EXPLORING INSTITUTIONAL SEXUAL ASSAULT, BETRAYAL, AND TRUST-BASED HARM By Kathleen Darcy A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Criminal Justice - Doctor of Philosophy 2022 ABSTRACT EXPLORING INSTITUTIONAL SEXUAL ASSAULT, BETRAYAL, AND TRUST-BASED HARM By Kathleen Darcy The dissertation investigated harm in the context of institutional sexual assault using trust as a lens to identify potential for harm (1) at different stages of the trust process and (2) in interpersonal versus institutional trust relationships. Findings illuminated how stages of the trust process aligned with different harms and trust relationships, adding understanding to imposed and focal vulnerability, trustworthiness, and emergent vulnerability. The scoping review identified harms for both interpersonal and institutional trust as concrete (e.g., physical, financial) where identity, context, and power mattered, and as amorphous (e.g., professional, privacy, emotional). Interpersonal trust articles identified mechanisms to understand emotional harm (moral injury and betrayal), whereas only (institutional) betrayal arose in institutional trust. Next, qualitative analyses using the victim impact statements given at the sentencing trial of Dr. Larry Nassar compared how victims who experienced different trust relationships (interpersonal and multilevel) spoke to vulnerability, trustworthiness, and harm. Vulnerability related to identity and context, differing by type of trust. Both groups described multiple reasons to trust, but only the multilevel group identified (failed) institutional factors. Moral injury and betray helped understand harm, but the multilevel group identified harms tied to institutional responses (e.g., putting institutional interests above its members). The final chapter connected empirical results to the trust process framework. Legal and practical implications of the findings are discussed, where improved understanding of victims’ experiences can better shape efforts at prevention, improve reporting, and better tailor legal remedies. ACKNOWLEDGMENTS First, thank you to all survivors, especially those who have bravely shared their stories with the world and who constantly remind me why this work matters. I want to give a huge thank you to my incredible network of scholarly support. First, Dr. Joe Hamm, I truly can’t thank you enough for pushing me, supporting me, and ultimately making me a better writer and scholar (and for fostering a begrudging passion for trust). You are an incredible mentor and set an outstanding example for how to support your students. To my committee members, thank you to Dr. Mary Finn, for helping me navigate research, classes, and parenting during the first years of the program (and a global pandemic), and for continually modeling ethical research and teaching. To Dr. Jennifer Cobbina-Dungy, I have learned so much from you about how to engage in quality research but, even more importantly, how to put that research into practice. Finally, Dr. Francine Banner, you (without knowing it) set me down this path when I first engaged with your incredible research. I am so lucky to call you a colleague and am so grateful for your support, kindness, and insight. Additionally, Dr. Gina Fedock, thank you for your brilliance and friendship; Dr. Chris Smith, thank you for your teaching wisdom and general support; and Dr. Christina DeJong, thank you for your unwavering support in every area. Additionally, thank you to my amazing cohort (with a special shoutout to Dr. Jin Lee, Dr. Ariel Roddy, and, especially this past year, my dear friend, Ashleigh LaCourse). I have also been so inspired from constant the brilliance and unwavering support of everyone in the TRUSST Lab, especially Amanda Osuna, whose excitement about the research kept me going. Thank you to my daughter Gabby, who was a (welcomed) distraction and motivation. A massive thank you to my parents, sisters, and in-laws (and friends who are like family) who have iii stepped in to provide childcare, found humor (even in dark times), and were sources of continuous strength over the past five years. Last, but far from least, thank you to my husband Mike. I never could have even begun to pursue this dream if I didn’t have your constant love, positivity, and advocacy motivating me through the tough times. I am so incredibly grateful for you and thank you from the bottom of my heart. iv TABLE OF CONTENTS LIST OF TABLES ....................................................................................................................... viii LIST OF FIGURES ....................................................................................................................... ix CHAPTER 1: INTRODUCTION ................................................................................................... 1 Literature Review........................................................................................................................ 5 Institutional Sexual Assault .................................................................................................... 5 Direct Harm ............................................................................................................................ 7 Trust-Based Harm ................................................................................................................... 8 Trust Process ......................................................................................................................... 11 Interpersonal Trust ................................................................................................................ 12 Institutional Trust .................................................................................................................. 16 Integrating the Literature .......................................................................................................... 19 CHAPTER 2: UNDERSTANDING INTERPERSONAL AND INSTITUTIONAL TRUST AND HARM: A SCOPING REVIEW ................................................................................................... 24 Abstract ..................................................................................................................................... 24 Introduction ............................................................................................................................... 24 Literature Review...................................................................................................................... 26 Understanding Trust.............................................................................................................. 27 Conceptualizing Harm .......................................................................................................... 28 Trustee................................................................................................................................... 30 Scoping Reviews................................................................................................................... 32 Current Study ............................................................................................................................ 34 Method ...................................................................................................................................... 35 Data Collection and Analysis................................................................................................ 35 Results ....................................................................................................................................... 40 Study Context ....................................................................................................................... 40 Study Design ......................................................................................................................... 40 Themes .................................................................................................................................. 41 Interpersonal Concrete Harms .............................................................................................. 41 Interpersonal Amorphous Harms .......................................................................................... 45 Multilevel Harms .................................................................................................................. 48 Institutional Concrete Harms ................................................................................................ 49 Institutional Amorphous Harms............................................................................................ 50 Discussion ................................................................................................................................. 51 Summary of Evidence ........................................................................................................... 51 Concrete Harms and Vulnerability ....................................................................................... 52 Amorphous Harms ................................................................................................................ 54 Mechanisms for Harm........................................................................................................... 56 Multilevel Harm .................................................................................................................... 56 Limitations and Gaps of the Knowledge Base ...................................................................... 58 v Conclusion ................................................................................................................................ 59 CHAPTER 3: “I THOUGHT I COULD TRUST YOU, BUT INSTEAD YOU ABUSED ME”: UNDERSTANDING VICTIM ACCOUNTS OF INTERPERSONAL AND MULTILEVEL VULNERABILITY AND TRUSTWORTHINESS ..................................................................... 61 Abstract ..................................................................................................................................... 61 Introduction ............................................................................................................................... 61 Literature Review...................................................................................................................... 65 Institutional Sexual Assault & Trust ..................................................................................... 65 Trust Process ......................................................................................................................... 66 Differences by Trust Relationship ........................................................................................ 67 Vulnerability ......................................................................................................................... 67 Trustworthiness ..................................................................................................................... 69 Current study............................................................................................................................. 72 Method ...................................................................................................................................... 74 Data ....................................................................................................................................... 74 Study Design ......................................................................................................................... 75 Data Analysis ............................................................................................................................ 76 Coding ................................................................................................................................... 77 Results ....................................................................................................................................... 78 Interpersonal Trust Group ..................................................................................................... 78 Multilevel Trust Group ......................................................................................................... 81 Interpersonal Trust Group ..................................................................................................... 87 Multilevel Group................................................................................................................... 90 Discussion ..................................................................................................................................... 93 Trustworthiness ..................................................................................................................... 97 Conclusion and Limitations .................................................................................................... 101 CHAPTER 4: THE POWER TO DAMAGE: UNDERSTANDING VICTIM ACCOUNTS OF INTERPERSONAL AND MULTILEVEL TRUST AND HARM ............................................ 103 Abstract ................................................................................................................................... 103 Introduction ............................................................................................................................. 103 Literature Review.................................................................................................................... 106 Trust Process ....................................................................................................................... 107 Differences by Trust Relationship ...................................................................................... 108 Emergent Vulnerability and Harm ...................................................................................... 109 Current Study .......................................................................................................................... 112 Method .................................................................................................................................... 113 Data ..................................................................................................................................... 113 Study Design ....................................................................................................................... 113 Data Analysis .......................................................................................................................... 114 Coding ................................................................................................................................. 115 Results ..................................................................................................................................... 116 Interpersonal Trust Group .................................................................................................. 116 Multilevel Trust Group ....................................................................................................... 119 Interpersonal Mechanisms for Understanding Harm .......................................................... 121 vi Multilevel Mechanisms for Understanding Harm .............................................................. 123 Interpersonal Group ............................................................................................................ 128 Multilevel Group................................................................................................................. 130 Discussion ............................................................................................................................... 136 Direct Harms ....................................................................................................................... 136 Mechanisms for Understanding Harm ................................................................................ 138 Moral Injury ........................................................................................................................ 140 Conclusion .............................................................................................................................. 141 CHAPTER 5: DISCUSSION AND IMPLICATIONS ............................................................... 143 Introduction ............................................................................................................................. 143 Insights from Empirical Chapters ........................................................................................... 147 Vulnerability ....................................................................................................................... 147 Trustworthiness ................................................................................................................... 148 Emergent Harm ................................................................................................................... 149 Trust Relationship ............................................................................................................... 150 Conclusion .............................................................................................................................. 153 APPENDICES…………………………………………………………………………………………………. 154 APPENDIX A: Figures for Chapter 1 …………………………………………………………... 155 APPENDIX B: Figures and Tables for Chapter 2 …………………………………………….. 156 APPENDIX C: Figures and Tables for Chapter 3……………………………………………... 171 APPENDIX D: Figures and Tables for Chapter 4……………………………………………...176 APPENDIX E: Figures for Chapter 5……………………………………………………………..181 REFERENCES………………………………………………………………………………………………….182 vii LIST OF TABLES Table 1: Understanding Harm Stemming from Trust Relationships……………………………….. 163 Table 2: Studies Included in Scoping Review………………………………………………………....... 164 Table 3: Themes for Scoping Review Articles………………………………………………………......169 Table 4: Demographic Characteristics…………………………………………………………………… 172 Table 5: An Example of the Three-Tiered Coding Framework Used for Vulnerability……...... 173 Table 6: Theoretical Constructs, Themes, Codes for Vulnerability…………………………………174 Table 7: Theoretical Constructs, Themes, Codes for Trustworthiness…………………………….. 175 Table 8: An Example of the Three-Tiered Coding Framework Used for Harm………………..... 177 Table 9: Theoretical Constructs, Themes, Codes for Direct Harm…………………………………. 178 Table 10: Theoretical Constructs, Themes, Codes for Betrayal……………………………………...179 Table 11: Theoretical Constructs, Themes, Codes for Moral Injury……………………………….. 180 viii LIST OF FIGURES Figure 1: A Visual Representation of Harm within Institutional Sexual Assault Experiences...155 Figure 2: Trust Process………………………………………………………………………………………..155 Figure 3: PRISMA Flowchart for Scoping Review Trust and Harm.……………………………….156 Figure 4: Database Search for Scoping Review…………………………………………………………157 Figure 5: Template for Extraction Review………………………………………………………………. 161 Figure 6: Trust Process and Potential for Harm Stage One………………………………………….. 171 Figure 7: Trust Process and Potential for Harm Stage Two………………………………………….. 176 Figure 8: Trust Process……………………………………………………………………………………….181 ix CHAPTER 1: INTRODUCTION Shocking stories of institutional sexual assault (ISA) (assault occurring in institutional settings or between institutional actors) have been brought to light in numerous settings, including prisons, the military, higher education, sports, medicine, religious institutions and more (Blakemore, Herbert, Arney, & Parkinson, 2017; Crossmaker, 1991). Research on ISA suggests it is widespread, difficult to report, and potentially impacts many victims over decades (Austen, 2022; Cave, 2022; Gerson, 2022; Svrluga, 2022). ISA has been linked to a number of direct negative impacts on victims1 (e.g., psychological, educational, emotional, economic) (Blakemore et al., 2017), which are redressable using a variety of legal options (Bublick, 2014; Bublick & Mindlin, 2009). However, part of the complexity in understanding sexual assault in institutional contexts is that there are a variety of types of relationships in which assault can occur, each of which can relate to distinct harms (Figure 1). Two types of trust relationships are particularly relevant: (1) interpersonal trust (trust in another person) and (2) institutional trust (trust in an institution). To explore harms associated with interpersonal trust, betrayal trauma theory posits that sexual assault in an interpersonal trusting relationship (e.g., caregiver/child) is related to mental health, physical health, and relationship harms separate from harms associated with sexual assault more broadly (Delker & Freyd, 2014; Delker, Smith, Rosenthal, Bernstein, & Freyd, 2018; Freyd, 1996). That is, distinct harms seem to emerge when abuse occurs in a relationship of interpersonal trust. Distinct harms may also stem from violations of institutional trust in the wake of ISA. The theory of institutional betrayal suggests that a trusted institution’s action or 1 Some individuals who have experienced sexual assault might prefer to identify themselves as survivors, however the research and legal terminology utilize the word “victim,” so for the sake of uniformity and clarity that is the term used throughout the dissertation. 1 inaction (e.g., not taking reports of assault seriously, punishing those who report) relate to similar mental and physical health harms for victims (C. P. Smith & Freyd, 2013). While trust is at the center of both betrayal trauma and institutional betrayal, its existence is merely presumed. The literature doesn’t explain why a trust relationship relates to distinct harms experienced by victims. There are important differences between mere cooperation compared trust that help explain unique vulnerabilities to harm stemming from trust (Ferrin, Bligh, & Kohles, 2008). Trust explicitly contemplates ceding some control and being vulnerable to the actions of another (Baier, 1986; Rousseau, Sitkin, Burt, & Camerer, 1998). Vulnerability relates to the necessary uncertainty and potential for negative outcomes that exist in all trust relationships, which creates the potential for trust to be abused through opportunistic behavior (PytlikZillig & Kimbrough, 2016; Schilke, Wiedenfels, Brettel, & Zucker, 2017). Thus, the trust literature provides a useful lens to understand the potential for harm(s) that are associated with trust in the context of ISA. Trust is commonly conceptualized as a psychological state within a trustor (the one doing the trusting) characterized by a willingness to be vulnerable to the actions of another party (the trustee) (Mayer, Davis, & Schoorman, 1995). This definition sets forth a complex process associated with a trust relationship which, when broken down, can provide insight into potential for harm (see, Figure 2). First, the definition positions vulnerability at the core of trust. Vulnerability is the general state of being susceptible to damage or harm (Misztal, 2012). There are many types of vulnerability that could exist for individuals. The limited research that has been done seeks to understand vulnerability that exists before deciding to trust (Baghramian, Petherbridge, & Stout, 2020; Misztal, 2012). Some research has suggested individuals differ on their potential for imposed vulnerability, that is, exposure to events over which they do not have control 2 (Montgomery, Jordens, & Little, 2008). Research increasingly recognizes that intersectional identity matters, where a variety of psychological, physical, social and economic elements shape a person’s imposed vulnerability (e.g., age, race, education, ability) (Kuran et al., 2020; Montgomery et al., 2008, Darcy et al., under review). Although this broad vulnerability provides important context for trust, it is only a subset of it that is directly relevant to trust. When an individual trusts, it is not the breadth of their vulnerability they are willing to accept, but only that vulnerability they experience specifically to the actions of the trustee. Hamm and Möllering (2022) argue that this “focal vulnerability” is similarly relational (tied to their relationships and status in society) and is the vulnerability they contemplate in considering a willingness to be vulnerable. Next, trustors are willing to accept focal vulnerability, in large part, because of evaluations of the trustee’s worthiness of being trusted. These trustworthiness perceptions are the characteristics of a trustee which help a trustor decide if they are willing to accept vulnerability to them. A significant body of literature connects these trustworthiness assessments to the state of trust (Colquitt, Scott, & LePine, 2007; PytlikZillig & Kimbrough, 2016; Rousseau et al., 1998), which then facilitates risk taking in the relationship (Mayer et al., 1995; Schoorman, Mayer, & Davis, 2007). Finally, risk is tied to the behaviors that manifest a willingness to be vulnerable (actions actually relying on the person, such as leaving a child in their care) (Baier, 1986; Mayer et al., 1995). Thus, trust as a psychological state characterized by a willingness to accept vulnerability, paradoxically facilitates behaviors that create new vulnerability for a victim by exposing them to additional risks (Misztal, 2012). For example, risk taking in a relationship includes decreased monitoring of the trustee, letting one’s guard down, or increased dependence (Misztal, 2012; 3 Weber, Malhotra, & Murnighan, 2004). This creates potential for additional vulnerability and harm. To explore this potential, I introduce the concept of emergent vulnerability. The trust process illuminates the different vulnerabilities and potential for harm that emerge in trust relationships. Research on trust also provides strong reason to believe that the trust process, and, therefore, harms, might also look different depending on whether there is trust in another person or an institution. 2 Vulnerability to an individual may create risk of exploitation, betrayal, financial damage, or emotional harm if trust is betrayed (Baier, 1986; M. L. E. Chan, 2009), whereas vulnerability to institutions might depend on the extent to which a person identifies with and cedes control of different aspects of their lives to the institution. For example, institutions can be an integral part of a trustor’s identity such that they might eat, sleep and work there, rely on the institution to govern disputes, and for protection (e.g., universities that act in loco parentis) (Banner, 2014; Beteille, 2012; Scott, 2010). Institutions therefore foster loyalty in a way that may foreground some proposed dimensions of vulnerability more than others (e.g., concreteness or severity; Hamm & Wolfe, 2021). In addition, trustworthiness perceptions differ. In an interpersonal relationship, trustworthiness is informed by assessing another person’s ability, benevolence, and integrity (Colquitt et al., 2007). However, different, more global factors are theorized to influence a trustor’s perceptions of trustworthiness of an institution (i.e., legal provisions, reputation, certification of exchange partners and community norms) (Bachmann & Inkpen, 2011). 2 Institutions can be broadly defined as “the prescriptions that humans use to organize all forms of repetitive and structured interactions” including those within families, neighborhoods, churches, governments (Ostrom, 2005 p. 3) and will be used as a broad term that encompasses specific organizations (a university, the Catholic Church, etc.) in this dissertation. 4 The goal of the dissertation was to investigate harm in the context of ISA using trust as a lens to identify different potential(s) for harm (1) at different stages of the trust process and (2) in interpersonal versus institutional trust relationships. To accomplish this goal, the dissertation is presented in a five-paper format. Each chapter will correspond to the overarching theoretical framework and goal presented in the introduction chapter, but they are intended to stand alone. Thus, some elements of the works (including relevant literature and methods) may overlap. The second chapter is a scoping review that explored how vulnerability, trustworthiness and harm were discussed in existing trust literature and how this differed by type of relationship. The third and fourth chapters explored statements by victims of ISA given at the sentencing trial of Dr. Larry Nassar. These chapters compared how those who experienced different trust relationships (interpersonal and multilevel trust) spoke to vulnerability, trustworthiness, and harm. The final chapter drew together insights from the previous chapters, identifying how the empirical data mapped on to the trust process in practice to build practical and legal implications. Experienced harm is concretely tied to harm as a legal phenomenon, so improved understanding of victims’ experiences can better shape efforts at prevention, improving reporting, and access to remedies. This multimethod approach helped fill gaps and break down silos between literatures to increase understanding of harm in ways that can better provide victims of ISA with resources and advocacy tailored to their needs. Literature Review Institutional Sexual Assault Sexual assault is widespread in the United States, with an estimated one in three women and one in six men experiencing some form of sexual violence in their lifetimes (S. Smith et al., 2018). Sexual assault can occur in a variety of relationships, such as romantic or acquaintance 5 relationships, with strangers, or even in relationships of caretaking or supervision (Abrahams et al., 2014; Brownmiller, 1975; Parrot & Bechhofer, 1991; Tjaden & Thoennes, 2006). Certain identities are at higher risk of sexual assault, where women, members of historically marginalized groups, and members of the LGBTQ+ community face increased risk (Callan, Corbally, & McElvaney, 2021; McCauley, Campbell, Buchanan, & Moylan, 2019). Sexual abuse is one of the most under-reported crimes, where victims experience higher levels of shame, stigmatization, and self-blame than victims of other crimes (Campbell, Dworkin, & Cabral, 2009; Dworkin, Menon, Bystrynski, & Allen, 2017; Herman, 2015). Increasingly, ISA research has specifically focused on how certain institutional contexts also increase vulnerability for sexual assault (Barter, 1997; Gil, 1982). While much of the research has focused on risk for sexual assault in higher education (see, e.g., Armstrong, Hamilton, & Sweeney, 2006; Moylan, Javorka, Maas, Meier, & McCauley, 2021; Muehlenhard, Peterson, Humphreys, & Jozkowski, 2017), victims face increased risk for sexual assault in a range of institutional settings. These include child care and residential facilities, correctional facilities, disabled adult institutional facilities, the military, higher education, and sports (Banner, 2014; Barter, 1997; Beck, Rantala, & Rexroat, 2014; Brackenridge, 1997; Cantor, David; Fisher, Bonnie; Chibnall, Susan; Townsend, Reanne; Lee, Hyunshik; Bruce, Carol; Thomas, 2015; A. Carr et al., 2010; Crossmaker, 1991; Department of Defense, 2019; McAlinden, 2006). The research suggests that ISA is widespread, characterized by repeat victimization, and has the capacity to continue for years unabated (Banner, 2014; Beck, Berzofsky, Caspar, & Krebs, 2013; Cantor, David; Fisher, Bonnie; Chibnall, Susan; Townsend, Reanne; Lee, Hyunshik; Bruce, Carol; Thomas, 2015; Crossmaker, 1991; Daigle, Fisher, & Cullen, 2008). 6 Direct Harm Institutional sexual assault has been linked to a complex web of direct harms experienced by victims (Blakemore et al., 2017). Victims of sexual assault report significantly worse psychopathology than those who have not been assaulted, where a recent meta-analysis found that victims are at particular risk of depression, PTSD, anxiety, disordered eating, substance abuse or dependence, suicide ideation or attempts, and bipolar and obsessive-compulsive conditions compared to non-victims (Dworkin et al., 2017). Certain characteristics of assaults may be associated with higher risk for mental health impacts (e.g., strangers as perpetrators, use of a weapon) (Campbell et al., 2009; Dworkin et al., 2017). Victims may also have increased risk of long term physical and reproductive health impacts (Gazmararian et al., 2000; Santaularia et al., 2014). When victims do report (even informally), increasingly research suggests they face unique harms based on the responses of those to whom they report (Ahrens, 2006; Campbell, Ahrens, Sefl, Wasco, & Barnes, 2001). Sexual assault is usually linked to direct harm stemming from an individual perpetrator, but ISA creates the potential for an institutional role in the perpetration of direct harm. In these examples, injury is being facilitated (but not specifically caused) by the institution. For example, institutional context can facilitate a perpetrator’s access to enable assault, allowing the direct harm to occur (McAlinden, 2006). There may be institution-specific direct harms associated with ISA, for example, sexual assault in higher education has been linked to sexual revictimization and decreased academic performance (Decker & Littleton, 2018). In short, victims’ experiences with abuse, reporting, adjudication, and recovery range greatly, in part because direct harms may differ based on their experience. Better understanding the types of harms associated with sexual 7 assault can improve efforts to mitigate harms by informing targeted interventions (Dworkin et al., 2017). Trust-Based Harm In addition to direct harms, research increasingly recognizes the potential for harm based in trust. When sexual assault occurs in a trust relationship, this can exacerbate or relate to distinct harms to victims. These include physical and mental health harms, but also impacts on revictimization, future relationships, and self-blame (Delker & Freyd, 2014; Delker et al., 2018; Gobin & Freyd, 2009; Monteith, Bahraini, Matarazzo, Soberay, & Smith, 2016; Smith & Freyd, 2014, 2017). Two theories have emerged to explain why institutional sexual assault occurring in a trust relationship is linked to distinct harms. Betrayal trauma theory suggests that sexual assault occurring in a close, trusting, or dependent relationship between an offender and victim results in worse mental health outcomes than assault perpetrated by a non-caregiver or non-trusted party (Birrell & Freyd, 2013; Freyd, 1996; C. P. Smith & Freyd, 2014). Birrell & Freyd (2013) examined assault that occurred in multiple relationships of dependence (e.g., caregiver or parent/child; coach/player; domestic partners; correction officer/parolee) and the common source of harm was identified as the “violation of trust” (Birrell & Freyd, 2013, p. 18). Coping mechanisms include a blindness to the assault or turning blame inward (Birrell & Freyd, 2013; Gartner, 1999). In research applying the theory, betrayal trauma was linked to decreased psychological well-being (e.g., PTSD, dissociation), substance use, revictimization, violent offending, and decreased future tendency to trust (Delker & Freyd, 2014a; Gobin & Freyd, 2014; Mackelprang et al., 2014; Platt & Freyd, 2015; Tang & Freyd, 2012). Taylor and colleagues (2020) found that chronically homeless individuals who reported a violation of trust as part of their self-identified “worst” lifetime 8 trauma (betrayal trauma) presented with significantly more PTSD symptoms than those who did not. Although rooted in interpersonal trust, it’s important to recognize that with ISA, betrayal trauma doesn’t happen in a vacuum. Instead it operates “in a larger context” of cultural attitudes, systems, and institutions (Birrell & Freyd, 2013, p. 35; Goldsmith, Martin, & Smith, 2014). For example, a “professional perpetrator” might use their job to target and “groom” victims, fostering an interpersonal trust relationship (McAlinden, 2006). Although this research integrates the institution in betrayal trauma, it still roots harms in the betrayal that stems from another person, even if they operate in an institutional context. Thus, trust is a critical element of betrayal trauma because it provides insight into the source of harm in these trust relationships. However, while betrayal trauma theory is built on an assumption that a trusting relationship is key to understanding harms experienced by victims, the research never actually explores or verifies the presence of trust in these relationships (Birrell & Freyd, 2013). The role of institutions in ISA can go beyond merely facilitating direct harm and betrayal trauma. Specifically, institutional trust violations are equally important sources of distinct harm in the context of ISA. In the 1990s, sexual assault research began to focus on how responses by institutions that a victim trusts can shape victim distress and increase harm and trauma (Campbell & Raja, 1999; Herman, 1992). These responses create the potential for secondary victimization, which is when a victim feels blamed or experiences negative or judgmental behaviors and attitudes, either through unresponsive treatment or victim-blaming practices by organizational actors (Campbell & Raja, 1999; Orth, 2002). This began an important trajectory expanding source of harm and trauma outside the interpersonal victim/perpetrator relationship and recognizing systemic trauma (Goldsmith et al., 2014). 9 Institutional betrayal theory continues that trajectory, explicitly hypothesizing that institutions may “elicit similar trust and dependency from their members as is found in interpersonal relationships” (C. P. Smith & Freyd, 2013, p. 119). In some cases, institutions provide protection (e.g., universities that act in loco parentis) or individuals may be dependent on the institution for survival (e.g., the military) (R. W. Baker, McNeil, & Siryk, 1985; Holliday & Monteith, 2019). Institutions can also be an integral part of a member’s identity; they govern disputes, one may eat, sleep and work there; and they may foster intense loyalty (Banner, 2014; Beteille, 2012; Sampaio, Perin, Simões, & Kleinowski, 2012; Scott, 2010). Institutional betrayal is perceived by the victim as a violation of the trust relationship they had with the institution (C. P. Smith & Freyd, 2014). Institutional betrayal theory postulates that harm occurs when institutions fail to prevent or supportively respond to abuses committed by members of the institution against other members (Smith & Freyd, 2014). These betrayals can be acts of commission (e.g., cover-ups, insensitive responses to reports of abuse), acts of omission (e.g., failing to prevent the abuse), systematic, or isolated events (Smith, Gómez, & Freyd, 2014; Smith & Freyd, 2014). The institution may create hostile environments to normalize abuse, make it difficult to report abuse, mishandle complaints, and/or retaliate against victims (Holliday & Monteith, 2019; Smith & Freyd, 2013). This theory focuses on how institutional action and/or inaction may cause distinct trauma for a victim (separate from trauma associated with sexual abuse generally, and separate from betrayal trauma in an interpersonal relationship) (C. P. Smith & Freyd, 2013, 2017). While institutional betrayal theory has largely been applied in higher education (Rosenthal, Smith, & Freyd, 2017; C. P. Smith, Cunningham, & Freyd, 2016; C. P. Smith & Freyd, 2013, 2017), it has also been explored in other institutional settings, such as the military, 10 high schools, healthcare settings, and the judiciary (Lind, Adams-Clark, & Freyd, 2020; Monteith et al., 2016; C. P. Smith, 2017; C. P. Smith et al., 2014). Institutional betrayal is linked to reduced help-seeking behaviors for ISA victims, as well as negative mental health outcomes, including PTSD, depressive symptomology, anxiety, borderline personality disorder, and increased suicidality (Holliday & Monteith, 2019; Monteith et al., 2016; C. P. Smith & Freyd, 2017). While research addressing betrayal trauma and institutional betrayal expand thinking on the variety of harms caused by ISA, they are limited in the extent to which they fully engage with the trust that is theorized to cause them. Trust is what makes harms associated with betrayal trauma and institutional betrayal distinct and is, therefore, a critical element of these theories (Birrell & Freyd, 2013). However, relationships of “dependence” are presumed to involve trust without further exploration (Birrell & Freyd, 2013, p. 18). To address this shortcoming, the wider literature on trust provides a lens unpack its nature and dynamics more fully. Trust Process The trust process stems from the common conceptualization of trust as “the willingness of a party to be vulnerable to the actions of another party” (Mayer et al., 1995). In operationalizing trust, most reviews of literature agree that trust involves a trustor and trustee that are somehow interdependent and involves a situation containing risks for the trustor (PytlikZillig & Kimbrough, 2016; Rousseau et al., 1998). This conceptualization has three important elements that each provide insight to potential harm(s). First, it positions vulnerability at its core, so understanding an individual’s vulnerability helps understand which harm(s) they are willing to accept by trusting. A victim will consider potential harm in their initial decision to trust, but trust is dynamic, so vulnerabilities and potential harm might change. Second, it draws 11 attention to understanding the reasons why one might be willing to accept that vulnerability, one of which includes characteristics of a trustee, referred to as trustworthiness. Finally, it argues that trust will facilitate risk taking in a relationship. This is the behavioral manifestation of trust, that is, acting in reliance on a person’s goodwill (e.g., giving them money, leaving your child in their care) (Baier, 1986; Mayer et al., 1995). Trust is dynamic, so even once its established, a person will continually consider the context and likelihood of positive or negative outcomes, and could involve letting their guard down or reducing monitoring (Frederiksen, 2014; Hamm, Cox, et al., 2019; Mayer et al., 1995; Sitkin & Pablo, 1992). These actions within the trust relationship might relate to distinct, emergent vulnerability. Understanding trust as an acceptance of vulnerability means that it creates vulnerability. Interpersonal Trust Vulnerability is the state of being prone to or susceptible to damage or injury, or being exposed to risk, and is an important precondition to trust (Colquitt et al., 2007; Ferrin et al., 2008; Mayer et al., 1995; Misztal, 2012; PytlikZillig & Kimbrough, 2016). A certain level of vulnerability exists generally for all individuals. Some research refers to this as imposed vulnerability. In the limited work that has explored this vulnerability, it considers events that are out of one’s control but which can lead to serious harm (e.g., natural disasters) (Misztal, 2012; Montgomery et al., 2008). This vulnerability integrates the potential for a wide variety of risks and often intersects with social status, where we might refer to “vulnerable children” or “vulnerable adults” to highlight how this vulnerability is linked not only to the hazard but also to the one at risk of injury from it (Misztal, 2012). Importantly, there may be qualitative differences in imposed vulnerability on the basis of intersectional identity, suggesting that systems of oppression relate to different levels of social vulnerability (Kuran et al., 2020). Numerous 12 intersecting elements, including social, physical, and economic elements, might impact intersectional vulnerability (e.g., the elderly might lack technology knowledge, possess lower capacity to react in a crisis, or may be at risk of relying on others for information) (Kuran et al., 2020; Montgomery et al., 2008, Darcy et al., under review). Imposed vulnerability therefore creates an important backdrop for trust but it is larger than the specific vulnerability that trustors feel willing to accept. When an individual is willing to accept vulnerability in a relationship, they create both opportunity and potential for harm for themselves (Kramer, Hanna, & Wei 2001). Thus, when one trusts, it is not all of their vulnerability they are willing to accept, but specifically their vulnerability to the deliberate decisions of the trustee. Recently, there have been attempts to distinguish the vulnerability one decides they are willing to accept when entering a trust relationship as focal vulnerability (Hamm & Mollering, 2022; Nienaber, Romeike, Searle, & Schewe, 2015). Focal vulnerability is theorized to be perceived, relational, and part of an ongoing evaluation that is shaped by a trustor’s social position (Hamm & Mollering, 2022). This necessarily requires some awareness of vulnerability to some potential harm that one is willing to accept. Recent conceptualizations of harm can help understand focal vulnerability. Types of harm can range from concrete, which refer to a demonstrable injury under the control of a trustee, to more amorphous, which refer to failures to facilitate feelings of security or perhaps emotional harm (Hamm & Wolfe, 2021). In applying this harm typology, Hamm and Wolfe (2021) found that when citizens think about harms that could stem from trust in police, concrete harms included citations (less severe) and arrest (more severe). However, amorphous harms included being misidentified as a perpetrator (more severe) and being the victim of a crime (less severe) (Hamm & Wolfe, 2021). Their results suggested that concreteness of harm and its 13 severity corresponded with different preferences in trustworthiness attributes of individual police officers (Hamm & Wolfe, 2021). In light of how they understand their vulnerability, a person then looks to signals that suggest they should trust the trustee. In other words, a trustor’s willingness to accept focal vulnerability is heavily tied to their evaluations of the trustee (Mayer et al., 1995). Research regularly shows that these evaluations are strongly related to trust levels (Colquitt et al., 2007), so “clearly, then, the concept of trustworthiness is central to understanding and predicting trust levels” (Colquitt et al., 2007, p. 910). Deeming another person worthy of trust might relate to past experiences with them, linguistic devices they use, or their characteristics (Mayer, Davis, & Schoorman, 1995b; Pate, Morgan-Thomas, & Beaumont, 2012; Sharp, Thwaites, Curtis, & Millar, 2013; Wubs-Mrozewicz, 2020). Although a wide variety of trustworthiness constructs have been proposed, there is a consensus that ability, benevolence, and integrity account for most of the variance in trustworthiness assessments (Colquitt et al., 2007). Ability refers to the group of skills and competencies that a party has in some relevant domain (Mayer et al., 1995). Benevolence refers to perceptions that the trustee wants to do good to or cares about the trustor, and integrity refers to perceptions that a trustee is complying with some set of principles that are important to the trustor (e.g., a sense of justice or consistency in actions) (Mayer et al., 1995). These factors are interrelated and continually predict trust in other people (Colquitt et al., 2007; Mayer et al., 1995; Poon, 2013; PytlikZillig & Kimbrough, 2016). In other words, the presence of one or even two attributes of trustworthiness would not necessarily result in willingness to be vulnerable, but these characteristics of a trustee help build a foundation to develop trust (Mayer et al., 1995; Schoorman et al., 2007). In general, the extent to which one perceives ability, benevolence, and 14 ability in a trustee then shapes their willingness to accept their own vulnerabilities and cede control to that trustee. Trustworthiness thus drives a willingness to accept focal vulnerability which, unlike imposed vulnerability, is electively engaged with by taking risk (cooperating) with another person (Montgomery et al., 2008). Risk taking in a relationship might relate to increased dependence, which can create a vulnerability to exploitation especially when relationships have a power imbalance (Misztal, 2012). Thus, once an individual decides to accept vulnerability in a relationship, they are more likely to take risks in the relationship (Mayer et al., 1995) and in doing so, create new vulnerability and potential for harm for themselves. While existing theory clearly acknowledges this potential (“the act of trust needs to be seen as offering both, a solution to the problem of our vulnerability, and as exposing us to more risks” (Misztal, 2012, p. 216)), this vulnerability is undertheorized. I suggest that it be considered emergent vulnerability, which is shaped by behaviors rooted in trust that are established with new information, increased dependence, or motivations to keep trusting. Emergent vulnerability does not need to be focal, however, it is inherent in a trust relationship. Conceptualizations of harm tied to emergent vulnerability have not been considered in existing literature. However, some research has explored the emotional experience of interpersonal trust breach and betrayal (within an existing trust relationship), finding that violations can be traumatic, relating to a sense of injustice or mistreatment (Robinson, Dirks, & Ozcelik, 2004). Similarly, betrayals impacted general beliefs about respecting other people and codes of conduct (Rousseau et al., 1998). The trustee’s motive may also affect the emotional impact of the betrayal, where incidental betrayal might be perceived as less harmful than intentional betrayal (M. L. E. Chan, 2009). Exploring harms associated with emergent 15 vulnerability is particularly relevant to understanding how vulnerability (and harms) might change throughout the trust process. Institutional Trust There is a growing body of literature that suggests important differences as a function of whether trust is in an individual or an institution. Institutional trust relies on a similar broad process as interpersonal trust: some level of focal vulnerability exists for a trustor, they assess the trustworthiness of the institution to handle it and, if they are willing to be vulnerable, they become more likely to engage in risk taking in the relationship which then creates an emergent vulnerability to harm (Schoorman et al., 2007). Like an individual, actions of an organization or institution can promote or constrain trust relations, and provide the opportunity for trust violations (Fukuyama, 1995; Rousseau et al., 1998). However, institutions may not rely on interpersonal interactions in the same way interpersonal trust is developed (Bachmann & Inkpen, 2011; McEvily, Weber, Bicchieri, & Ho, 2006). Instead, a trustor might rely on proxies such as third parties (e.g., other institution members like coworkers) or develop role based trust, looking to the role the institutional actor plays in an organization, which might signal trustworthiness (Kramer, 1999). The specific target of institutional trust can be difficult to disentangle for trustors. For example, a trustor may not be able to tell whether they trust a physician or the medical system that trained them (including its educational tradition and curriculum they needed to pass before they met patients) (Bachmann, 2018). Thus, some have pushed to focus on multilevel trust instead of institutional trust (Fulmer, 2018; Lumineau & Schilke, 2018). Others have attempted to explicitly differentiate trust in an institution as a whole as compared to institutional actors (Tan & Lim, 2009; Tan & Tan, 2000). 16 Additionally, vulnerability differs in institutional trust. While decades of theory has explored the coercive or restrictive role of institutions on individual behavior (Foucault, 1977, 1982; Goffman, 1961), modern institutional theory suggests that members voluntarily seek out membership in institutions for purposes of self-improvement and accept a certain level of control over their lives in exchange (Scott, 2010). By giving up control to the institution an individual is disempowered, and accepts their vulnerability to and dependence on the institution, similar to an interpersonal relationship (Misztal, 2012; Schoorman et al., 2007). Institutions can be central to individuals’ identity formation, education, or vocation (e.g., higher education, religious institutions) and even, in some cases, foster dependence for safety or survival (e.g., police, military) (Banner, 2014; Beteille, 2012; Scott, 2010). This increases individuals’ level of identification with the institution, and as a result, members may be vulnerable to harm. For example, they may stay silent to harms they experience in an institutional setting for fear retaliation, fear of damaging the institution's credibility, or reluctance to exert their rights at the expense of the institution with which they identify (Banner, 2014; Barter, 1997; Nunno & Motz, 1988). Much of the research on institutional trust applies the same trustworthiness factors (ability, benevolence, and integrity) to assessments of institutions (Elsbach, 2004; Pirson & Malhotra, 2011; Schoorman et al., 2007), but some have argued that different factors shape trustworthiness perceptions of institutions. Bachman and Inkpen’s (2011) work on institutional trust suggests there are different mechanisms by which institutions act to foster development of trust by signaling trustworthiness. Two of these are formal mechanisms: (1) legal provisions (the backdrop of laws that promise sanction if one breaches trust); and (2) reputation (organizations may be incentivized to avoid acting in a way that will damage their reputation, fostering trust). 17 Two other mechanisms are informal: (3) certification of exchange partners (guidelines and norms set by professional or industry associations which promote behavioral norms and established rules, like a medical licensing board) and (4) community norms, structures, and procedures (these are set by the institution and shape behavior of members of the community, which acts as a proxy for trustworthiness of the overall community). Reflecting the complex nature of institutional trust, people rely on different trustworthiness signals for different trustees, where trust in a supervisor was associated with ability, benevolence, and integrity while trust in an organization was associated with more “global variables” (Tan & Tan, 2000). Investigating how these differences shape a trustor’s perceptions and decision-making in trusting an institution can help better understand harms experienced. Very little research has explored what risk taking in an institutional trust relationship looks like or what vulnerability might emerge with institutional trust. Much like in interpersonal relationships, risk taking in a relationship with an institution could involve giving up control or acting in reliance on the institution (Baier, 1986). However, unlike most interpersonal relationships it might also involve increased participation, embeddedness, and identification with an institution (Scott, 2010). Like interpersonal trust, emergent harm has not been explored in the institutional context. Some research suggests harm can occur when an institution has failed to fulfill its perceived obligations to it members, which constitutes a psychological contract breach (Kramer, 1999; Morrison & Robinson, 1997; Robinson, 1996). This seemingly flags an emergent vulnerability associated with perceived broken promises or obligations of an institution. Robinson’s (1996) study examining perceptions of 125 newly hired managers identified a vulnerability that only occurred once they entered into a trust relationship with the organization. Namely, when 18 employees started at a company with initial levels of trust in their employer, they later perceived violations of the psychological contract through unmet expectations (Robinson, 1996). Later work builds on these harms and suggest that psychological contract breach stemming from unmet expectations on the part of the institution had an emotional impact on the trustor, namely, betrayal and anger (Morrison & Robinson, 2006). There is a dearth of literature on institutional trust and emergent harms, but existing research suggests institutional trust, like interpersonal trust, can be tied to complex emotional harms. Integrating the Literature Betrayal trauma and institutional betrayal theories help explain harms experienced by victims of ISA by identifying abusive violations of trust as a distinct source of harm. Problematically, however, they simply presume the existence of trust. To help provide insight into trust-based harm, trust research suggests that a victim would identify from among their imposed vulnerabilities the specific focal vulnerability they are willing to accept to another’s actions (Hamm & Mollering, 2022; Montgomery et al., 2008). They then evaluate the characteristics of the trustee for evidence of its trustworthiness which, when identified, typically leads to a willingness to accept vulnerability that facilitates risk taking in the relationship (Colquitt et al., 2007; Mayer et al., 1995; Tan & Tan, 2000). This risk taking is then associated with a new, emergent, vulnerability that builds upon (but is distinct from) imposed and focal vulnerability. This likely differs whether a victim trusts in another person or an institution (Bachmann & Inkpen, 2011; Tan & Tan, 2000). Exploring the trust process is key to understanding harm and trust in betrayal trauma and institutional betrayal to be able to better craft remedies, interventions, and provide resources to victims who experience myriad harms. The goal of the dissertation is to investigate harm more fully in the context of ISA by exploring 19 harms associated with the stages of the trust process in individual versus institutional trust relationships. To this end, Chapter 2 evaluated how the literature spoke to harms that arise throughout the trust process as may differ by type of trust relationship. A scoping review was performed to address two research questions: (1) What is known about potential harms associated with interpersonal trust? (2) What is known about potential harms associated with institutional trust? and (3) What are the limitations and gaps of the knowledge base? A scoping review can be used to systematically synthesize literature across topics for reinterpretation (Baumeister & Leary, 1997; Siddaway, Wood, & Hedges, 2019). Collection of data (e.g., search terms and inclusion/exclusion criteria) was informed by Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) checklist and guidelines (Moher, Liberati, Tetzlaff, & Altman, 2009) and resulted in 50 articles for inclusion. Thematic analyses provided a method by which to identify prominent or recurrent themes in the literature and summarize findings related to these themes (Braun & Clarke, 2006; Clarke & Braun, 2017). Results suggested that for both trust relationships, articles fell into two general categories: concrete and amorphous harms. Concrete harms included abuse, physical health, financial harms, and multiple overlapping harms. These were particularly linked to different identities, situations, and contexts, providing insight into imposed and focal vulnerability. Amorphous harm emerged as professional, security, and relational harms. The interpersonal articles also identified amorphous emotional harm, mapping on to theory for moral injury and betrayal. Some multilevel trust articles arose, suggesting that identifying harm stemming from one type of trustee is difficult in institutional contexts. Finally, the results indicated that very little research has explored harms explicitly linked to institutional trust. The few articles that emerged primarily described amorphous emotional harms that were 20 linked to institutional betrayal as a mechanism for understanding harm. This method contributes to goal of the dissertation identifying qualitative differences between research on interpersonal and institutional trust and adding nuance to the conceptualization of harm as ranging from concrete to amorphous. Chapters 3 and 4 qualitatively examined the lived experiences of victims to evaluate how victims spoke to harm that stems from the trust process as differs by trust relationship. Specifically, I examined 127 victim impact statements (VIS) given in the Ingham County Court in Michigan in 2018 by women who experienced institutional sexual assault from Nassar. I applied constant comparative and grounded theory methods (Charmaz, 2006; Glaser & Strauss, 1967), adapting Auerbach & Silverstein’s (2003) three-level categorization system to identify low-level text-based categories, develop middle-level themes, and develop higher-level theoretical constructs. This method allowed for comparison within and between groups who either described (1) interpersonal trust (in the perpetrator, Nassar) or (2) multilevel trust (in both interpersonal and institutional trustees). For Chapter 3, the research questions explored were: (1) How do victims in the interpersonal trust group (versus in the multilevel trust group) understand trustworthiness? (a) in what ways does trustworthiness differ for those who described trusting individuals (versus multilevel)? (b) in what ways is trustworthiness similar for those who described trusting individuals (versus multilevel)? Victims in both the interpersonal and multilevel group described how identities and situations shaped their willingness to be vulnerable (e.g., were young, naïve, often injured, lacked medical knowledge, had family situations that made them vulnerable). This suggests that (intersecting) identities shaped vulnerability and decisions to trust, especially in terms how they bump up against institutional structures (e.g., governance). With these 21 vulnerabilities in mind, individuals looked to multiple intersecting trustworthiness signals to help them find reasons to be willing to be vulnerable. Both groups described multiple strong reasons to trust Nassar and institutions (e.g., institutional context and physical space mattered for both groups when assessing trustworthiness). The multilevel trust group relied more explicitly on institutional safeguards (and other institutional actors) to help them be willing to accept vulnerability to Nassar and associated institutions. For Chapter 4, the research questions explored were: (1) How do victims in the interpersonal group (versus in the multilevel group) describe how trust facilitates harm? (a) in what ways does harm differ for those who described trusting individuals (versus multilevel)? (b) in what ways is harm similar for those who described trusting individuals (versus multilevel)? For both groups, physical (iatrogenic) harms were prevalent, where they feared Nassar’s abusive behavior either exacerbated their injuries, prevented him from being motivated to treat their injuries, or both. Even in negative cases (where Nassar’s treatment did help them heal) many experienced emotional harm tied to the abuse or realizing the treatment was tied to abuse. Both groups also identified a wide array of very serious mental health impacts, that impeded their ability to sleep, have relationships, keep jobs, and function. Both groups also described injuries to relationships (through the mechanism of betrayal) impacting trust in others, dreams, careers, and intimacy, as well as injury to identity (through moral injury). The multilevel group uniquely described harm stemming from institutional action or inaction that silenced, disempowered, or blamed them for the abuse. Chapter 5 brought together insights from the previous chapters to answer the central question of how the trust process can provide insight into potential for harm for victims of ISA in different trust relationships. The results of the scoping review provided insights into concrete 22 and amorphous harms; looked at the role of identity and power in imposed vulnerability; identified potentially multiple focal vulnerabilities; and identified a dearth of research on institutional trust and harms. However, it did not provide insight into signals of trustworthiness or differences in emergent versus focal vulnerability. The victims’ accounts did provide insight into trustworthiness and differences in imposed and emergent vulnerability, but didn’t help clarify focal vulnerability, nor did harms map as clearly on to concrete and amorphous harms. The chapter concluded with a discussion of how insights into trust-based harm can help improve better coordinated responses for prevention, advocacy, and justice. In particular, the chapter presents a discussion of how existing legal resources could remedy the harms that were identified by the scoping review and victims’ accounts (and where gaps exist). 23 CHAPTER 2: UNDERSTANDING INTERPERSONAL AND INSTITUTIONAL TRUST AND HARM: A SCOPING REVIEW Abstract Institutional sexual assault (ISA) that occurs in trusting relationships with another person or an institution is related to distinct harms experienced by victims. However, ISA research doesn’t engage with the vast, dispersed, and multi-disciplinary research on trust to explain why these harms might emerge as distinct, or how they might differ for interpersonal or institutional trust. This chapter reports the findings of a scoping review of literature on the potential for harm that stems from different relationships of trust. A total of 50 different studies were identified that explored harms stemming from relationships of interpersonal and institutional trust. Both types of trust were associated with concrete and amorphous harms, and results provided insight into how different identities, contexts, and power in relationships, and mechanisms shaped potential for harm. Advocacy and prevention implications are discussed. Introduction Recent news has brought to light horrifying examples of sexual assault that occur in institutional settings or between institutional actors (e.g., in universities, churches, and the military) (Austen, 2022; Gerson, 2022; Svrluga, 2022). Part of what makes this ISA distinct is it is widespread, difficult to report, and relates to distinct harms for victims (Blakemore, et al., 2017; Crossmaker, 1991). To explain why harm might be distinct, betrayal trauma and institutional betrayal theories suggest that when abuse occurs in a trusting relationship with a person and or an institution (respectively) victims experience negative mental health, physical health, and relationship impacts distinct from sexual assault more generally (Birrell & Freyd, 24 2013; Platt & Freyd, 2015; Smith & Freyd, 2013). While trust is theorized to set this abuse apart, this literature doesn’t explain why trust might relate to these harms for victims. Trust has been explored in a variety of disciplines, including psychology, economics, philosophy, criminal justice, in ways that can provide insight into trust-based harm. Trust is typically defined as a willingness to be vulnerable to the actions of another, where vulnerability is the potential for harm (Mayer et al., 1995; Misztal, 2012). The limited research exploring vulnerability suggests it is both tied to factors outside of a person’s control, (e.g., natural disaster, identity) and relates to subjective awareness of harm from the actions of a trusted other (focal vulnerability) (Hamm & Mollering, 2022; Kuran et al., 2020; Montgomery et al., 2008). Building on that definition, a rich body of research on trust has explored different types of harm, and how these relate to different trustees (interpersonal or institutional) (see Table 1). Type of harm can range from concrete (e.g., demonstrable injury under the control of a trustee) to amorphous (e.g., failure to facilitate feelings of security, emotional harm) (Hamm & Wolfe, 2021). The type of trust relationship likely shapes the injury under control of the trustee: one might trust a person to do a distinct task (e.g., trusting a stranger to give correct directions) at risk of being disappointed, betrayed, or cheated (Baier, 1986). Alternatively, one may trust an institution to protect them from harm or protect their interest (e.g., financial interests) in exchange for complying with its rules and norms (Bachmann & Hanappi-Egger, 2012; Scott, 2010). Institutional trust increasingly is recognized as actually implicating multilevel trust, which is particularly relevant for victims of institutional sexual assault who may both trust a perpetrator and an institution to protect them from harm (Fulmer & Dirks, 2018; Kramer, 1999; Tan & Tan, 2000). 25 Clearly, a central tenet of both betrayal trauma and institutional betrayal theories is that violations of trust relate to distinct harms for victims (Birrell & Freyd, 2013; C. P. Smith & Freyd, 2014), but trust is merely presumed, and the ways trust makes this abuse distinctly harmful is not explored in depth. Examining the vast interpersonal and institutional trust literature can provide insight into these harms. However, trust research is broad, often siloed, and applies multiple interpretations of trust, struggling to come to consensus on definitions or concepts (Bachmann & Zaheer, 2013; PytlikZillig & Kimbrough, 2016). This research has not been systematically gathered with a goal of understanding types of harm(s) stemming from different trust relationships. The current research is a scoping review, intended to take stock of the broad research on harm stemming from interpersonal and institutional trust. Only when the state of the research on these harms is assessed can we inform literature on institutional sexual assault by getting to the heart of trust relationships that are betrayed. Literature Review Increasingly, research has identified the complex dynamics of ISA and corresponding impacts of the abuse on victims (Clark & Fileborn, 2011; Crossmaker, 1991). Importantly, ISA potentially implicates multiple potential relationships of trust that, when betrayed, relate to harms experienced by victims (Freyd, Klest, & Allard, 2005; C. P. Smith & Freyd, 2013). Research is still in the process of understanding these harms, but the theories of betrayal trauma and institutional betrayal suggest they are rooted in trust. Betrayal trauma theory suggests that sexual assault in an interpersonal trusting relationship (e.g., caregiver/child, priest/parishioner, teacher/student) is related to increased PTSD, depression, anxiety, dissociation, delay of disclosure, revictimization, and substance use (Delker & Freyd, 2014; Delker et al., 2018; Freyd, 1996). 26 Similarly, institutional betrayal theorizes that an institution can compound harms for victims of sexual assault by its action or inaction surrounding a sexual assault. This includes: not taking proactive steps to prevent this type of experience; creating an environment where abuse appears to be normalized; creating an environment where abuse seemed likely to occur; making it difficult to report; responding inadequately to the experience if reported; covering up the experience; punishing for reporting (C. P. Smith & Freyd, 2014). Institutional betrayal is related to PTSD, suicidality, difficulty trusting, and more (Monteith et al., 2016; Owen, Quirk, & Manthos, 2012; C. P. Smith & Freyd, 2017). It is the intersection of trust, institutional sexual assault, and harm that is central to understanding the experiences of victims of ISA. However, the trust relationships at the core of these theories has been largely neglected, presumed, or conflated with other concepts such as dependence or cooperation (Birrell & Freyd, 2013; PytlikZillig & Kimbrough, 2016; C. P. Smith & Freyd, 2014). Understanding Trust Trust as a concept has been explored in broad literatures (e.g., economics, philosophy, criminal justice), and while this literature has faced struggles to find consensus on definitions or concepts of trust, it can help provide essential insight into potential for harm (Kostis & Nasholm, 2020; PytlikZillig & Kimbrough, 2016; Welter, 2012). Across literature, trust has been conceptualized in many ways and as operating at different levels, including interpersonal trust, multilevel trust and institutional trust (A Fulmer, 2018; Rousseau et al., 1998; Welter, 2012). However, a commonly used definition of trust is “the willingness of a party [the trustor] to be vulnerable to the actions of another party [the trustee]”(Mayer et al., 1995). This definition helps identify both vulnerability and the reasons why one might accept that vulnerability (trustworthiness) as central to understanding harms that stem from that relationship (Baghramian 27 et al., 2020; Colquitt et al., 2007; Mayer et al., 1995; PytlikZillig & Kimbrough, 2016). Trust then leads to risk taking in a relationship, which has been conceptualized as increased dependence, decreased monitoring, or other behavior in reliance on the trustee (Mayer et al., 1995; Schoorman et al., 2007). Vulnerability is the state of being susceptible to some damage or harm (Misztal, 2012) and “one leaves another an opportunity to harm when one trusts, and also shows confidence they will not take it” (Baier, 1986, p. 235). Some of the limited research on vulnerability suggests it can be imposed, meaning it is related to factors outside of human control (e.g., natural disasters, intersectional identities) such that some groups (e.g., historically marginalized groups) might be more vulnerable to some types of harms because of these factors (Kuran et al., 2020; Montgomery et al., 2008). Vulnerability can alternatively be related to actions in control of a trustee—this is the type of vulnerability that is contemplated by most of the trust literature. Some recent theorizing has pushed for better understanding of the trustor’s subjective experience of their vulnerability, or, focal vulnerability, suggesting it is key to understanding the risk of harm one might contemplate in a trust relationship (Hamm & Möllering, 2022). Importantly, this focal vulnerability is relational, that is, is shaped by one’s relative power and position in society and is limited to the vulnerability that one is willing to accept to enter a trust relationship (Hamm & Möllering, 2022). However, trust is related to risk taking in a trust relationship, potentially creating the opportunity for additional harm within the relationship (Colquitt et al., 2007; Mayer et al., 1995). Conceptualizing Harm The first key to understanding harm is exploring how perceptions of harm differ across individuals. The range of potential harms might depend on the level of discretionary powers a 28 person or institution is trusted with and whether the power is unequal in the relationship (Bachmann, 2001; Baier, 1986; Möllering, 2019). Additionally, a person or institution might control multiple types of harm depending on the type of interaction. Broadly, these can range from concrete to amorphous, where concrete harms might include physical injury impacting just one individual trustor, whereas amorphous might include psychological harms that affect society more broadly (Hamm, Trinkner, & Carr, 2017). Some research has empirically examined these ranges of harm. A recent meta-analysis identified a range of themes in terms of harm that youth could experience from police, ranging from more concrete physical danger (injury or abuse), to amorphous (feeling police are ineffective, controlling, and prejudiced) (Nordberg, Crawford, Praetorius, & Hatcher, 2016). Research by Hamm and Wolfe (2021) builds on this conceptualization, suggesting harm can both range in concreteness and in terms of severity. In their research, concrete harms were linked to demonstrable injuries in control of police, including citations (less severe) and arrest (more severe) whereas amorphous harms included being misidentified as a perpetrator (more severe) and being the victim of a crime (less severe). Their results suggested that concreteness of harm and its severity did matter to the public’s preferences on procedural justice dimensions and shaped the importance of different trustworthiness signals from police. Additionally, perceptions may differ depending on the identity of the trustor, where individuals of color might perceive risk and potential harm from trusting police differently than other groups (Brunson, 2007; Brunson & Wade, 2019; Hamm & Wolfe, 2021). This reiterates the takeaway of imposed vulnerability research, suggesting identity and other factors shape perceptions of a range of harms. 29 Trustee The second key to understanding harm is understanding the type of trustee, where different trustees are perceived to control different risks (Hamm et al., 2019). Interpersonal trust refers to trust in a specific individual, and it is generally shaped by a trustor's propensity to trust and their personal interactions with that individual (Mayer et al., 1995; McEvily et al., 2006; Patent & Searle, 2019). This trust is developed in stages that are shaped by interactions in a range of types of relationships, including close romantic relationships, friendships, acquaintances, and business and professional relationships, where each bring potential for harm (Lewicki & Bunker, 1996). The process of deciding whether to accept vulnerability to another person is shaped by whether that person is deemed trustworthy. When a trustor assesses another person’s trustworthiness they focus on whether they appear capable of performing the task that they are trusted with (ability) whether they care about the trustor or people like the trustor (benevolence) and whether they adhere to a set of principles that the trustor finds acceptable (integrity) (Colquitt et al., 2007; Mayer et al., 1995; Poon, 2013). Different interpersonal relationships confer different benefits, and similarly create potential for different harms. In their study on betrayal in mateships (a heterosexual romantic relationship), friendships, and coalitions, Shackelford and Buss (1996) found that the greatest betrayal was perceived in the mateship context. Other research has conceptualized multiple types of harm in control of an interpersonal trustee, where harm can be internal or external (Hamm, Searle, Carr, & Rivers, 2021). For example, in trusting police, one can be vulnerable to external harm that police are responsible for protecting them from (e.g., crime) or internal harm (e.g., harms that arise from the actions of police) (Hamm et al., 2021b). Finally, some research has explored the differences in interpersonal trust betrayal whether it is perceived to be intentional or 30 incidental (the goal of the perpetrator is not to harm the victim, but it is a byproduct of their action) (M. L. E. Chan, 2009; Morris & Moberg, 1994) This limited research flags differences in vulnerability depending on dynamics of a relationship that can give insight into differences in harm when trust is betrayed. Existing literature suggests there are separate processes for developing trust in an institution. Institutional trust has been conceptualized numerous ways in the literature, exploring this as trust in a collective of individuals or as trust in the overarching system or framework in which a specific relationship is operating (Bachmann & Inkpen, 2011; Kramer, 1999; McEvily et al., 2006). Assessing trustworthiness of an institution relies less on personal interactions. Some research has explored how one perceives the trustworthiness of an institution using ability, benevolence and integrity perceptions (Pirson & Malhotra, 2011; Schoorman et al., 2007) while others argue there are different processes entirely for developing trust in institutions (Harris, Keevil, & Wicks, 2013; McEvily et al., 2006). Instead of ability, benevolence and integrity, people are theorized to rely more on structural factors such as legal provisions, reputation, certification of exchange partners and community norms in deciding to trust an institution (Bachmann & Inkpen, 2011). Research on institutional trust has struggled to differentiate between trust in aspects of an organization (e.g., leaders, coworkers) and trust in the organization or institution. Tan and Tan (2000) found that trust in supervisors was shaped by different antecedents (ability, benevolence, and integrity) while trust in organization was shaped by more structural factors, such as procedural and distributive justice. Additionally, trust in individuals within an organization might be related to trust in the organization as a whole, where employees’ trust in coworkers and their perceptions of an organization predicted trust in an organization (Tan & Lim, 2009). 31 Increasingly research suggests that considering multilevel trust is essential to truly understand trust in institutional contexts (Fulmer & Dirks, 2018; Lumineau & Schilke, 2018). Finally, separate benefits, and harms are associated with institutional trust. Unfortunately, the extent and type of harm that arises from the institutional trust relationship is unclear. In some research on political trust, harms tied to the deliberate actions of the government and could include persecution, impediments to a citizen’s access to resources or means, or intrusion into private areas of their lives (Hamm, Smidt, & Mayer, 2019). Organizational trust research suggests that an institution could cause harm when they fail to fulfill perceived obligations to it members, which constitutes a psychological contract breach (Kramer, 1999; Morrison & Robinson, 1997; Robinson, 1996). Psychological contract breach was related to harms associated with emotional impact (e.g., anger) (Morrison & Robinson, 2006). This type of harm may change depending on the stage of the trust relationship. Robinson’s (1996) study examining perceptions of 125 newly hired managers found that when they entered employment with trust in the employer, they perceived violations of the psychological contract through unmet expectations (a vulnerability that only occurred once they entered the organization). Importantly, multiple trustees can potentially be sources of harm in the context of an institution or organization (e.g., police). For example, harm can stem from individual officer-level decisions or institutional agency-level decisions (e.g., policy), suggesting that for institutional trust, some multilevel harm might be implicated (Hamm, Searle, Carr, & Rivers, 2021). Scoping Reviews The trust research is dispersed across disciplines, siloed by type of trust, and has struggled to settle on consistent definitions or concepts (PytlikZillig & Kimbrough, 2016; Welter, 2012). To use this research as a lens to understand harm that stems from trust, a systematic synthesis of 32 this research is necessary. A scoping review is a relatively new methodology that is gaining popularity as a method of knowledge synthesis in Criminology and Criminal Justice (Callan et al., 2021; Grant & Booth, 2009). This type of review is traditionally undertaken to address a broad research question and gain a clearer view of the scope or scale of existing research with a goal to provide the information to interest groups (e.g., policy makers, practitioners, researchers) (Arksey & O’Malley, 2005). They are particularly useful to clarify key concepts or conceptual boundaries that underpin a research area (Peters et al., 2015). This is accomplished by mapping out concepts across a diverse literature in order to uncover gaps in that knowledge (Arksey & O’Malley, 2005; Mays, Pope, & Popay, 2005; Moher et al., 2009). Scoping reviews present a rigorous and distinct methodology to answer the research questions compared to other methods of literature review. Where a traditional literature review may or may not include comprehensive searching, focuses on more recent literature, and is more susceptible to bias (Grant & Booth, 2009), a scoping review follows a structured approach to identifying, searching, and screening literature and to analyzing the data and is reproducible (Tricco et al., 2018; Wilson & Lemoine, 2021). This type of review is well-suited to address broader topics where many types of study designs might be applicable, as opposed to other types of systematic literature review (e.g., meta-analyses), which are instead better suited to focus on well-designed questions from a narrow range of quality-assessed studies (e.g., evaluations) (Arksey & O’Malley, 2005). Finally, “scoping studies often require 'sense-making' across fields of enquiry that are complex and which lend themselves to interpretation through many academic and theoretical disciplines” (Anderson, Allen, Peckham, & Goodwin, 2008, p. 6), making it well- suited to examine broad concepts that have not been explored together (i.e., trust and harm). 33 There are five steps in a scoping review according to Levac, Coloquhoun and O’Brien (2010). (1) Identify the research question: the question will be broad in scope but must contain enough clarity and focus to provide a roadmap and inform the later stages of the research process. (2) Identify relevant studies: this stage requires comprehensiveness and breadth as built into the search strategy and involves developing a plan for where to search, which terms to use, types of sources to be included and year or language limitations. These limitations can be by concept, context and population, and will be guided by the research questions (Peters et al., 2015). (3) Selecting studies: this stage refers to an iterative process of searching the literature, refining search strategy, reviewing studies and developing inclusion and exclusion criteria. (4) Chart the data: this includes collecting uniform data from each article in the final sample. These can include details about the sample, method, county of research, terms, and/or narrative or descriptive summaries. Finally, (5) collate, summarize and report the results: this is where thematic analysis method is used to sort the articles into themes which provide an overview of the literature and a narrative of the results (Levac, Colquhoun, & O’Brien, 2010). Current Study Harms experienced by institutional sexual assault victims are theorized to relate to relationships in which trust is betrayed. Trust research suggests that understanding harm requires an exploration of vulnerability, reasons to accept that vulnerability, and potential for harm that stems from trust. Some research has conceptualized harm as concrete or amorphous, where concreteness maps on to a demonstrable injury under control of the trustee and amorphous harms are less linked to demonstrable injury, and map on to more disperse, less specific actions within the control of the trustee (e.g., failure to facilitate feelings of security, emotional harm). Because there are important differences in vulnerability and signals of trustworthiness for institutional 34 versus interpersonal trust, it is likely that there are differences in amorphous or concrete harm depending on the type of trustee, but that has not been explored by the literature (Bachmann, 2011; Bachmann & Hanappi-Egger, 2012; Colquitt et al., 2007; PytlikZillig & Kimbrough, 2016). Only by bringing together disparate strains of literature exploring trust, harm, and different trustees can we begin to fill this gap in our understanding. This scoping review addresses three research questions: (1) What is known about potential harms associated with interpersonal trust? (2) What is known about potential harms associated with institutional trust? and (3) What are the limitations and gaps of the knowledge base? Method Data Collection and Analysis (1) Identifying the Research Question. Trust has been theorized to be the mechanism in betrayal trauma and institutional betrayal that creates the potential for distinct harm when trust is violated. A wide literature on trust exists across disciplines that helps understand decisions to accept vulnerability by looking at type of harm and trust relationship. However, while definitions of trust imply some level of risk to harm (Mayer et al., 1995) the specific harms are not explicitly discussed. Therefore, the research questions sought to identify the potential for specific harms that are identified in the trust literature for interpersonal and institutional trust and identify gaps in this research. (2) Identifying Relevant Studies. The PRISMA extension for completing scoping reviews provided an exhaustive checklist which guided the data collection and coding (Tricco et al., 2018). The PRISMA flowchart is presented in Figure 3 and summarizes the study identification and selection process. The systematic review software Covidence was utilized to aid in data collection and coding. To identify relevant articles for extraction and thematic analysis, a two- 35 stage data collection and coding process was performed. This included (1) a search of relevant databases and (2) identification of non-traditionally published studies (often called “gray literature”) via other methods. For database search, in consultation with a staff librarian, five potentially relevant databases were selected: Criminal Justice Database; ProQuest Dissertation and Theses; Psych Info Including Psych Articles; Web of Science Core Collection, and Sociological Abstracts. The literature on trust was used to draft search terms that were then tailored to each database’s search norms to capture both interpersonal and institutional trust (including trustworthiness and vulnerability) and any potential for harm or damage that might stem from that trust. Searches were performed in August of 2021 and were iterative to test and develop the most effective terms to answer the research questions. The final search strategy can be found in Figure 4, but an example of the search terms used are: trust, (trustworth* OR vulnerab*) harm OR damage OR risk OR "potential negative outcome" OR "adverse effect" OR "adversely affect" (title) or harm OR damage OR risk OR "potential negative outcome" OR "adverse effect" OR "adversely affect" (abstract). To fully explore the relationships, no publication date limit was placed on articles in the available literature. The final search results were exported into Covidence, and duplicates were removed by the software, resulting in 2,883 references imported. In addition to identifying relevant studies through a search of relevant databases, identification of studies via other methods was also necessary (Tricco et al., 2018). Two distinct search strategies were used: (1) citation chaining and (2) hand searching. Citation chaining required identifying articles that are relevant to the research question and then collecting all the research that they referenced, as well as all the later research that cited the original article. To identify articles that were highly relevant to understanding harm that stemmed from trust 36 foundational pieces that served as the basis of the dissertation (e.g., Mayer, Davis, & Schoorman, 1995), articles identified in the database search (e.g., Chan, 2009), and prior knowledge of research (e.g., Nienaber, Hofeditz, & Romeike, 2015) were relied upon. In total, 8 articles3 were identified and collected, as well as the research cited in their references and all research citing those articles. In hand searching, four trust-related journals were searched to identify articles: Academic Management Review (AMR), Academy of Management Journal (AMJ) , Journal of Trust Research, and Risk Analysis. Two of the journals (AMJ and AMR) also allowed access to conference proceedings, insights, and annals which were included in the search. General search terms (e.g., “trust AND harm” or “trust AND injury”) were used to yield the largest number of possible articles (Tricco et al., 2018). For both citation chaining and hand searching, after duplicates were removed, a total of 519 articles were imported into Covidence. (3) Selecting Studies. Once articles had been identified by database search and other methods, they were separately screened by title and abstract in Covidence. Inclusion and exclusion criteria helped guide the screening process to yield the final articles for extraction. As 3 Bachmann, R., & Inkpen, A. C. (2011). Understanding institutional-based trust building processes in inter- organizational relationships. Organization Studies, 32(2), 281-301. Baghramian, M., Petherbridge, D., & Stout, R. (2020). Vulnerability and trust: An introduction. International Journal of Philosophical Studies, 28(5), 575-582. Chan, M. E. (2009). “Why did you hurt me?” victim's interpersonal betrayal attribution and trust implications. Review of General Psychology, 13(3), 262-274. Kuran, C. H. A., Morsut, C., Kruke, B. I., Krüger, M., Segnestam, L., Orru, K., Naevestad, T. O., Airola, M., Keranen, J., Gabel, F., Hansson, S., Torpan, S. (2020). Vulnerability and vulnerable groups from an intersectionality perspective. International Journal of Disaster Risk Reduction, 50, 101826. Landa, N., Zhou, S., & Tshotsho, B. (2019). Interrogating the role of language in clergy sexual abuse of women and girls in Zimbabwe. Journal for the Study of Religion, 32(2), 1-20. Mayer, R. C., Davis, J. H., & Schoorman, F. D. (1995). An integrative model of organizational trust. Academy of Management Review, 20(3), 709-734. Nienaber, A. M., Hofeditz, M., & Romeike, P. D. (2015). Vulnerability and trust in leader-follower relationships. Personnel Review.44(4), 567-591. Smith, C. P. (2017). First, do no harm: institutional betrayal and trust in health care organizations. Journal of Multidisciplinary Healthcare, 10, 133. 37 to inclusion criteria, only papers with an English language abstract that discussed the potential for harm that stemmed from a trust relationship were included. Because the aim of the scoping review is to identify a broad array of relevant research, multiple types of research design were included so there were no exclusion criteria related to research design. The exclusion criteria were guided by the research question and were continually updated during screening. Exclusion criteria typically fall under three categories: participants, concept, and context (Peters et al., 2015). Participant-related exclusion criteria require that relevant characteristics of participants are identified which match the research question and identify them as appropriate for the review, for example, participants were limited to individual trustors because exploring potential harm in the context of victims who trust 4 was desired (Peters et al., 2015). Next, concept criteria address the fit to the phenomena being explored (Peters et al., 2015). In this scoping review, the phenomenon of interest is harm that might result from the willingness to accept vulnerability inherent in trust. Five final exclusion criteria related to the concept were developed iteratively to omit articles that did not match the type of harm and vulnerability of interest (e.g., articles that merely discussed harm but did not engage with trust; articles on medical vulnerability to disease). Finally, context will vary based on the objectives of the scoping review and may include consideration of the setting or discipline (Peters et al., 2015). In this scoping review, exclusion criteria around context were shaped by definitions of trust that formed the foundation of this dissertation (i.e., Bachmann & Inkpen, 2011; Mayer et al., 1995) which define trust as a psychological state. Thus, other models of trust (e.g., economic trust games) were excluded. After applying exclusion criteria at the abstract and title stage, a total of 181 studies were identified as relevant from database search methods and 77 were 4 A full list of exclusion criteria and number of articles omitted under each criterion is provided in Figure 3. 38 relevant from the other search methods. These were combined and when full texts were then assessed for eligibility, a total of 50 studies were included for final extraction. (4) Charting the Data. The next stage involved extracting data uniformly from each article and charting, summarizing, and reporting the findings (Levac et al., 2010; Tricco et al., 2018). Extracted information included attributes of articles such as the year each article was published, title, authors, country in which research was performed, type of trustor, specific trustee, study design, etc. A template of the data charting form was created using the Covidence software and is available in Figure 5. (5) Summarizing and Reporting the Findings. Finally, a thematic analysis was applied to identify prominent or recurrent themes in the articles and summarize findings related to these themes (Dixon-Woods, Agarwal, Young, & Sutton, 2005; Thomas & Harden, 2008). Thematic analysis is a method used to inductively identify, analyze, and report patterns across a data set that capture something important in relation to the research question(s) (Braun & Clarke, 2006). The six stages of thematic analysis identified by Braun & Clarke (2006) were utilized, which entailed becoming familiar with the data, developing initial codes, searching for themes, reviewing the themes, defining and naming themes, and then producing the report. Familiarizing myself with the data occurred at the abstract and full text review stages. In these stages initial codes were developed as “tags” within the Covidence software (e.g., “negative health outcome”), which helped flag similarities between articles. Next, when the final 50 articles were selected, a return to the data set allowed a search for themes, which were then compared against the tags/codes. Central to this method was identifying themes inductively (which began at the full text review stage) and then using an iterative process to continually return to the to the data set (at the charting data and summarizing stages) and 39 updating themes to reflect the trends identified. Next, the themes were reviewed, which required thinking about the relationship between codes and themes and different levels of themes, which ultimately resulted in identifying meta themes, general themes, and sub-themes. In this stage themes were also edited to better reflect the articles encompassed in the theme (e.g., “health”). Finally, the themes were organized narratively and the results drafted. Utilizing this method meant the themes were grounded in the data without trying to fit them into a pre-existing coding frame (Braun & Clarke, 2006). To ensure reliability, memoranda were created at all phases of thematic analysis (Yin, 2015). Results Study Context Table 2 provides details about attributes of the final sample of articles. As one of the inclusion criteria, all articles necessarily involved an individual trustor. As to trustees (the person or entity being trusted), most of the articles (N = 39) examined potential for harm from an individual trustee (i.e., interpersonal trust). A total of 2 articles did not fall directly into interpersonal or institutional trust and these were considered multilevel trust. Finally, there were 9 articles that explicitly examined potential harm from solely an institutional trustee. The specific trustees considered in the studies ranged from general (e.g., “others”) to specific (e.g., firefighters, students, police, patients, romantic partners). Almost half the studies (N = 23) were conducted in the United States, but other locations included Australia, Canada, Brazil, China, Jamaica, Kenya, Pakistan, Zimbabwe, South Africa, and the UK as well as multinational sites. Study Design In terms of study design, a total of 14 articles relied on a solely quantitative methodology while 5 articles utilized mixed methods (here, meaning both qualitative and quantitative methods 40 or proposing a theoretical model and some empirical application of theory). A total of 11 articles were qualitative studies and 2 articles were either traditional or systematic literature reviews. The most prevalent study design (N = 18) was “theoretical” examination, meaning it proposed or engaged with a theory or model or contained a non-empirical discussion of potential harm stemming from a trust relationship. Combined, most of the articles applied some sort of empirical examination of harm that stemmed from different trust relationships, and the majority of these were exploring potential harm that stemmed from interpersonal trust. Themes Overall, meta-themes around concrete and amorphous harms help understand the multi- faceted ways harm might emerge depending on relationship and context. Within these meta- themes, a total of 11 themes arose inductively, the vast majority of which (8 of 11 themes) arose from articles on interpersonal trustees. The concrete harms for interpersonal trustees included abuse, health, financial, and multiple harms (Table 3). The amorphous harms included emotional, professional, security/privacy, and relational harms. A theme of multilevel harms arose inductively, which implicated both interpersonal and institutional harms and trustees. Finally, for institutional trustees, the only concrete harm was financial, and the amorphous harm was emotional. Interpersonal Concrete Harms The first type of concrete harm associated with interpersonal trust related to abuse (N = 5). Within this theme, a subtheme of exchange arose, which implicated a trustor offering up information that could be used abusively to foster a relationship of trust. Exchange could look like offering up one’s safety to others during a cultural ritual, such as moshing (“a seemingly aggressive colliding of bodies that occurs with members of the audience at concerts”) (Palmer, 41 2005, p. 147). In this situation, the participants trust other individuals to be merely acting out parodies of violence to allow cathartic expression of repressed motives and facilitate entrance into an insular community (Palmer, 2005). A different view of exchange involved willingly offering something that could be used abusively (e.g., details on past crimes/deviance, self-made private sexual images) to facilitate friendship or intimacy (Amundsen, 2022; Flashman & Gambetta, 2014). The potential for harm was explicitly tied to vulnerability, where a participant described: “I think sending like explicit pictures, you have to have like ultimate trust with the person that you’re sending them to. [ ... ] they’re so like vulnerable and they’re so intimate that it’s like; it’s a lot to think about” (Amundsen, 2022, p. 122). Physical risk was a second subtheme for abuse that relied heavily on vulnerability tied to the context of the trust relationship. This was relevant to individuals engaged in particularly risky occupations, where Pratt and colleagues (2019) found that trust in other firefighters related to potential physical injury in the line of duty. To cope with the leap of faith that was necessary to facilitate trust for a fire station to function, they found supporting dynamics (telling stories about fighting fires) and sustaining dynamics (ignoring or not seeking out new information about colleagues) helped firefighters take the leap of faith and accept the physical risk of harm in control of the colleague (Pratt et al., 2019). While those harms might have been at the forefront of firefighter decision-making, other contexts made individuals less aware of physical risk. Landa and colleagues (2019) explored how emotional or religious dependency and trust in clergy created the potential for sexual abuse for women. The women and girls may not have been as aware of the risk of abuse in entering trust, but situations such as poverty, unemployment and the worsening economic environment in the country created imposed vulnerability (Landa et al., 42 2019). The articles within the abuse theme suggest that context and multiple vulnerabilities shaped these harms. Related to physical injury, physical health impacts were identified as a potential harm stemming from interpersonal trust (N = 6). Most of these articles concluded that increased trust related to decreased condom use, which increased the potential for harm in the form of sexually transmitted infections, unwanted pregnancy, or HIV (A. E. Bailey & Figueroa, 2018; Bhana & Pillay, 2018; De Aguiar & Camargo, 2014; Luchters et al., 2013; Maia, Guilhem, & Freitas, 2008). Different contexts created different vulnerability, and a population that was identified as particularly high risk of potential harm through the negative health outcome of HIV were female sex workers (A. E. Bailey & Figueroa, 2018; Luchters et al., 2013). Additionally sexuality and gender, having a common law relationship, or being in love were contexts that were tied to trust and decreased condom use, creating vulnerability to health-related harms (Bhana & Pillay, 2018; De Aguiar & Camargo, 2014; Maia et al., 2008). A different concrete harm from interpersonal trust was financial harm, which was present in 6 articles. Of these, 4 articles aligned with traditional economic literature and explicitly tested fraudulent acts as outcomes of trust or trustworthiness perceptions (e.g., investing money, being cheated, or weight loss/other scams) (P. E. Bailey et al., 2016; Butler, Giuliano, & Guiso, 2016; Castle et al., 2012; Judges, et al., 2017). However, slightly different contexts shaped different financial harms, where trust in a salesperson related to potential financial harm for a customer through quality of the product (Brockway, 1993). Alternatively, citizen trust in an individual CEO could have far reaching financial impacts, as high CEO trust was tied to less diligence by financial analysts, leading to fraud and dropping stock prices (Gu, Liu, & Peng, 2022). 43 As with the above themes, one article hypothesized that financial harm could be tied to vulnerable identity or situation, but this research came to different conclusions. Engaging with the potential for harm that exists due to vulnerability in ageing, Judges and colleagues (2017) explored a sample of older adults, focusing on their unique potential for financial harms. These harms were measured as “scams” (i.e., weight loss scam, miracle health product scam, prize or lottery fraud, fraudulent work-at-home programs, charity scam, credit repair fraud, fraudulent business opportunity, advance free loan scam, counterfeit check scam) (Judges et al., 2017). However, their hypothesis that trustful individuals would be more inclined to believe others and put themselves at increased victimization risk was not supported (Judges et al., 2017). Thus, financial harm emerged as a concrete harm but the hypothesis that some vulnerable identities (i.e., age) played a strong role was not universally supported. Finally, there were 6 articles that discussed multiple concrete potential harms that stem from a trust relationship. In Chan (2009)’s article, these factors emerged as a theoretical framework where multiple harms were theorized to differ depending on the type of trust violation. If the trust violation was incidental compared to intentional this was theorized to relate to different harms (e.g., broken promises, stealing others’ ideas and claiming credit for it) (M. L Chan, 2009). Alternatively, exploitation (cognitive and emotional, or exploitation in numerous contexts) was linked to injuries to numerous things we value (e.g., injuries to life, health, reputation, shared goods, theater, market exchange) (Baier, 1986). Again, as with the above themes, identity and context shaped which concrete harm(s) trustors were aware of, (e.g., exploring age, and how that intersects with global pandemic restrictions; exploring relationships in medical teaching institutions) (Damodaran, Jones, & Shulruf, 2021; Han & Mosqueda, 2020). Interestingly, Han and Mosqueda (2020) proposed a model that considered increased dependency 44 of elderly people on others because of COVID-19, which confined them to their homes, adding a level of vulnerability beyond advanced age, and thus creating higher potential for abuse by a trusted other. Where multiple potentials for harm exist, different vulnerabilities may be more or less salient depending on the context. Interpersonal Amorphous Harms Interpersonal trust created the potential for emotional harm (N = 8) which was associated with betrayals that stemmed from specific trust relationships. These articles suggested that betrayals or breaches damaged some sense of security, future trust, or health outcome (Bearman & Molloy, 2017; Bianchi, 2015; Gobin, 2012; Huddy, 2015; Koehler & Gershoff, 2003). Providing insight into the process by which emotional harm can occur through betrayal, Bearman and Molloy (2017) suggested that teachers should purposefully express vulnerability through sharing with students their confusions, mistakes, and deficits as a pedagogical strategy (“intellectual streaking”). In portraying their vulnerabilities and thought processes (compared to polished final products) it puts them at risk of emotional harm in the form of loss of face, but, in doing so, they match the vulnerabilities they are asking students to put forth in trusting their teachers (Bearman & Molloy, 2017). Additionally, harms tied to betrayal were linked to the closeness of the relationship. Gobin (2012) examined a sample who had experienced betrayal trauma by exploring differences in closeness to the trustee and found that betrayals in close relationships (high betrayal trauma) disrupted socio-emotional functions including the ability to judge trustworthiness and might be related to increased risk for revictimization. Asymmetry within interpersonal relationships uniquely related to emotional harm and betrayal in these articles. In both Huddy (2016) and Koehler and Gershoff (2003), betrayal in a relationship of dependence or protection was linked to unique harms. Namely, Koehler and 45 Gershoff (2003) explored how trust may be uniquely positioned to create potential for harm between those who have a professional duty to protect and those who were receiving that protection. The authors conducted five studies exploring different types of betrayal (criminal betrayals, safety product betrayals, and the risk of future betrayal by safety products) and found that people react more strongly with negative emotions when trust in a relationship of duty to protect is betrayed (Koehler & Gershoff, 2003). A way to consider harm in another asymmetrical relationship was as “emotional risk,” looking at students’ trust in business professors which placed them at risk of emotional harm if it was betrayed (Huddy, 2015). Emotional injury was also associated with loss of identity, sense of self, and connection to others, which comprised the theme of moral injury5 (N = 4). In these articles, trust was related to mutual recognition of personhood, and damage to this trust resulted in emotional injury and damage to identity (Bernstein, 2011; Brennan, 2021; Rotenberg, 2010). Brennan (2021) provides insight into how moral injury as a mechanism operates differently from the articles on betrayal. When moral injury occurs, there are three tiers of harm: the betrayal or injury, the loss of security and connection to others, and finally an injury to self-image, and these injuries degrade the victim’s sense of her own value that results from the loss of security (Brennan, 2021). Similar to the betrayal theme, asymmetrical relationships emerged as important to understanding emotional harm in these articles. Kutsyuruba and Walker (2017) explored specific harms that stemmed from trust in leaders (e.g., damage to the civic order, damage to sense of identity) which related of different types of trust violation (e.g., breach of rules and norms governing 5 The type of harm that arose was different from abuse even though there are some similarities to “abusive exchange,” which had included offering information that would create the potential for harm to facilitate intimacy or friendship. Namely, some of the literature in exchange referred to a risk of loss of face or emotional consequences to the information being shared (e.g., sexts being disseminated) but exchange implicated some more concrete (possibly criminal) consequence than emotional harm and was differentiated from solely emotional harm here. 46 behavior, honor violations). The asymmetry of power is central to understanding harm, where those in “authority and leadership can and do betray the trust of their constituents in highly destructive ways” (Kutsyuruba & Walker, 2016, p. 135). In some interpersonal trust, a trustee’s behavior could violate some notion of privacy or professionalism, for example, by sharing some private information which would have harmful consequences (N = 2) or cause some professional damage (N = 2). Privacy concerns were related to sharing of personal information during contact tracing in the COVID-19 pandemic (Chen et al., 2021) and trusting mental health professionals who were also mandatory reporters (Sapiro, 2020). Youth described how trust in a mental health professional opened them to risk of being taken from their parents’ care, or other behaviors that felt punitive, where a participant said: “I feel like there’s a lot of fear of therapists and psychiatrists, and like you say this and this will hurt your life forever” (Sapiro, 2020, p. 6). Kramer (2012) concluded that unique professional harms stemmed from trust in an interpersonal trust relationship in higher education (e.g., status and reputation). However, Kerler and Killough (2009) presented an important negative case for professional harm stemming from trust, concluding that accountants’ trust in an audit client does not degrade the ability of an accountant to perform their professional duties. Finally, relational trust explicitly discussed how amorphous harms stem from reciprocal interpersonal trust relationships (where both parties trust one another) and therefore both have potential for harm(s). These articles draw on the growing literature on reciprocal trust, which refers to “the trust that results when a party observes the actions of another and reconsiders one’s trust-related attitudes and subsequent behaviors based on those observations” (Serva, Fuller, & Mayer, 2005, p. 625). Relational trust implicates some of the same emotional risk identified in betrayal. For example, Novack (2018) explains that there is a potential to harm to “psychic 47 safety” when an analyst enters into a trust relationship with a patient. Bianchi (2015) engaged with the reciprocal relationship of trust between art buyers and sellers and found object betrayal can be as acute as interpersonal betrayal, and both are implicated in the reciprocal trust of art markets. Again, power imbalances shaped vulnerability in these relationships, and the articles show that this may occur reciprocally in ways that shape distinct harms. For example, Carr and Sandmeyer (2018) explored how a relationship of trust between a patient and analyst entails an interplay between both safety and danger for both parties, but a willingness to enter and engage in this relationship can promote healing and growth for both parties. Bianchi (2015) discussed that in art markets, there is an imbalance in terms of access to information (where sellers have increased access compared to buyers) and Buchman and Ho (2014) explored an asymmetrical relationship between physicians and patients. Buchman and Ho (2014) explored the use of “opioid contracts,” which describe conditions a patient must meet to be prescribed opioids to prevent opioid abuse. This reciprocal trust creates physician vulnerability (when they trust a patient’s symptoms are truthful) to professional and legal consequences, public health risks, and related concerns such as patient violence. It also creates patient vulnerability when they trust in a physician’s diagnosis and treatment plan to “iatrogenic harms,” which refer to harms caused by medical treatment (Buchman & Ho, 2014). The articles on relational trust uncovered complex and interdependent harms, some of which also implicated power asymmetry in relationships. Multilevel Harms Some articles did not fall neatly into interpersonal or institutional trust but pushed the conversation about harm in ways that are applicable outside their specific contexts. Hamm and colleagues (2021) sought to unpack vulnerability implicated in trust in police. Namely, they 48 differentiated between external harm (arising from deliberate acts of others, such as criminals) and internal harm (arising from deliberate acts of police) and concluded that individuals with minority identities evaluated harm differently than those in majority identity groups. Wolfensberger (2016) engaged broadly with trust in medicine, and theoretically explored how betrayals are associated with “disappointments of moral expectations” but concluded that most people cannot disentangle targets of trust (i.e., trust in physicians compared to trust in medical institutions or the broader medical field) (Wolfensberger, 2016, p. 52). While these multilevel articles may have implicated some of the same harms discussed in the articles above (e.g., emotional harm), they expand the conversation by looking outside a specific context and engaging with multilevel trust and harm. Institutional Concrete Harms Only financial harms were identified as concrete harms stemming from institutional trust. Bachmann and Hanappi-Egger (2014) presented a discussion of how an “unhealthy” form of institutional trust (without regulations) created the risk of fraud and wide-reaching harm in the form of the global financial crisis (Bachmann & Hanappi-Egger, 2012). Part of their solution involved promoting institutionalized control which should promote a level of distrust that keeps actors wary of fraud (Bachmann & Hanappi-Egger, 2012). Similarly, Dupont and Karpoff (2020) discuss the threat of fraud and opportunism and the potential impacts on financial market activity suggesting the “trust triangle” as a solution (Dupont & Karpoff, 2020). The trust triangle sets out the pathway by which people develop trust and relies on similar types of institutionalized control to Bachmann and Hanappi-Egger (2014), namely, market forces and relational capital; laws, institutions, and regulators; and personal ethics, integrity, and culture (Dupont & Karpoff, 2020). 49 This institutional trust has far-reaching financial impacts when violated, and thus, relates to more institutional-level suggestions for prevention. Institutional Amorphous Harms Amorphous harms were associated with trust in institutions through institutional betrayal. In the articles here, Smith and colleagues (2017; 2013, 2014) built on institutional betrayal theory to explore potential harms that stemmed from this relationship. The authors found that betrayal by a trusted institution exacerbated what was already a traumatic experience for women in an undergraduate sample (2013) and that betrayal might occur through commission or omission and be either isolated or systemic (2014). Specifically, the harms were to psychological well-being and included dissociation, anxiety, sexual disfunction, and other trauma-related outcomes (C. P. Smith & Freyd, 2013, 2014). However, these did not engage with the mechanisms for harm, merely applying institutional betrayal theory without exploring trust or sources of harm. A different vein of articles expanded institutional betrayal outside the context of sexual assault to provide more insight into the mechanisms of harm. In a sample of adult patients in a healthcare setting, Smith (2016) found that when patients realized their physician had acted counter to their best interests or taken advantage of them (what the author calls a betrayal of fiduciary trust) they may experience an emotional response rooted in moral outrage. Similar to the results in interpersonal trust (Gobin, 2012), betrayal also related to more tangential effects: patients who had experienced institutional betrayal reported lower trust in their own physicians, doctors in general, healthcare organizations, and worse physical health (Smith, 2016). French and colleagues (2022) and Brewer (2021) explored institutional betrayal in clinical staff who worked during the COVID-19 pandemic and nursing, respectively. Brewer (2021) performed a 50 concept analysis exploring the extent to which research had explored institutional betrayal in nursing, finding harms associated with nurses’ trust in employers included hostile management, unsafe working environments, or gaslighting nurses who might experience negative events. This work gives insight into the mechanism of harm, where institutional betrayal explicitly “differs from moral injury” ( p. 1088). Instead, institutional betrayal aligns with psychological and ethical distress—for example, when institutional actions resulted in constraints to nurse behavior (meaning they could then not provide adequate patient care), nurses felt distress. Finally, some concerns around institutional betrayal were uniquely linked to security and confidentiality harms at an institutional level. Namely, Kraft and colleagues (2018) used focus groups to explore trust in the research enterprise and identified concerns about use of research to affirm racist stereotypes and deny access to healthcare later due to racial discrimination and profiling. This perspective identified fears that an institution (e.g., research institution with medical research information) may be believed to influence larger societal or cultural processes through research. As might be expected, the institutional betrayal articles largely focused on institutional level harms. Discussion Summary of Evidence The present study set out to identify how the literature had explored how trust in interpersonal and institutional trust relationships creates potential for harm. Much of the research focused on potential harms in interpersonal trust relationships compared to institutional trust (only 9 of the 50 articles grappled with institutional trust). This trend aligns with gaps that have been identified in this literature and calls for more multilevel or institutional trust research in the larger trust literature (Bachmann, 2011; Fulmer & Dirks, 2018). 51 Second, overall, meta-themes around concrete and amorphous harm helped understand the multi-faceted ways harm emerged differently based on relationship. Namely, more concrete harms emerged from interpersonal trust, and this was more dependent on context and identity as shaping vulnerability, while more amorphous harms emerged under institutional trust, and these built heavily on institutional betrayal. However, perhaps more concrete harms arose for interpersonal trust (e.g., injury, financial or health impacts, or multiple concrete harms) because they were linked to context-specific interpersonal trust relationships (e.g., sex work, firefighters, salespeople, art buyers) in ways the literature has not acknowledged. Concrete Harms and Vulnerability For interpersonal trust, identity (e.g., A. E. Bailey & Figueroa, 2018; Han & Mosqueda, 2020; Landa et al., 2019) and various contexts (e.g., Brockway, 1993; Pratt et al., 2019) closely shaped vulnerability. Namely, these results add to our understanding of imposed vulnerability, that is, vulnerability not tied to the actions of a trustee (Montgomery et al., 2008). The interpersonal concrete harms of abuse, negative financial impacts, health impacts, and multiple harms differently implicated identity (e.g., age, gender, and status in society). This aligns with prior literature suggesting that identity matters, where historically oppressed groups of individuals might view harms in different situations differently for dimensions of procedural justice and trustworthiness (Hamm & Wolfe, 2021). These results additionally align with some of the growing literature on intersecting vulnerabilities (Kuran et al., 2020) and suggests the literature should spend time on dissecting the salience of these vulnerabilities. Adding a different dimension to imposed vulnerability, the results on concrete harm suggested that context of the trust relationship was important to shaping imposed vulnerability. For example, jobs at risk of physical harm (i.e., firefighting), cultural norms, and purposeful 52 exchanges of material that could be used abusively (e.g., past deviance) all shaped vulnerability as closely tied to the context in which one was trusting another person (Flashman & Gambetta, 2014; Palmer, 2005; Pratt et al., 2019). Identity and context were less prevalent in the institutional group, where the only concrete harm was financial. Thus, explicitly examining imposed vulnerability in institutional trust is an area for future research, especially because institutions are linked to potential for societal harms (i.e., research used to perpetuate racism). These results also shed light on which vulnerability might be focal, that is, the subjective and relational vulnerability one is aware of in deciding whether to enter a trust relationship (Hamm & Möllering, 2022). For interpersonal trust, participants could identify and anticipate concrete vulnerability to harm in certain contexts (e.g., shared private images, physical harm) more easily than others (e.g., sexual abuse). It could be that vulnerability to harm was at the forefront due to an exchange relationship, where some articles discussed engaged in trading vulnerability very purposefully to garner intimacy or friendship, suggesting context is important to understanding focal vulnerability (e.g., Amundsen, 2022). Or, it may be that individuals viewed harms as focal due to organizational norms (e.g., firefighter storytelling), culture (e.g., trust in clergy), or education on the topic (e.g., STI education on campuses) that brought certain harms to their attention (Bhana & Pillay, 2018; Landa et al., 2019; Pratt et al., 2019). However, this concept is not engaged with directly in these articles, suggesting that there is a need for more foundational work in this area, trying to understand individuals’ perceptions of focal potential harm. The findings provide insight into how focal vulnerability might be complicated by multiple harms. When multiple harms were mentioned, some articles merely provided a (presumably) non-exhaustive list of examples (see, Baier, 1986) while others incorporated an 53 exhaustive list into a model (see, M. L. E. Chan, 2009). Other contexts brought specific vulnerabilities to the forefront (even when multiple vulnerabilities existed). For example, Han and Mosqueda (2020) argued that potential for elder abuse might be more at the forefront of a trustor’s mind because of the additional vulnerability to COVID-19 and the stay-at-home orders (meaning they were shut off from resources and confined to their homes). For institutional trust, financial harm was the only concrete harm identified, but was similarly shaped by context, where individuals relied on coping strategies shaped by the context. Namely, they balanced trust and institutionalized distrust in the wake of past financial crises to ensure appropriate oversight such that negative financial impacts can be avoided (Bachmann & Hanappi-Egger, 2012) or relied on external factors such as laws, culture, and ethics to navigate trust in institutions (Dupont & Karpoff, 2020). Taken together, these results suggest that individual trustors may be understanding vulnerability as an amalgamation of harms. Additionally, focal vulnerability is heavily shaped by context. Additional research on how different contexts bring certain harms to the forefront of one’s mind and the ways individuals cope with uncertainty and risks is necessary. Amorphous Harms Amorphous hams were tied to asymmetrical power within a relationship. In the interpersonal trust articles, vulnerability to specific harms related to relative power, (such as patients who trust doctors or students who trust teachers) but the relational trust theme showed that there was potential for harm even in the inverse (e.g., teachers exploring vulnerability to students, doctors trusting patients) (Bearman & Molloy, 2017; Buchman & Ho, 2014). Importantly, if a betrayal occurred in a relationship that particularly close or one of protection from harm, this could relate to negative emotional reaction (Gobin, 2012; Koehler & Gershoff, 54 2003), especially if the harm is the very thing the trustee was supposed to protect the trustor against (Koehler & Gershoff, 2003). For institutional trust, the power asymmetry was linked to an individual compared to an institution (exemplified by the very title of French, Hannah & Huckle’s (2022) article—" If I Die, They Do Not Care” discussing nurse’s disempowered relationships with healthcare systems). The results on amorphous harms and asymmetrical power lend support to the core tenets of betrayal trauma and institutional betrayal theories which theorize that dependence and trust relate to trauma or emotional harm when betrayed (Freyd, 1996; C. P. Smith & Freyd, 2014) but add important nuance by exploring power. While the concrete harm articles pointed to contexts and identities that contributed to vulnerability, these articles add a dimension by considering vulnerability relative to the power of the trustee. When a trustee has more relative power or a trustor is more dependent on that trustee, these articles suggest there is potential for emotional harm. Future research should explicitly explore trust in asymmetrical relationships of power and explore differences in both vulnerability to and awareness of potential harms. A novel theme arose in the interpersonal trust articles surrounding relational harm, which explicitly looked at different harms tied to the different parties in reciprocal trust relationships. It is telling that three of the four articles under this theme related to the medical field (including psychoanalyst/patients and opioid contracts between doctors and patients), as there is a power imbalance between professionals and patients, especially in light of the COVID-19 pandemic (see, H. Y. Chan, 2021). The trends in the literature suggest that especially when a more powerful party expresses a willingness to be vulnerable in a relationship that can be beneficial to foster better healthcare, teaching, and intimacy in interpersonal relationships. 55 Mechanisms for Harm Finally, an important takeaway from the articles addressed mechanisms for understanding emotional harm. Some form of betrayal as a mechanism arose in both groups, but moral injury only arose in interpersonal trust. Moral injury and betrayal are two different mechanisms that require some teasing apart as they both make important but slightly different contributions to our understanding of harm. Moral injury had less to do with context and more to do with understanding multiple harms that might stem from a trust relationship (Brennan, 2021). For example, whether it be “intellectual streaking” to put one’s mistakes and thoughts up for scrutiny to build trust (Bearman & Molloy, 2017) or a moral injury that stems from a loss of sense of security after a breach (Brennan, 2021), the literature suggests that moral injury is complex and could incorporate multiple harms. However, betrayal was strongly tied to the specific behavior one was trusted to perform (or the specific harm in control of the trustee). The most salient example being a breach of trust when the person allows harm from the very thing they had promised protection (Koehler & Gershoff, 2003). Again, moral injury was explicitly not identified as the mechanism in institutional trust and betrayal. Instead, constraints by an organization that thwart ability to provide care mapped on to distress and betrayal more broadly (Brewer, 2021). Much more research into mechanisms for harm, especially in institutional and multilevel trust is necessary, but the results suggested betrayal and moral distress are key to understanding harm. Multilevel Harm An important takeaway is the growing work identifying the role of multilevel trust and potentially multilevel harm. As Hamm and colleagues (2021) and Wolfensberger (2016) both identified, actors in an institutional setting (e.g., physicians, police) are difficult to disentangle 56 from their context and could relate to multiple types of harm. Similarly, while privacy and security did not identify “multilevel” harm, these concepts arose in both interpersonal and institutional trust and implicated institutional structures in an important way. An institution might be trusted to protect private information (in the case of contact tracing) but research struggled to define what an institution was comprised of, or which actors were in charge of safeguarding the information (Chen et al., 2021). Or, institutional safeguards put in place to protect members might actually be viewed as potential harms themselves—as with marginalized youth’s perceptions about mandatory reporting and harms to their family structures (Sapiro, 2020). Privacy and security also implicated perhaps the most amorphous types of harm, (e.g., racism) where participants were concerned about broad misuse and diffuse impacts that would stem from their trust in and participation with research (Kraft et al., 2018). Research in this area is undeveloped as relates to trust, but privacy violation or personal data misuse seem to align with the complex and growing literature on multilevel harm, especially in interactions on the internet. It is important to note that articles were included even if they hypothesized that trust would create potential for harm but ultimately these hypotheses were not supported (i.e., Kerler & Killough, 2008; Judges 2012). Importantly, Kerler and Killough (2009) concluded that while theory suggested that an accountant’s trust in their client could decrease skepticism and interfere with their ability to properly audit that client, the data did not support that hypothesis. No professional vulnerability stemmed from trust in that context, but professional harm did arise in exploring a higher education context (Kramer, 2012). Perhaps there are important differences between accounting and higher education that can explain these differences in harm from trust (e.g., different legal provisions or cultural norms governing behavior). In Judges and colleagues’ 57 (2017) work, the hypothesis that trustful older individuals would be more inclined to believe others and put themselves at increased victimization risk was not supported. However, they asked participants about past fraud victimization but measured current levels of trust such that prior experiences of fraud might cause decreased current trust (and increased awareness of vulnerability to types of trustee), so potential for fraudulent victimization could still be a potential harm tied to trust despite these findings (Judges et al., 2017). These negative cases bring attention to the complex ways trust and harm relate in these specific contexts and suggest not all harm is equally salient in different types of relationships. Limitations and Gaps of the Knowledge Base A major limitation was that interpersonal trust was the predominant focus of the literature and potential for harm in institutional trust was largely absent from this literature. Of the articles that did grapple with potential for harm from institutional trust, three discussed financial harms that were wider reaching than those identified by interpersonal trust (e.g., the global financial crisis), but much more work must be done to explore explicit harms in this realm. While more articles arose on amorphous harm stemming from institutional trust, the articles primarily applied different iterations of institutional betrayal theory and very few engaged with the mechanisms for harm stemming from institutional trust (for exception, see, Brewer, 2021). This suggests institutional betrayal theory is extremely formative in understanding harm that stems from institutional trust but the research is limited in terms of scope (i.e., focus on sexual assault) (C. P. Smith & Freyd, 2014). While some (e.g., French et al., 2022; C. P. Smith, 2016) considered institutional betrayal in healthcare settings, Wolfensburger (2016) suggested this setting uniquely implicates multilevel trust, a concept that these authors did not engage with. Thus, there are gaps in research qualitatively exploring the differences in 58 specific types of harms that might stem from multilevel, interpersonal, and institutional trust, and theorizing why these might be different. Future research should focus on understanding the extent to which institutional members trust institutions, what they consider as “institutional trustees” and how harms might be distinct for different types of trustors (e.g., comparing victims to nurses). Finally, while concrete and amorphous emerged as meta-themes for both interpersonal and institutional trust, many articles struggled to identify one specific harm that might stem from a trust relationship. Some of the articles began to flag how context shaped salience and type of harm (e.g., Pratt and colleagues (2014) identified that physical harm might be at the forefront of concern to firefighters) but others struggled, perhaps listing multiple (non-exhaustive) lists of harms or engaging with them only superficially (e.g., Baier, 1986; M. L. E. Chan, 2009). Prior research supports that vulnerability might be tied to identity (Kuran et al., 2020), institutional role (Kutsyuruba & Walker, 2016), or context (S. M. Baker, Gentry, & Rittenburg, 2005), and can be cumulative and multidimensional (Misztal, 2012). However, there were gaps in terms of clear boundaries around what would be concrete versus amorphous harms in these articles. This is a specific area that future research can seek to clarify, as concrete harms would have different recommendations for resources for victims (e.g., physical health, financial resources) than amorphous harms (e.g., mental health, emotional support resources). Conclusion Central to understanding harms that are distinct in Institutional Sexual Assault is understanding potential for harm in a trust relationship. This scoping review identified a broad array of literature that engaged with interpersonal, institutional, and multilevel trust and harm. Across research questions, a meta-theme of concrete and amorphous harms helped organize the 59 themes that emerged for articles on interpersonal and institutional trust. Concrete harms were linked to the behavior of the trustee, for example, when a trustee perpetrated fraud, passed along an STI, or sexually abused a trustor. However, the results provided important insight into how identity and context shaped imposed and focal vulnerability for concrete harm. As with concrete harms, part of understanding emotional harms that stem from interpersonal trust includes understanding vulnerability as shaped by context. In these articles, dynamics of relationships, namely, asymmetrical relationships, emerged as important to understanding potential harms. Very few institutional trust articles arose, and they implicated different mechanisms to explain amorphous emotional harm than interpersonal trust. Results improve insight into differences in vulnerability as shaped by identity and context, different types of harm, and how these differ by trustee suggest there are a possible array of harms that a victim is experiencing in the wake of trust betrayal. 60 CHAPTER 3: “I THOUGHT I COULD TRUST YOU, BUT INSTEAD YOU ABUSED ME”: UNDERSTANDING VICTIM ACCOUNTS OF INTERPERSONAL AND MULTILEVEL VULNERABILITY AND TRUSTWORTHINESS Abstract Victims’ accounts of ISA can help provide unique insight into distinct harms associated with different trust relationships tied to an assault. This research applied constant comparison and grounded theory methods to explore 127 victim impact statements given at Dr. Larry Nassar’s sentencing trial. Specifically, the research compared victim descriptions of vulnerability and reasons to accept that vulnerability (trustworthiness) across individuals who did and did not address institutional relationships. Results suggested that when assessing their vulnerabilities, both groups described aspects of their identity, situations, and (to differing extents), institutional structure that shaped their vulnerability to harm when trusting either Nassar or Nassar and the institution(s). Importantly, their intersectional identities contributed to differential vulnerability. There were differences in the two groups, where the multilevel group uniquely stressed how institutional safeguards helped shape their perceptions of vulnerability in that they assumed (wrongly) these would protect them. For trustworthiness, both groups described a hybrid set of interpersonal/institutional trustworthiness factors that influenced their willingness to be vulnerable to Nassar and/or the institutions in which they were members. Implications for research, especially as relate to institutional trustworthiness used in grooming, and policy as relates to institutional space and multilevel solutions to address vulnerability are discussed. Introduction In September of 2016, the Indystar newspaper brought to light shocking allegations of sexual abuse couched as legitimate treatment perpetrated by renowned gymnastics doctor Larry Nassar (Evans, Alesia, & Kwiatkowski, 2016). The subsequent months brought forward 61 hundreds of victims who had experienced similar abuses over decades from Nassar. The corresponding criminal charges and plea agreement allowed all impacted by the abuse to provide victim impact statements during Nassar’s sentencing, which received local, national, and international attention (Crawford & Haneline, 2018). The victim impact statements offered a unique opportunity for victims to share the events leading up to the abuse and the harms they experienced. The victims described multiple reasons to trust Nassar (his status, his kindness, trust by their peers) and they identified multiple institutions as settings of abuse to which they attributed accountability (Abrams & Potts, 2020). These included educational institutions like universities and high schools, medical institutions, gymnastics organizations, including local gymnastics clubs and national organizations like USA Gymnastics. This was one of the most highly publicized events of institutional sexual assault (ISA) (assaults occurring in institutional settings and/or by intuitional actors) that provided victims an opportunity to explore the people and institutions that contributed to abuse and caused harm. Existing research on ISA suggests that when assault occurs in a relationship of trust with another person or with an institution it relates to distinct harms for victims (Freyd, 1996; Smith & Freyd, 2013). When assault occurs in a trust relationship with an offender and victim, it results in worse mental health outcomes than assault perpetrated by a non-caregiver or non-trusted party (Birrell & Freyd, 2013; Freyd, 1996; Smith & Freyd, 2014). Distinct harm also occurs when institutions fail to prevent or supportively respond to abuses through acts of omission or commission through institutional betrayal (Smith & Freyd, 2014). ISA can potentially implicate both interpersonal trust and institutional trust, meaning multilevel trust (trust in a person and an institution) could relate to distinct harms. While trust is central to understanding harm associated 62 with ISA, research has failed to explain what these harms might be and how they might differ based on the type of trust relationship in which abuse occurs. Examining the steps of the trust process that lead to trust can help understand nuances in harms from trust. Trust is commonly defined as a psychological state characterized by the willingness of a party (the victim) to be vulnerable to the actions of another party (the person or institution) (Mayer, Davis, & Schoorman, 1995). Once trust is established, it facilitates the trustor in a belief that efforts to self-protect are not needed, which, in turn, creates the potential for additional (emergent) vulnerability to harm (Mayer et al., 1995). However, understanding potential for harm entails teasing apart the first stage of the trust process, namely, victims’ vulnerability before they decide to trust and the factors that help them accept that vulnerability in deciding to trust (trustworthiness). Vulnerability generally refers to the state of being susceptible to damage or injury (Misztal, 2012) and can be imposed, that is, implicating events and systems outside of a victim’s control (Montgomery, Jordens, & Little, 2008). Only a subset of that vulnerability, called focal vulnerability is most relevant to understand trust. Focal vulnerability contemplates a victim’s subjective experience of vulnerability to the trustee (Hamm & Mollering, 2022). Vulnerability differs by type of trust relationship, where different risks may be associated with trust in a person (e.g., being cheated or betrayed) (Baier, 1986; Chan, 2009; Mayer et al., 1995) than an institution (e.g., financial scandals, damage to institutional identity) (Bachmann, Gillespie, & Priem, 2015; Scott, 2010). Multilevel trust implicates multilevel vulnerabilities (to a person, and perhaps differently, to the institution), but little research has explored those vulnerabilities (for exception, see, Fulmer, 2018). Victims’ intersectional identities here (they were primarily young, female, 63 and patients) could uniquely create opportunities for harm related to ISA, but the types of vulnerability victims experience in entering different trust relationships is under-explored. Trustworthiness refers to attributes of a person or institution that help foster a victim’s willingness to be vulnerable, and this also differs in interpersonal versus multilevel trust. Typically, trustworthiness is assessed by looking at another person’s ability, benevolence, and integrity (Colquitt, Scott, & LePine, 2007; Mayer et al., 1995; Schoorman, Mayer, & Davis, 2007). However, recently research has proposed that macro-level factors help a victim understand multilevel trust, looking to things like an institution’s reputation, safeguards, or culture (Bachmann & Inkpen, 2011; Fulmer, 2018; Tan & Tan, 2000). Reflecting the complexity of multilevel trust, multiple trustees (institutional representatives, coaches, the institution as a whole) and feelings of loyalty to the institution might factor into a victim’s decision to trust (Carvalho & de Oliveira Mota, 2010; A Fulmer, 2018; Kramer, 1999; Tan & Tan, 2000). The current research uses the first stage of the process of trust to understand harm by drawing from experiences of victims of ISA perpetrated by Nassar. This abuse implicated both interpersonal and institutional betrayal (Citrino, 2018; Hauser & Astor, 2018; Martindale, 2020; Tracy, 2018). However, no research has explored how the trust process can provide insight into unique harms associated with ISA for this subset of victims. When considered through the lens of trust, these victim impact statements provided a unique opportunity to discuss vulnerability and trustworthiness tied to interpersonal trust (e.g., trust in Nassar) and multilevel trust (e.g., trust in Nassar and an institution). Using constant comparison and grounded theory methods, I relied on 127 victim impact statements to explore victim perceptions of vulnerability and trustworthiness in interpersonal and multilevel trust. Insights help build theory around the trust 64 process, inform literature on harm associated with ISA, and shape suggestions for better tailored resources for recovery and prevention. Literature Review Institutional Sexual Assault & Trust Decades of research have identified a range of harms associated with sexual assault (Armstrong, Hamilton, & Sweeney, 2006; Dworkin, Menon, Bystrynski, & Allen, 2017; Salter, 1995) but research suggests that when ISA occurs in relationships of trust, this relates to distinct harms for victims (Freyd, 1996; Smith & Freyd, 2013). Assault occurring in a close, trusting, or dependent relationship between an offender and victim (betrayal trauma theory) is associated with blindness to betrayal as a coping mechanism and a number of negative impacts (e.g., mental health, relationship, and substance use issues) (Birrell & Freyd, 2013; Delker & Freyd, 2014; Freyd, 1996; Smith & Freyd, 2014). Harm can also occur when institutions fail to prevent or supportively respond to abuses (institutional betrayal theory)(Smith & Freyd, 2014). Both types of betrayals are perceived by the victim as a violation of the trust relationship they had with the person or institution (Goldsmith, Martin, & Smith, 2014; Smith & Freyd, 2013, 2014).). ISA is particularly complicated because it might implicate multilevel trust, that is, interpersonal trust in a perpetrator and institutional trust. In the VIS given at Nassar’s sentencing trial, victims pointed out that Nassar fostered a trusting or dependent relationship, and in many cases they lamented that they were “blind” to the abuse for many years, key attributes of betrayal trauma theory (Birrell & Freyd, 2013; Rahal & Kozlowski, 2018). Additionally, many victims described behaviors of institutions that exacerbated the impact of the abuse: institutions and institutional actors were viewed as creating an environment where this experience seemed common; they didn’t take proactive steps to 65 prevent this experience; created an environment in which this experience seemed more likely to occur; made it difficult to report the experience; responded inadequately; covered up the experience; punished them in some way for reporting—each an element of institutional betrayal (Banta, 2019; Smith & Freyd, 2014). There is a critical nexus among the institutional setting, interpersonal, and institutional trust relationships in which ISA occurs. Trust is theorized to provide insight into the source of harm in these trust relationships (namely, the violation of trust). However, trust is merely presumed to be present in the existing literature. For the most part, dynamics of trust relationships have not been investigated any further in relationships of dependence at the interpersonal or institutional level (Freyd, 1994; Smith & Freyd, 2014; Taylor et al., 2020). Even in research that identified that trust and power discouraged disclosure for ISA victims (Eiler et al., 2019), trust was not explored in any depth as to how and why it was linked to these distinct experiences and harms. Trust Process The commonly used definition of trust describes a willingness to be vulnerable to another as its core tenet, centering the focus on vulnerability (Mayer et al., 1995, p. 712). A commonly used definition of vulnerability is the state of being susceptible to damage or injury (Misztal, 2012). Vulnerability draws attention to the potential for harm that a trustor is aware of when they enter a trust relationship (Hamm & Möllering, 2022). Once aware of vulnerability, in assessing their willingness to be vulnerable, a victim looks to attributes of the trustee that signal trustworthiness. Thus, when a victim trusts, it is usually because they believe that they have identified good reasons for a belief that their vulnerability is protected and which often facilitates a belief that future self-protective behavior is not needed (Schoorman et al., 2007). For example, a supervisor 66 might trust an employee to work with less supervision once trust is established (Mayer et al., 1995). Emergent vulnerability is theorized to occur at a separate stage of the trust process, but this might differ from imposed and focal vulnerability and bring potential for different harms. Differences by Trust Relationship Vulnerability and trustworthiness differ based on the type of trust relationship, where most literature contrasts interpersonal with institutional trust (Bachmann & Inkpen, 2011; McEvily et al., 2006). However, when ISA implicates institutional trust, it is more difficult to disentangle from interpersonal trust. This is because an individual is necessarily implicated in perpetrating the abuse (which could implicate both interpersonal and institutional trust). Thus, the most accurate comparisons in ISA are between interpersonal trust and multilevel trust (trust in a perpetrator and an institution). Multilevel trust is increasingly being recognized as the way trust operates in many institutional settings (Fulmer, 2018; Fulmer & Dirks, 2018). Multilevel trust draws attention to interpersonal relationships that are necessarily embedded in larger institutional contexts. It similarly focuses both on the trustor’s relationship with other peers in the institution, supervisors or representatives of the institution, the organizational climate, and the cultural values of society (Fulmer, 2018). The existing literature suggests that trust operates slightly differently in interpersonal versus multilevel relationships. Vulnerability Vulnerability in interpersonal trust relationships has first been explored as imposed vulnerability. Imposed vulnerability is shaped by situations outside of a person’s control (e.g., natural disasters) (Montgomery et al., 2008). Increasingly, imposed vulnerability considers identity, where certain groups might feel more vulnerable to harm in certain contexts. For example, vulnerability has been linked to individual characteristics (e.g., age, gender, 67 race/ethnicity, cognitive deficiency), states (e.g., grief, instability) or external conditions (e.g., discrimination, social status (Baker et al., 2005; Kuran et al., 2020). An intersectional perspective has been particularly useful for understanding how multiple intersecting identities could relate to multiple intersecting vulnerabilities (Crenshaw, 1991; Kuran et al., 2020; Darcy et al., under review). Along those lines, ISA literature suggests that certain identities (e.g., women of color, queer/trans individual) face increased vulnerability to sexual assault (Cantor et al., 2019; Department of Defense, 2019; McCauley, Campbell, Buchanan, & Moylan, 2019; Rantala & Rexroat, 2018). An individual’s vulnerability in a relationship can be explored by looking at the interdependence in relationships, namely, as the extent to which they depend upon, are willing to rely upon, give control to a trustor (PytlikZillig & Kimbrough, 2016). In the course of their lives, a person can be vulnerable to multiple hazards, some of which they might be more or less aware of. Some research on trust focuses on focal vulnerability, which refers to potential harms in control of the trustee that one is aware of in entering the relationship (Hamm & Mollering, 2022). Focal vulnerability also implicates identity and relative social status, and likely incorporates some imposed vulnerability, but focuses solely on the vulnerability one contemplates in deciding whether to enter a trust relationship (Hamm & Mollering, 2022). The details of a victim’s imposed vulnerability (as related to identity, situation, or other factors outside of their control) and which types of vulnerability they are aware of in deciding to trust a perpetrator has not been explored. Multilevel trust adds an institutional dimension to understanding vulnerability. By choosing to enter a relationship with an institution, one may also accept vulnerability to the actions of that institution. Members may rely on institutions to protect them from harm in 68 exchange for complying with their rules and norms (Scott, 2010). Increasingly, people may voluntarily seek out membership in institutions for purposes of self-improvement or finding community, and as a result of this voluntary acceptance, they accept a certain level of control over their actions, creating vulnerability to actions of the institution (Scott, 2010). Some ISA research suggests that victims’ vulnerability to assault can be controlled by an institution. Sexual geography is a concept introduced in Hirsch and Kahn’s (2020) work on understanding campus sexual assault. The authors explored how space controlled by a university has social power and certain behaviors may be more likely to occur in certain settings. For example, the physical layout of campus dormitory rooms is such that there is often no space to sit when visiting a room except on a bed, or, in going “off campus” this may mean not having access to transportation and feeling compelled to spend the night somewhere. As the authors note, “sexual geography is part of what produces vulnerability” to assault and is shaped by institutional decisions and designs (p. 144). Exploring physical space and sexual geography can provide insight into vulnerability to harm related to trust by considering a novel aspect of institutional context. To apply this concept to ISA, certain aspects of physical space for treatment of patient/victims (treatment rooms, spaces in gymnastics training facilities, hotel rooms) could shape victim vulnerability. Physical space and sexual geographies have not been explored to understand trustworthiness and vulnerability. Trustworthiness Trustworthiness refers to the signals that a victim might look to in determining willingness to be vulnerable to another party. Interpersonal trustworthiness has been conceptualized in this literature as having many dimensions, but three distinct dimensions continually arise: ability (perceived technical competence in the domain of interest); benevolence 69 (that the perpetrator cares about you or group of people like you); and integrity (they follow a set of internalized values trustor finds acceptable) (Mayer et al., 1995). These perceptions in conjunction with a victim’s vulnerability are part of what informs their willingness to be vulnerable. Trust and trustworthiness are strongly related in empirical research on trust, such that trustworthiness is often a strong signal of trust (Colquitt et al., 2007; Mayer et al., 1995). An individual trustee’s behavior can intentionally signal trustworthiness. For example, leaders can purposefully use self-presentation behaviors, language, and physical appearance to portray images of trustworthiness (Elsbach, 2004). The way this may manifest in ISA is through grooming behaviors. These are intentional acts by a perpetrator that might be used to foster trust of victims, which helps facilitate assault and make it difficult for victims to report (Craven, Brown, & Gilchrist, 2006; McAlinden, 2006). Grooming, then, is the process by which a perpetrator manipulates a victim using trust to create a situation where they can be more readily assaulted and will be less likely to disclose (Salter, 1995). It may be that a professional perpetrator uses their job to target and groom victims (Sullivan & Beech, 2002), perhaps by purposefully projecting attributes of trustworthiness (ability, benevolence and integrity). However, trustworthiness had not been explicitly examined by grooming literature. As with vulnerability, there are institutional analogues of trustworthiness. When a person assesses the trustworthiness of an institution, they look to structural factors (Bachmann & Inkpen, 2011). Part of the way institutions contribute to a trustor’s willingness to be vulnerable is by having certain structures reducing the risk of misplaced trust (Bachmann & Inkpen, 2011). Bachman and Inkpen’s (2011) work on institutional trust suggests there are four different mechanisms by which institutions act to foster development of trust: (1) legal provisions; (2) reputation (3) certification of exchange partners and (4) community norms, structures, and 70 procedures. The extent to which an institution or institutional actor signals legal provisions, reputation, certification of exchange partners or community norms has been minimally explored (for exception, see, Bachmann & Inkpen, 2011; J. D. Harris et al., 2013; Poppo, 2013). Similarly, the extent to which an institution might signal trustworthiness in a way that mirrors grooming by an individual is unclear.6 A multilevel lens provides additional insight into trustworthiness of a person in an institutional role (e.g., supervisor, team doctor). In assessing trustworthiness of a person in an institutional role, a trustor relies on proxies for direct knowledge and interaction. Kramer (1999) explains that proxies can be in the form of third parties (e.g., other organization members) or in the form of role-based trust, looking to the role the trustee plays in an organization, which indicates their capabilities and signals trustworthiness (Kramer, 1999). This type of trust is innately related to the system of expertise in which one was trained (e.g., medicine) and which is expected to shape role-appropriate behavior (Kramer, 1999). In this case, it is not the person in the role that is necessarily trusted, but the system of expertise that regulates and maintains role- appropriate behavior of that person (Barber, 1983, Dawes, 1994). For example “we trust engineers because we trust engineering and believe that engineers are trained to apply valid principles of engineering, moreover, we have evidence every day that these principles are valid when we observe airplanes flying” (Dawes, 1994, p. 24). This could apply to a physician in the same way, especially a physician who was licensed and employed by numerous institutions. Different trustees in a multilevel institution may be trusted differently (e.g., trust in a supervisor but not in an organization), and research suggests trustors may rely on different 6 While some have referred to institutional grooming itself (McAlinden, 2006), it does not refer to an institution grooming a victim, but refers to a perpetrator grooming an institution to continue abuse. 71 trustworthiness signals for different trustees (Tan & Tan, 2000). Namely, trust in a supervisor was associated with ability, benevolence, and integrity, while trust in an organization was associated with “global variables” such as perceived organizational support and perceptions of justice (Tan & Tan, 2000). Research that has applied a multilevel lens to specifically doctor- patient relationships found that meso-level factors (i.e., trust in medical profession, benevolence, competence and integrity) influenced trustworthiness perceptions of a doctor, but were embedded in system trust (i.e., trust in health care system, trust in other people, trust in technology and procedures) (Krot & Rudawska, 2016). A final factor that should shape trustworthiness perceptions within multilevel trust is the level of loyalty one feels to the institution. In higher education, trust and loyalty differ depending on the type of trustee. Carvalho and Mota (2010) found that students’ trust in personnel (faculty and staff) and management (as reflected in administrative policy) differentially influenced perceptions of the value of the institution, which then influenced their loyalty to the institution. As perceived value of an institution increases, so does trust, and therefore, so does loyalty (Carvalho & de Oliveira Mota, 2010a; Sampaio, Perin, Simões, & Kleinowski, 2012). Thus, multilevel trust could incorporate both interpersonal and institutional signals of trustworthiness. Because different trustworthiness perceptions exist for the complex and embedded trustees in institutional settings, it is likely that different harm (or different mechanisms for producing harm) stem from trust relationships. Current study In 2016, the media uncovered widespread abuse by Nassar which impacted hundreds of victims abused under his care as university faculty, physician, and gymnastics trainer, resulting in his conviction for criminal sexual conduct (Evans et al., 2016). Existing ISA research 72 theorizes that abuse in relationships of trust relate to distinct harms, however, research has struggled to explain how trust relates to harm and how harms may differ by relationship (Freyd, 1994; Smith & Freyd, 2014; Taylor et al., 2020). Despite, the nexus between institutional setting, interpersonal, and institutional actors trust has not been applied as a lens to understand harms stemming from ISA. Examining the trust process as differs by type of relationship can provide insights into the differences in vulnerability and perceptions of trustworthiness that cause a victim to accept willingness to be vulnerable to a person or an institution (Bachmann & Inkpen, 2011; Mayer et al., 1995). The goal of this chapter was to explore how victims spoke to the question of how trust was developed, focusing on their vulnerability and the signals of trustworthiness for both interpersonal and multilevel (interpersonal and institutional) trust. I applied constant comparison and grounded theory methods (Auerbach & Silverstein, 2003; Glaser & Strauss, 1967) to understand differences in harm experienced based on type of trust relationship described by victims. To do so, victims were split into groups of those who discussed only interpersonal trust of the perpetrator and those who described both interpersonal and institutional trust (multilevel trust). I examined 127 victim impact statements given at a criminal sentencing of Nassar to explore the following questions: (1) How do victims in the interpersonal trust group (versus multilevel trust group) understand trustworthiness? (a) in what ways does trustworthiness differ for those who described trusting individuals (versus multilevel trust)? (b) in what ways is trustworthiness similar for those who described trusting individuals (versus multilevel trust)? 73 Method Data Data for this investigation come from 127 victim impact statements made in Michigan’s Ingham County Court by victims of sexual assault by Nassar. The VIS were given as part of the sentencing proceedings from Nassar’s criminal plea to multiple charges of criminal sexual conduct under the Michigan Compiled Laws (Lutz, 2018). In addition to victims, friends, family, coaches, and others who were impacted by the abuse provided victim impact statements. While a total of 156 people provided victim impact statements in Ingham County Court over the course of four days the final sample was limited to victim impact statements that were either read or written by victims and for whom transcripts were available7 (N=143). Since the current study is primarily focused on harms that stemmed from trust placed in individuals and/or institutions, only those victims who mentioned trust were included in the sample. Ultimately, the study includes individuals who either (1) explicitly used some form of the word “trust” (N=120) or (2) who described key elements of trust (e.g., benevolence, integrity, vulnerability, reputation) (N=7).8 This resulted a total of 127 statements by 127 victims. Table 4 presents demographic information about the sample. The sample included only women or girls, with 118 adults and 9 minors. Minors were required to have parental consent to give impact statements and so their ages were captured in the transcripts. Racial identities and age of abuse onset were not uniformly provided, but participants described abuse that began in childhood, adolescence, and adulthood. Many experienced the sexual assault during medical examinations for injuries (sports-related or otherwise) and only one participant was not a patient 7 Five statements made over video were not transcribed in the legal record. 8 These indicators of trust were drawn from the theoretical frame of the dissertation. 74 of Nassar. Details about the athletic experience of the participant were often mentioned as the initial reason for the first appointment, and victims included gymnasts, swimmers, dancers, and a range of other collegiate/non-collegiate athletics. While there were a number of other abuses reported, most of the participants reported that Nassar improperly applied a medical treatment, myofascial release, to perpetrate digital vaginal penetration on his patients without consent or appropriate protective measures (Rahal & Kozlowski, 2018). Abuse occurred in multiple settings: offices on Michigan State University’s campus, high schools, other universities, in local gymnastics gyms, in national gymnastics training camps, in hotel rooms during the Olympics and more. Study Design The study relies on victim impact statements, as they are a rich narrative source which allow a search for meanings “to illuminate individuals’ interpretations in a social, cultural, and personal context” (Orbuch, 1997, p. 461). Detailed analysis of individuals’ accounts has been used to allow scholars to arrive at meanings based on the ways people organize views of themselves, others, and their social worlds (Brunson & Wade, 2019; Orbuch, 1997). Victim impact statements are also one of the few spaces in the legal system where victims may provide evidence that is not allowed under evidentiary rules, allowing them to cover a broad array of experienced harms. While individuals were not explicitly asked about trust, trust and trustworthiness literature were used as a sensitizing concept (Bowen, 2006). Sensitizing concepts can be useful to provide a general sense of reference and guidance in grounded theory analysis (Blumer, 1954; Bowen, 2006). This study investigates how victims understand trustworthiness and vulnerability, and how these patterns vary by group (interpersonal vs. multilevel). 75 To make comparisons across different groups, I separated the VIS into two separate groups: those victims who endorsed trust and harm associated solely with an individual trustee (e.g., Nassar) (N=51) and participants who discussed trust and harm associated with both an individual trustee (e.g., Nassar) and an institutional trustee (e.g., an organization) (N=76). To identify these groups, I read through the VIS for each victim and sorted those who only mentioned trusting an individual into the “interpersonal” trust group. Victims in this group may have mentioned institutional settings (e.g., offices at MSU) but if they did not engage with an institution beyond mentioning it, that was not deemed sufficient to indicate multilevel trust. A codebook captured parameters and examples for each group. Participants were sorted into the “multilevel group” if they described trust in an individual (Nassar) and some level of trust, reliance, blame, or harm ascribed to an institution. Ostrom (2005)’s broad definition of institutions as prescriptions that are used by humans to organize social interactions was applied to identify institutions, which meant that references to specific organizations (e.g., MSU, Twistars, a local gymnastics gym) as well as references to broader institutions (.e.g., “gymnastics community” and “the system”) were included. 9 Data Analysis The victim impact statements were analyzed inductively using constant comparative and grounded theory methods to compare how those in the interpersonal group differed from the multilevel group in relation to trustworthiness and vulnerability perceptions (Glaser & Strauss, 1967). This type of qualitative research relies on subjective experiences to inductively develop a pattern of meaning (Cresswell, 2003). Essentially, grounded theory methods utilize systematic 9 Because there is overlap even in how trust scholars think about institutional trust as collectives, groups, organizations, and systems, (e.g., McEvily, Weber, Bicchieri, & Ho, 2006)), identifying trust in institutions was purposefully over-inclusive. 76 inductive guidelines for collecting and analyzing data to build middle range theoretical frameworks that explain the collected data (Charmaz, 2000). While traditional grounded theory methods require simultaneous data collection and analysis, there has increasingly been a recognition of the utility of using grounded theory techniques to analyze rich secondary sources, which are relied upon here (Glaser, 1963). Coding Several inductive analytic techniques were used to strengthen the internal validity of the analysis. This included the use of grounded theory methods that entailed a constant comparative approach and the search for deviant cases. The use of a constant comparison generates theoretical properties of the category and enhances reliability and validity by ensuring the researcher systematically checks the data and monitors analyses and interpretation (Glaser, 1965; O’Neal & Hayes, 2020). To apply the process of constant comparison, I adapted Auerbach & Silverstein’s (2003) step-by-step approach to data analysis. This process culled the victim impact statements into relevant text to analyze and entailed a three-level categorization system analyzing the content (Table 5). First, I identified low-level text-based categories, then developed middle- level themes, and, finally, developed higher-level theoretical constructs (Auerbach & Silverstein, 2003). Each stage of coding and analysis was accompanied by extensive memoranda writing. The first level of analysis included reading the content of each victim impact statement line-by-line using the process of open coding. Similar words or phrases that depicted the same concept were extracted and labeled as repeating ideas. Repeating ideas are those expressed in relevant text by two or more individuals (Auerbach & Silverstein, 2003). Separate files were created for the interpersonal and multilevel groups, and they were inspected separately for patterns. 77 The next step entailed organizing the repeating ideas into larger groups (themes). I made systematic comparisons across themes related to trustworthiness and vulnerability between groups to identify similarities and differences in participant experiences. Finally, from the themes, overarching theoretical constructs were identified to encompass all the themes. This method allowed me to move from a lower to more abstract level of understanding about perceptions of trustworthiness (Auerbach & Silverstein, 2003). In the analysis, care was made to ensure that the concepts provided typified the most common patterns in respondents’ accounts. However, effort was made to search for and expound on deviant cases. Where present, I made notes of any subtle patterns in the analyses that follow. The percentage of individuals in each group that endorsed a theme are presented in the results, while the number of individuals within a group who endorsed a repeating idea (code) are presented in the results. A full table of theoretical constructs, themes, and codes as compared by group are provided for vulnerability and trustworthiness (see, Table 6, Table 7). Although not generalizable, the study provides further insight into how participants experienced harm stemming from trust. Results Interpersonal Trust Group Trust theory suggests that a victim is willing to accept their vulnerability based on perceptions of trustworthiness, so, before assessing trustworthiness, we must first explore the victims’ perceptions of their own vulnerability. Almost half of the victims in this group (47%) described factors that influenced feelings of vulnerability, often considering their vulnerability in relation to Nassar (e.g., relative positions, identities, power). Certain attributes of identities contributed to vulnerability. These included age, innocence, and their status as patients. This was 78 especially relevant in comparison to Nassar’s more powerful status (older, experienced, physician). For example, V19510 described a vulnerability as related to her relatively disempowered status, namely, being young, innocent, and naïve (e.g., “my young, innocent self”). A few (N=3) explicitly described his position of power in comparison to their vulnerability. As KS said: I thought I had no right to speak up. I thought, what is he doing? Why isn't he wearing gloves? This is disgusting. Why is this happening? But he was the doctor and I was the child so he knew what he was doing and he knew what was best (emphasis added). These identity-based vulnerability factors align with existing literature on imposed vulnerability but also show how vulnerability is relational in helping understand their own potential for harm (Baker et al., 2005; Kuran et al., 2020; Wildavsky, 1988). A novel aspect of vulnerability relating to their different situations was salient to some victims. This may have had to do with their life circumstances (e.g., a parent’s illness; family car crash). For others, it had to do with lack of experience: I grew up extremely sheltered. I had never kissed a boy at that point. I had never seen a gynecologist. I hadn't even taken sex education at school because my parents didn't feel that teaching of safe sex was appropriate (AG). Or, because many of the victims in this group described coming to Nassar for healing to return to athletics, they expressed vulnerability in their desperation to get better. As JC described: “I was in a state of great desperation when I paid my visit to Doctor Nassar.” Nassar seemed like their last hope for healing. This was explained by MH: 10 Victims are identified using their initials, or, if no name was given, by a shortened version of the name used by the court (e.g., Victim 195 as V195). 79 He was the only one I was wildly assured who could really help me, help me further than any other doctor had helped me so far. He was just so talented and brilliant, they said. Indeed, he did seem impressive… Larry made me believe that this was my only hope. That he was my only hope. Situational vulnerability may have contributed to them feeling motivated to accept that vulnerability and place trust in Nassar in light of these multiple intersecting vulnerabilities (Weber, Malhotra, & Murnighan, 2004). Interestingly, even in this solely interpersonal trust group, codes related to vulnerability implicated the institutional context and safety. The theme of institutional and non-institutional safeguards was endorsed by one-third of the individuals in this group (33%). This theme broadly referred to the expectation that both institutional and non-institutional safeguards shaped understanding of vulnerability. For example, they referenced how Nassar’s general position of trust and power was related to his institutional position(s). As KJ explained, Nassar’s relative authority and power compared to her vulnerability (innocence) within the doctor/patient relationship was highly relevant: “You took advantage of your authoritative powers as my doctor and, as a result, took my innocence instead of healing me.” As a result, they did not feel they could question him. As MA asked, “Who am I to question the best gymnastics doctor in the world? Who am I to question his medical treatments?” Finally, victims in the interpersonal group described how physical space contributed to vulnerability and informed trustworthiness perceptions. For some (N=5) they were further disempowered by the fact their parents were in the room: 80 When he abused me, my mother was in the room, and even though I know now she had no knowledge of the assault, at the time I felt even less power to speak up because I assumed if something were wrong, she would do something about it (HW). For others, institutional reputations were used manipulatively to create the illusion of a safe space: “You hid for years behind Olympic rings and a Spartan head, but there is now no where left for you to hide, Larry (KP).” Physical space was even described as something that KJ relied on to maintain trust despite discomfort: “I was uncomfortable at the appointments without my father but maintained trust in you as the framed images of the patients on your wall told an incredible story of a great doctor who could heal anyone.” Institutional context, including Nassar’s role and presence, as well as physical attributes (a parent in the room, photos of other patients on the walls, a symbol) implicitly helped shape perceptions of vulnerability even for those who endorsed only interpersonal trust. Multilevel Trust Group In the multilevel group, even more (65%) victims described vulnerability related to their status or identity (Table 6). Like the interpersonal group, this related to their age and innocence. As V127 stated: “I was young, innocent, and terribly confused.” They directly linked this vulnerability to the way they viewed Nassar: I think the problem all this time was that I still saw Larry with a child's eyes, a child who looked at him with love and admiration, who was grateful to have his attention and affection. I was blind to what happened 81 because I trusted him implicitly like a child, but now I can see the truth. (CB) However, AH explained that vulnerability to abuse was not part of her focal vulnerability (the vulnerability she contemplated when entering the relationship), noting “I had never heard of a doctor purposely hurting someone.” Similarly, SR noted “I cannot blame myself for trusting my physician to do his job.” This group was unique in engaging directly with focal vulnerability and discussing how harm that stemmed from abuse was explicitly not something they considered when entering the trust relationship, providing insight into this concept. Some in the multilevel group also endorsed vulnerability related to their life circumstances (e.g., physical pain due to an injury or having parents who were sick). As BR reflected, “At the time my father was terminally ill with cancer so my parents were unable to attend this appointment. Mr. Nassar took advantage of my circumstances and of me.” A unique code that related to situational vulnerability in this group had to do with ignorance or lack of experience with doctors. As RM summarized: I had never seen a gynecologist. I never had a pap smear, and at the time I thought this is what it meant to be a woman going to the doctors, awkward and embarrassing and uncomfortable but just part of the deal. This aligns with descriptions of elite gymnastics as isolating, which contributes to victims’ lack of knowledge (even about sexual abuse) (Pesta, 2019). Victims described how institutional socialization conditioned them to respect authority (either through the culture of gymnastics or influence of specific coaches), thus creating their 82 vulnerability. For example, competing in the Olympics and representing the US Gymnastics team, AR said: Imagine how all of us feel. Imagine how it feels to be an innocent teenager in a foreign country hearing a knock on the door and it's you. I don't want you to be there but I don't have a choice. Treatments with you were mandatory. You took advantage of that. Or, as BL rationalized when she was originally uncomfortable with the abusive behavior: “This is the gymnastics world, though, things aren't by the book.” However, compared to the interpersonal group, only a few (N=3) in the multilevel group described Nassar as their last hope for healing. Their vulnerability was perhaps not as linked to desperation but tied to the institutional culture, socialization to expect mandatory treatments, and conditioning to respect authority of their coaches and doctors. Their experiences suggest that identities, situations, and institutional socialization intersected to shape vulnerability. Unique to the multilevel group, almost half (45%) of the victims endorsed the theme of institutional safeguards. Within this theme, they described expecting some sort of institutional- level protection which lowered their perceived risk of harm. For example, they referenced policies that are intended to prevent abuse (e.g., having a nurse in the room; mandatory reporting policies). However, these safeguards failed the victims as no one was implementing them. "If anyone -- if one or any of the stipulations had been enforced by anyone overseeing Nassar, this would have never happened (VD).” As CS questioned Why did MSU continue to employ Larry? He wore MSU affiliated shirts. He had an office in MSU Sports Medicine. He had business cards with MSU Sports 83 Medicine letterhead. He was your employee, and MSU should be responsible for his actions as the employer; under their roof, in their halls, on their campus. Instead, Nassar was able to still rely on the institutional reputation without any of the safeguards implemented to protect victims from abuse. Building on this notion, an important difference emerged in the multilevel group where some individuals (N=15) described trusting Nassar due to the culture fostered in gymnastics. The context is essential to understanding vulnerability. As JA explained: It's hard for an outsider to understand the world of elite gymnastics and to understand how a man like Larry could gain the trust of so many young girls and sexually abuse them for so many years. For a young girl away from her home being worked into exhaustion by screaming coaches, a kindly doctor offering relief from pain and a little sympathy was easy to like. I was raised in the culture of gymnastics where we were taught your voice doesn't matter. You follow instructions and never complain, especially about treatment. Victims described extreme reliance and obedience to gymnastics coaches as central to gymnastics culture. For example, CW concluded “coaches are trusted and obeyed without question” and described how cultural aspects of elite gymnastics, such as “obedience, unimaginable pain, and silent suffering, were expertly manipulated by Larry Nassar to identify, abuse, and control his victims, not once, but systematically over their lifetimes in the sport.” The victims described that they may have been expected to comply with these cultural norms in exchange for protection, which ultimately failed. “Thanks to USA Gymnastics there was not a soul there who may have protected me from you (JH).” For some, Nassar’s relationship to 84 prominent Gymnastics coach John Geddart emerged as compounding their vulnerability (N=6). Nassar emerged as trustworthy and safe in comparison to Geddart: You were the good cop to coach John Geddert's bad cop. When he broke us mentally and physically, depriving us of water on a hot summer's day in the un-air conditioned gym or pushing us to practice on broken bones, you were the one who stepped in. You defended us. You stood up to him on our behalf. You protected our bodies from further pain. You literally and metaphorically put us back together (V125). This culture and institutional actors shaped vulnerability of victims in the multilevel group in unique ways from the interpersonal group. Also unique to this group, victims had suggestions for what institutions could do in the future to foster trust by bolstering (or complying with) institutional safeguards. Suggestions came in the form of legal provisions, for example I then attended the hearing by the Senate Judiciary Committee which led to the introduction of legislation to require Olympic governing bodies, including USAG, to immediately report sexual abuse allegations to local or federal law enforcement agencies. I was shocked to learn that this law did not already exist (ML). Or, they suggested external investigations (e.g., “We need an independent investigation of exactly what happened, what went wrong, and how it can be avoided for the future. Only then can we know what changes are needed. Only then can we believe such changes are real” (AR)). Further, changes in culture, norms, and policies were suggested: It upsets me also that Michigan State did not have a nurse present in the room with a female patient. If there was a question of his misbehavior and he was put 85 on probation, why didn't he have a nurse present? At most female checkups doctors ask if a woman would like to have a nurse present. I find it uncalled for that a university that touts itself as a prominent health care provider would not have such a protocol in place to protect its patients and its doctors alike. (V136) These suggestions for change align with the indicators of institutional trust identified by Bachmann and Inkpen’s (2011) framework for fostering trustworthiness in institutions, which suggests institutions can foster trust through: corporate reputation, legal provisions, certification by exchange partners, and changes to culture. Victims suggested institutions could rely on these mechanisms to signal trustworthiness, or in other words, garner trust from victims that institutions could actually protect them from harm in the future. Finally, like the interpersonal group, victims in the multilevel group described the importance of the physical space in contributing to vulnerability. Not only were medical rooms associated with safety and healing (“To sexually abuse me, a little girl, right there in his office in the safest and warmest of places with such an overlying sense of healing and recovery MH”) but they were adorned with evidence of his ability to help others, bolstering trustworthiness. As CW explicitly noted: One small yet significant detail that strengthened my trust in Mr. Nassar's intentions were the actual treatment rooms designated to him at the MSU Sports Medicine facility. His particular treatment rooms were adorned with photos of Olympic gymnasts and other prominent athletes who he claimed to medically treat. I would walk into his room thinking that if he treated all these athletes, gymnasts, that I was undoubtedly in the right hands. 86 The results from both the interpersonal and multilevel group add to the growing research identifying how sexual geographies relate to vulnerability to assault and harm (Hirsch & Khan, 2020). Interpersonal Trust Group As to why victims would be willing to accept the vulnerabilities discussed above, the interpersonal group relied on benevolence as a signal of trustworthiness. Signals of benevolence were prevalent in this group (65%), however, in hindsight, these “benevolent” behaviors were identified by victims as grooming. As 139SB reflected, “He inspired me, helped me, but ultimately, and unfortunately, used me. He groomed me while gaining mine and my family's trust.” Benevolence emerged in a few distinct ways. First, they described how Nassar maneuvered himself to play multiple important roles in their lives, which helped convince them he cared for them and was doing what was best for them (benevolence) (Mayer et al., 1995). As V138 described: “From the time I was eight through 33 years old, you were in my life; a doctor, a friend, a second father, a person I confided in to make everything all right.” Playing multiple roles built evidence that he cared and wanted what was best for them. In addition, certain behaviors were used manipulatively. He gave gifts, treated them after hours, made jokes, or portrayed confidence and calmness as indicators of benevolence. For example, V48 lamented: Larry Nassar manipulated me psychologically. He made me deny how I felt when I was screaming inside to stop what was happening because he acted so calm like nothing ever happened. He made me trust him and not second-guess his medical procedure (ID). 87 He also did them “favors.” AG explained: Later he was willing to stay late after hours. He began fitting me in at the end of the day after my gymnastics practices, even after all the rest of the staff at the medical center had left. By doing this, I didn't even have to sign in or pay a co- pay. He was doing us a favor and not charging us. That's what he told me…. Larry gave us gifts, backpacks, shirts, a water bottle. Nassar also capitalized on their ignorance of medicine or medical terms and used “benevolent” acts to convince them to trust. Elaborating, AG said: He dumbed down his medical terms and was really goofy with us to try to get us to trust him, and we did, and the one thing that always sticks in my mind is at the end of every appointment he would give me this really long hug and assure me that everything was going to be okay and we were on the right track, and I always felt sick about that hug. Additionally, contributing to perceptions of benevolence was the fact that others in their peer group had also experienced the abuse. Only a few (N = 3) in the interpersonal group described how Nassar’s abusive “treatment” emerged as widely-accepted reality among peers, but where present, this might have helped assuage any doubts about his benevolence when abuse was occurring. These signals of benevolence built from the vulnerabilities victims described and convinced them to trust him. Slightly more than half (51%) endorsed codes around Nassar’s ability. Interestingly, assessments of ability were tied to institutional context by victims in the interpersonal group. These reflected his medical training as well as referrals by other professionals (coaches and other doctors). As AE explained, “[Y]ou were the doctor. You were also someone that many people 88 told me could fix my medical problems.” As with vulnerability, some described how institutions socialized them to trust doctors and their medical ability. As SR explained, “society teaches us that we should be able to trust doctors, and I thought I could trust you, but instead you abused me.” As she reflected, she was relying on conditioned assessment of his ability, but in the end, “you were not trustworthy.” Some pointed to trusted institutional actors referring them to Nassar, which bolstered perceptions of his medical ability through the implicit stamp of approval. “Prior to my visit I had been recommended by gymnastic coaches. I heard words such as, you'll love him. He's a miracle worker. He can fix anyone or anything” (JC). Related to ability in institutional context, the theme of reputational trust was endorsed by 45% of the victims in this group. His reputation signaled trustworthiness due to first-hand stories of him treating their friends, peers, or even idols. This may have implicated numerous people in their closest circles: “I trusted you. My teammates trusted you. My parents trusted you.” (V195). However, even in the interpersonal group, being known as “the gymnastics doctor” was a factor that promoted trust. This tied his reputation to institutions (gymnastics gym, USAG, USOC). As KJ explained, his reputation and the fact that others trusted him was a factor in assessing both her vulnerability and his trustworthiness: “I remember telling myself that you were safe and I should trust you with my body as you were a doctor for athletes across the country that I looked up to.” Other local institutions also bolstered his reputation and signaled trustworthiness, including colleges (“since Mr. Nassar was so close to CMU [Central Michigan University]” (SU)), high schools (“he was at the Holt High School as their trainer” (AG)), and gymnastics gyms (“every Monday he would come to the gym” (BH)). An important negative case that emerged was in the case of V48, who clarified: 89 Honorable judge, I did not choose to have him as my doctor. I was not star struck by the magazine covers with him in them framed in medical exam rooms. I didn't know who he was. I was a no name high school athlete that got injured during practice and was sent to see him due to a referral. Instead, she relied on referrals of others to his care. The interpersonal group only passively relied on institutional reputations or associations (i.e., they did not rise to the level of institutional trust), but these institutional factors were still relevant to some in assessments of ability. Multilevel Group Signals of benevolence in the multilevel trust group were similarly perceived as grooming. Although fewer victims in the multilevel group endorsed this theme, still over half of victims (54%) described benevolence. BP discussed how Nassar’s roles represented his morals, fostering trust: “It is hard to even know where to start. I can start with trust. I trusted you, Larry. I trusted you as an adult figure, a mentor, and someone who had good morals, especially someone coming from a medical background.” As a result, he seemed to be “the most trustworthy doctor I ever could have imagined” (MH). Again, Nassar’s actions used to signal benevolence were flagged by this group. Specifically, his behavior considering his world-renowned reputation made the individual connections more meaningful. As LB explained “I felt just as important as the Olympians you treated.” In hindsight, certain actions seemed designed to dissipate any doubts and convinced them to be vulnerable to his “treatment.” Similarly, as MP elaborated, actions like hugs, joking, and more contributed to his perceived benevolence: Larry was a craftsman of manipulation using his power and status to control and take advantage of children who he groomed to essentially worship his needs. He 90 hugged me when I walked into the room, made jokes with me, asked me questions about my personal life, about my boyfriends. He established a relationship of trust with me. This went on for over six years of regular appointments. Finally, like the interpersonal group, they relied on their networks when assessing benevolence, but the experiences of peers were a powerful factor in the multilevel group, where more victims endorsed this code than in the interpersonal group (N = 11 versus N = 3; or 14% versus 6%). This meant that even before the abuse occurred it was normalized as benevolent behavior, namely, essential treatment to help them heal. As CH said, before my first appointment I had talked to other gymnasts who had been treated by Nassar. They told me what his treatments included, but they, too, were young, innocent, and naive. We all thought it was okay. We all trusted him. As BH explained “I didn't know what was happening was wrong at the time because it was happening to all the girls who saw him.” They assumed he was doing what was best for them because others suggested this was the case based on their own abuse experiences. While reputation and ability emerged as separate themes in the interpersonal group, the multilevel group overtly discussed how his reputation implicated institutional contexts (68%). For example, he was intertwined with institutions as part of his identity; specifically mentioned were gymnastics gyms and the US Olympics. As above, “he was the gymnastics doctor, physician to the Olympians.” (V10) Referrals operated similarly to the interpersonal group in signaling ability, but these came from a broader array of institutional actors. For example, coaches or other physicians who seemed to trust his treatment bolstered his trustworthiness. As explained by V127, referrals acted in conjunction with his reputation: “My coach sent me to see you. This was my first 91 encounter with you. I had heard so many great things about this doctor and how you have helped so many gymnasts.” Again, the fact that peers or others close to them trusted Nassar contributed to his ability and reputation and fostered trustworthiness. As V125 summarized, “You earned our trust. You earned my trust. But, more importantly, as I was just a little girl, you earned my mother's trust.” His behaviors align with research on grooming, which is a process by which a perpetrator manipulates a victim using trust to make them easier to assault (Craven et al., 2006; McAlinden, 2006). However, by additionally fostering the trust of so many others (many of their parents also trusted him) it also aligns with research on familial grooming (McAlinden, 2006). Finally, many victims in this group (68%) uniquely discussed institutional loyalty, where trust and loyalty to institutions both signaled his ability and fostered trust in Nassar. As V11 explicitly stated: My family bleeds green. With most of them, including extended family, having worked or gone to MSU, I was confident that I would be getting the best care that I could. After all, I believed in MSU and I trusted MSU would only have the best doctors working for them and the safest conditions. In fact, it was his personal reputation in addition to institutional reputation that signaled trustworthiness. I thoroughly believed that Doctor Nassar, with his reputation as a world-renowned sports physician, coupled with Michigan State University's reputation at the time for top care of osteopathic care, I was headed in a positive direction to receive the best care possible (V136). 92 Because institutional safeguards such as culture, actors, and safety contributed to vulnerability, it makes sense that institutional loyalty, the experiences of peers, and associations with specific institutions shaped victims’ perceptions of trustworthiness of both Nassar and the institutions. Discussion Vulnerability is “a ‘state’ of being prone to harm, injury, or damage” which can then be exploited, causing harm to the trustor (Misztal, 2012, p. 27). Vulnerability motivates trust, so understanding the different (potentially multiple) vulnerabilities a victim experiences can set the scene for shaping decisions to trust. Additionally, understanding nuances of focal vulnerability (the vulnerability one considers when entering a trust relationship) was theorized to provide insight into harm. The perspective of hindsight in giving the VIS after the abuse had occurred allowed victims in both groups to identify factors that they viewed as contributing to their vulnerability, but it was difficult to identify vulnerability that may have been considered focal at the time they were deciding to trust Nassar or institutions. So, while the results in general do not provide much insight into differences in focal versus imposed vulnerability, the two groups looked very similar in terms of vulnerability before entering the trust relationship. Surprisingly, though, victims in the interpersonal group did describe how institutional context (e.g., reputation, position of power, elements of physical space) was relevant even without explicitly endorsing institutional trust. The primary differences emerged, as expected, with the multilevel group describing how institutional actors, rules, policies, and safeguards more overtly shaped their perceptions of vulnerability. These results add nuance to the growing literature on how identity and situation shape imposed vulnerability (Montgomery et al., 2008). Both groups described additional factors than 93 are traditionally considered, such as factors linked to identity (age) or experience (naïve, innocent, didn’t know much about medical procedures), or life situation (a sick parent, desperate to get better from an injury). Situational vulnerability has been less explored in imposed vulnerability literature, but here was tied to family (injured/sick parents who couldn’t attend appointments) and desperation to have injuries healed. This emerged more prevalently in the interpersonal group, where only 3 individuals in the multilevel group endorsed a similar code. Perhaps the multilevel group had different situational vulnerability because they had trust in institutions to provide other options and were less desperate to trust Nassar as an individual. However, taken together, for both groups, imposed vulnerability may have been linked to numerous intersecting aspects of their identities or situations (e.g., young and a sick parent). This supports the recent move in trust research to explore intersectional vulnerability (Kuran et al., 2020, Darcy et al., under review). Additionally, vulnerability was shaped by victims’ disempowered status relative to Nassar and/or the institution. This aligns with research on understanding how risk of harm and perceptions of safety are relational, suggesting relative status (in comparison to the trustee) is integral to shaping perceived vulnerability (Wildavsky, 1988). However, victims were less explicit in discussing their power in relation to institutions. This is a limitation of the sample generally, where victims were not asked specifically about institutional trust so perhaps did not contemplate their power in relation to it. However, this reflects a larger struggle in the literature in understanding vulnerability to institutions (Hamm, Smidt, & Mayer, 2019). While some research has considered the role of power in trust relationships (Luhmann, 1979; Möllering, 2019), more work should explicitly examine how vulnerability relates to relative power in different types of relationships. 94 A novel facet of imposed vulnerability that arose in both groups was the role of institutional structure, namely, physical space. The results align with theory on sexual geographies as related to risk for sexual assault in institutional settings (Armstrong et al., 2006; Hirsch & Khan, 2020). As Hirsch and Kahn (2020) explain, “sexual danger is an unintended consequence of university policies that are intended to be gender neutral” (p 488). This means that certain characteristics arise as a result of policies (e.g., residence halls and how they are regulated) which shape student behavior (e.g., push students to party off campus as bars, parties, and fraternities). To contribute to vulnerability in the context of trust, physical space arguably needs to be in control of the trustee. The institutional spaces described by victims were purposefully leveraged by Nassar to shape perceptions of vulnerability and bolster trustworthiness (he wore MSU shirts, flashed the Olympic logo, put photos on his wall of well-known gymnasts). Institutional affiliation helped victims understand vulnerability, in part signaling safety (e.g., MSU’s campus, in back rooms of gymnastics gyms or high schools). Additionally, the visual aspects of the space connoted safety by displaying photos of previous patients—including gymnasts and Olympians--many of whom were known to or admired by the victims. Finally, physical space intersected with other forms of trustworthiness for many, including having a parent present in the room, which bolstered trustworthiness of Nassar (for both the participants and their parents). However, Nassar manipulated the space by blocking the parents’ view of the procedure and performing it in their presence without permission or precautions. Physical space as related to perceived vulnerability is a rich area for future research, but also could relate to policy changes that are more aware of the role of space in sexual assault prevention (Hirsch & Khan, 2020). 95 Finally, some unique themes arose specifically in the multilevel group around institutional safeguards they relied on to understand their vulnerability. Victims described ignorance or lack of experience with doctors in ways the interpersonal group did not. Perhaps they relied more strongly on the structures in place to ensure doctors were properly trained to understand their vulnerability. For example, they pointed to structures that reduced the risk of misplaced trust like the reputation of an institution, laws or policies designed to facilitate reporting, or the culture/norms of an institution (Bachmann & Inkpen, 2011). Specifically, they discussed how they trusted MSU to have the best doctors, or how Nassar was a fixture in their gymnastics gyms and was known as the “gymnastics doctor.” These factors helped inform their vulnerability in light of these institutional safeguards they believed would govern his behavior. They also pointed to the importance of individuals who represented institutions or were particularly formative in their perceptions of vulnerability (e.g., gymnastics coach John Geddart). This adds to existing understanding of multilevel trust by considering how reputation might intersect with role-based trust and shape perceptions of vulnerability to individuals and institutions (Kramer, 1999). Ultimately, though, these safeguards failed when no one was implementing restrictions against Nassar after internal investigations or mandatory reporters failed to report abuse. The results on vulnerability from the multilevel group provide insight into avenues for prevention and addressing sexual assault directly from those who it has directly impacted. Victims in the multilevel group organically discussed suggestions of how to prevent abuse from occurring, which aligned with Bachmann and Inkpen’s (2011) framework for institutional trust building. These more macro-level factors (e.g., corporate reputation, legal provisions, certification by exchange partners, and community norms and structures) can signal trust 96 (Bachmann & Inkpen, 2011; Carvalho & de Oliveira Mota, 2010b; Harris et al., 2013). Namely, they suggested institutions could improve their safety moving forward by improving culture and training of institutional actors. Specifically, they identified vulnerabilities inherent in the culture of gymnastics related to silence and obedience, and instead urged transparency and external oversight. Additionally, victims were actively working to change legal provisions surrounding vulnerability in institutional spaces, testifying in Congress to amend relevant legislation. While Bachmann and Inkpen (2011) do not necessarily engage directly with an individual’s vulnerability to an institution, they do directly discuss how institutions can facilitate trust and their theoretical processes are supported by the evidence here. Additionally, these policy, legal, and cultural suggestions to reduce victims’ vulnerability and promote trustworthiness are rich examples of concrete action grounded in victim experience to which institutions should take heed. Trustworthiness A primary takeaway from both groups was that victims identified multiple layers of trustworthiness that intersected with and shaped their willingness to be vulnerable. What they flagged in hindsight was that trustworthiness signals were used abusively to groom them for abuse and deter reporting. It has been suggested that “criminals, not moral philosophers, have been the experts at discerning different forms of trust” (Baier, 1986, p. 234) and that was the case with Nassar. Grooming is described as an “abuse of trust” where the perpetrator pretends to be friendly and trustworthy (Ben-Yehuda, 2001; McAlinden, 2006). The behaviors he used to signal benevolence and ability mapped on to the literature on grooming, which often relies on deception (e.g., manipulation), establishing affection (e.g., getting to know their interests) and bestowing inducements (e.g., favors, gifts) (McAlinden, 2006). These results suggested that perpetrators of 97 ISA might intentionally display attributes of interpersonal trustworthiness (benevolence and ability) while stressing the reputations of the institutions in which they operate to facilitate grooming. While broad literature has focused on how institutions or individuals might foster or rebuild trust in the institutional context, the reality is that ability and benevolence can be weaponized to groom individuals in the context of ISA (Stevens, MacDuffie, & Helper, 2015). These results suggest that trustworthiness, namely, benevolence and ability and institutional reputation can be a frame for understanding grooming. Secondly, the types of factors that the different groups relied on in assessing trustworthiness add nuance to existing theory. Typically, in an interpersonal relationship, trustworthiness is informed by assessing a trustee’s ability, benevolence, and integrity (Colquitt et al., 2007). However, the literature disagrees as to what forms trustworthiness perceptions of an institution. One may look to ability, benevolence, and integrity of the institution (Hamm et al., 2019), or to more macro factors, as discussed above (Bachmann & Inkpen, 2011; Tan & Tan, 2000). The multilevel group described both benevolence (a traditionally interpersonal factor) as well as institutional reputation, culture, and certification by exchange partners (more traditionally institutional factors). The interpersonal group described both benevolence and ability, but also alluded to some more traditional institutional factors in ways that were surprising because they did not endorse institutional trust. It is perhaps not surprising that the interpersonal factor of benevolence applied in both the interpersonal and the multilevel trust group. Mayer, Davis and Schoorman’s (1995) model of trust was intended to be a muti-level model, and “all three factors of ability, benevolence, and integrity can contribute to trust in a group or organization.” (p. 345). However, a novel finding is that integrity was not described by either group, which could be because integrity (that the 98 trustee adheres to a set of principles that the trustor finds acceptable) was wrapped into benevolence (Schoorman et al., 2007). Because benevolence for both groups implicated institutional setting, perhaps the institutional culture, norms, and rules that informed perceptions of Nassar’s ability also signaled his integrity. Additionally, research has suggested there is a high correlation between integrity and benevolence (Colquitt et al., 2007; Schoorman et al., 2007). Here, the results suggested that ability (especially as tied to both Nassar’s and institutions’ reputations) shaped trustworthiness before a participant even met Nassar, but benevolence was also portrayed through his behavior or roles he came to occupy over time. Ability only arose in the interpersonal group but was unexpectedly tied to institutional context. For example, victims alluded to institutional contexts (e.g., in Nassar’s passive associations with other institutions), or institutional actors (e.g., referrals by other professionals) even though this group did not describe institutional trust, blame, or accountability. This was the case even for those who were not aware of Nassar’s reputation—they still relied on others’ referrals, arguably implicitly drawing on institutional certifications or associations for assessing ability. Ability was signaled by evidence of his skill, competence, and influence as a medical professional, specifically, a sports medicine physician (Mayer et al., 1995). However, these signals intersected with institutional reputation and culture and certification by exchange partners (e.g., the medical licensing board) (Bachmann & Inkpen, 2011). The fact that ability did not appear in the multilevel group was surprising, but it was likely because they endorsed a separate code about institutional loyalty, and Nassar’s associations with institutions with which they identified and trusted influenced their perceptions of his medical ability to help them heal. Some unique findings emerged in the multilevel group which complicate our understanding of institutional trust. Institutional loyalty was identified as an important factor by 99 this group, influencing trustworthiness perceptions. For example, victims noted that they trusted that the institutions with which they had existing affiliations, love, and loyalty (e.g., MSU, Olympics, USAG), would only have the best doctors working for them. This directly contributed to the trustworthiness of Nassar. Relatedly, socialization in specific institutions shaped trustworthiness perceptions. For example, elite gymnasts described how they were trained to obey coaches and not question authority, which gave more weight to referrals by these actors and influenced their willingness to be vulnerable. This troubling socialization of gymnastics culture was described as promoting obedience, unimaginable pain, and silent suffering has been identified as a risk factor for abuse and changes have recently been suggested to promote transparency in these institutions (Pesta, 2019). However, victims described how socialization and loyalty was perceived as weaponized by Nassar in subtle ways. For example, he had offices or space at numerous institutions, he wore branded clothing, and gave gifts with institutional branding (e.g., USA Olympic towels) in ways that bolstered their trust of these institutions and of him. Similarly, signals of Nassar’s skill were explicitly and closely tied to institutional reputation. The ways in which he bolstered trust in these institutions is seemingly also implied. Nassar was a “world-renowned doctor”(AA) and he likely helped elevate these institutions’ reputations working with and for them, which then contributed to trustworthiness of both in eyes of victims. That is, institutions employing a world- renowned physician must be trustworthy institutions. By leveraging or flaunting his associations with institutional contexts to portray ability and benevolence Nassar may have been performing institutional grooming (McAlinden, 2006), but the symbiotic relationship made it difficult to disentangle trustworthiness and explains complexity in identifying sources of harm. Clearly loyalty and trust are closely related (Carvalho & de Oliveira Mota, 2010b), but how institutional 100 loyalty might influence trustworthiness perceptions is an area for future research to explicitly engage. Conclusion and Limitations There are some important limitations to the conclusions. The victim impact statements are secondary data, and the participants were not directly asked about trust or trustworthiness of individuals or institutions in the context of their institutional sexual assault. While the groups were split into those that described only interpersonal and those that described multilevel trust, clearly some themes implicating institutional context arose even in the interpersonal group. This could be a limitation of the data or could just be inherent to the nature of institutional sexual assault, which necessarily involves some institutional actor or setting. Limitations of the data were outweighed by the novel contribution and victim-grounded insights from this analysis. However, future research should attempt to differentiate between interpersonal, multilevel and institutional vulnerability and trustworthiness as relates to victim harms. Understanding harm that stemmed from trust relationships required a closer look at the first stage of the trust process, namely, understanding a victim’s vulnerability and the reasons why they might accept that vulnerability (trustworthiness perceptions). These results provided essential insight into both imposed and focal vulnerability, where identity, situations, and institutional structure intersected to shape perceptions of vulnerability. Importantly, institutions arose (to a lesser extent) even for the individual group. But institutions do shape socialization and perceived vulnerability, where victims stressed the importance of physical space. As to trustworthiness, instead of relying clearly on either interpersonal or institutional trustworthiness signals, both interpersonal and multilevel groups blurred lines between traditional indicators of interpersonal and institutional trustworthiness. However, traditional signals of trustworthiness for 101 interpersonal trust (ability and benevolence) and institutional trust (reputation, laws) were identified as signals that were used to groom the victims in ways that literature has not explored. 102 CHAPTER 4: THE POWER TO DAMAGE: UNDERSTANDING VICTIM ACCOUNTS OF INTERPERSONAL AND MULTILEVEL TRUST AND HARM Abstract Institutional sexual assault continually comes to light in multiple settings and is associated with distinct or exacerbated harms for victims. This research examined victim descriptions of harm that stemmed from interpersonal and multilevel trust (trust in a person and trust in an institution) by applying constant comparison and grounded theory methods to 127 victim impact statements given at Dr. Larry Nassar’s sentencing trial. Results suggested that both groups identified physical and mental health impacts that stemmed from the breach of trust, but the multilevel group identified broader impacts (e.g., feeling uncomfortable in their body). Two mechanisms emerged to explain how harm was experienced by victims. Betrayal largely stemmed from changes in worldview and impacted relationships (with oneself, with others, with institutions) and was tied to isolation and voicelessness for both groups. Victims in the multilevel group described feeling institutionally disempowered and expressed a novel harm to their trust in institutions. The mechanisms of moral injury related to damage to identity for both groups but was surprisingly tied to institutional and amorphous harm even for the interpersonal group. For the multilevel group, guilt and fear emerged as emotional responses and damage to institutional identity emerged as a novel harm. Practical and legal implications are discussed. Introduction The strong words of Rachael Denhollander summarize the potential for harm that stems from trust relationships tied to institutional sexual assault (ISA): "I gave them the power to damage me when I gave them my trust" (Denhollander, 2019, p. 116). By sharing her experiences of ISA in 2016, she motivated hundreds of victims to come forward, identifying a 103 range of assaults by Nassar over decades and intersected with numerous institutional settings (Levenson & Joseph, 2018; Winowiecki, Lafond, & Wells, 2018). Victims described trust in both Nassar and a range of institutions (e.g., universities, local gymnastics clubs, medical organizations, national sports organizations) that was betrayed (Tracy, 2018). Legal scholars, community members, victims, and reporters all agreed: the victims described elements of both interpersonal and institutional betrayal (Citrino, 2018; Hauser & Astor, 2018; Martindale, 2020; Tracy, 2018). In their statements, victims pointed out that Nassar fostered a trusting or dependent relationship, which “blinded” them to the abuse for many years, key attributes of betrayal trauma (Birrell & Freyd, 2013; Rahal & Kozlowski, 2018). Additionally, many victims described behaviors of institutions that exacerbated the impact of the abuse (e.g., institutions and institutional actors didn’t take proactive steps to prevent this experience; created an environment in which this experience seemed more likely to occur) elements of institutional betrayal (Banta, 2019; C. P. Smith & Freyd, 2014). Trust is at the center of betrayal trauma and institutional betrayal theories (Birrell & Freyd, 2013; C. P. Smith & Freyd, 2013), but how trust relates to harm is not explored. Trust is typically conceptualized the willingness of a trustor to be vulnerable to the actions of another party (Mayer et al., 1995). Only in understanding the different stages of the trust process is it possible to tease apart distinct harms stemming from these trust relationships and better understand how to help this subset of victims. The prior chapter explored the vulnerabilities and the multiple signals of trustworthiness (at multiple levels) that contributed to victims’ decision to trust. That stage is essential to understanding what harms they might have contemplated and why they accepted vulnerability. Namely, focal vulnerability might include only parts of broader imposed vulnerability, and trustworthiness perceptions help an individual 104 accept a willingness to be vulnerable (Mayer, Davis, Schoorman, 1995; Schoorman, Mayer, Davis, 2007). However, key to understanding harms that stem from a breach in a trust relationship is understanding the second stage of the trust process (Figure 6). Once trust is established, it may make a victim more willing to accept vulnerability in the ongoing trust relationship (Mayer et al., 1995). This, then, might create different vulnerability than existed before trust (emergent vulnerability). For example, relying on their previous trustworthiness assessments (and increased dependence on the person or institution), a victim might feel an increased willingness to be vulnerable, perhaps by monitoring the trustee less (Colquitt et al., 2007; Mayer et al., 1995; Misztal, 2012; Scott, 2010). This potential additional vulnerability that emerges within trust is underexplored in the literature. While no research has explicitly looked at emergent vulnerability, some disparate research has examined harms associated with breach of trust. Complex emotional feelings can be wrapped up in interpersonal betrayal (Ben-Yehuda, 2001; Birrell & Freyd, 2013; M. L. E. Chan, 2009), indicating that emotional harm might be tied to trust breach. This has been less explored in institutional or multilevel trust, but some research similarly suggests a breach of psychological contract might stem from breach in institutional trust (Morrison & Robinson, 1997; C. P. Smith & Freyd, 2013). Some prior literature has explored discussions of pain, hurt, betrayal and accountability specifically in experiences of ISA victims, concluding that trust did emerge as relevant in these accounts (Abrams & Potts, 2020; Eiler, Al-Kire, Doyle, & Wayment, 2019). Finally, some theorizing on moral injury stemming from other institutional contexts (e.g., the military) suggests there might also be damage to one’s identity that stems from trust violations (Jinkerson, 2016). However, no research has specifically examined how the trust 105 process (trust and emergent vulnerability) can provide insight into unique harms associated with ISA depending on type of trust relationship. The current research provides insight into harms that stem from process of trust by drawing from the experiences of victims of ISA by Nassar. Victims described how the abuse implicated both interpersonal and multilevel trust which implicated both interpersonal and institutional betrayal (Citrino, 2018; Hauser & Astor, 2018; Martindale, 2020; Tracy, 2018). Constant comparison and grounded theory methods were used to examine 127 victim impact statements (VIS) given at the sentencing trial of Nassar. This method allowed for comparison of individuals that endorsed interpersonal trust (e.g., trust in Nassar) with those who endorsed multilevel trust (e.g., trust in Nassar and an institution) to explore insights into emergent vulnerability and harm stemming from different trust relationships. The results provide novel insight into emergent vulnerability and mechanisms for understanding harm to better provide this subset of victims with prevention, recovery, and advocacy resources. Literature Review When ISA occurs in a relationship of trust, it is theorized to lead to distinct harms for victims (Freyd, 1996; C. P. Smith & Freyd, 2013). Betrayal trauma theory suggests sexual assault occurring in a close, trusting, or dependent relationship with another person relates to distinct harms, whereas institutional betrayal suggests that an institution’s response to rape and sexual assault can result in similar harms (Goldsmith et al., 2014; C. P. Smith & Freyd, 2013, 2014). Negative impacts for both types of betrayals include PTSD, depression, dissociation, suicidality, physical health issues, relationship, and substance use issues (Birrell & Freyd, 2013; Delker & Freyd, 2014b; Deprince, 2001; Freyd, 1996; Freyd et al., 2005; Monteith et al., 2016; C. P. Smith & Freyd, 2014). Both betrayals are perceived by the victim as a violation of the trust 106 relationship they had with the person or institution, and harm stems from that betrayal. ISA is particularly complicated because it might implicate multilevel trust, that is, interpersonal trust in a perpetrator and institutional trust. Limited research has explicitly explored harms that are uniquely related to ISA. Abrams and Pott (2020) explored the sources and types of harm described by victims of Nassar and found that naming strategies for Nassar implicated institutional structures and actors, signaling that institutional context provides insight to understand harm (Abrams & Potts, 2020). Harms were linked to emotional actions, states, and processes, as well as psychological actions, states and processes (Abrams & Potts, 2020) suggesting complexity of emotional harm. Additionally, Eiler and colleagues (2019) identified differences in the experiences of those who disclosed sexual violence in sports context compared to those who disclosed in other contexts and found that disclosures in sports context endorsed themes of power and trust. Despite flagging the importance of context and trust, existing research does not engage with the process of interpersonal or institutional trust, nor does it tease apart how that might relate to differential harm. Trust Process As mentioned in the previous chapter, trust refers to the willingness of a party to be vulnerable to the actions of another (Mayer et al., 1995). The previous chapter explored the first two stages of the trust process, focusing on vulnerability expressed by victims and signals of trustworthiness from both Nassar and different institutions that helped them to be willing to be vulnerable. The vulnerability that is relational, subjective, and which the trustor is aware of when they enter a trust relationship is focal vulnerability (Hamm & Möllering, 2022) while imposed vulnerability is shaped by identity, situation, or relationship (e.g., power) (S. M. Baker et al., 107 2005; Montgomery et al., 2008). When deciding to accept that vulnerability, a victim looked to attributes of the trustee that signaled trustworthiness, which differ based on an interpersonal or institutional trustee (Bachmann & Inkpen, 2011; Mayer et al., 1995; Schoorman et al., 2007; Tan & Tan, 2000). However, the next stage of the trust process could bring risk of additional harm. When a trustor ultimately decides to trust, this increases the likelihood of risk taking in the relationship (Mayer et al., 1995; Schoorman et al., 2007). For example, in a trust relationship, one might increase dependence on the trustee or reduce monitoring of their behavior, creating additional vulnerability (Baier, 1986; Mayer et al., 1995). This creates the potential for emergent vulnerability, which is essentially the potential for additional harm based on an ongoing assessment of risks within a trust relationship. Differences by Trust Relationship Again, research supports that there may be differences in harm in interpersonal versus institutional trust. Because institutional trust is difficult to disentangle from interpersonal trust in ISA victims’ accounts, the most accurate comparisons are between interpersonal trust and multilevel trust (trust in a perpetrator and an institution). Multilevel trust is increasingly being recognized as a reality in many institutional settings (Fulmer, 2018). To inform harm, multilevel trust draws attention to the ways in which interpersonal relationships are embedded in larger institutional contexts. For example, trust and harm could relate to peers in the institution, individuals acting in certain roles, representatives of the institution, and the institution as a whole, (Fulmer, 2018; Kramer, 1999). 108 Emergent Vulnerability and Harm Theory on betrayal provides a useful frame to help explain harms that could be linked to emergent vulnerability. First, betrayal trauma theory suggests that mental health impacts stem from violations in trust relationships, including PTSD, dissociation, substance use, and impacts on future relationships (Babcock & DePrince, 2012; Delker & Freyd, 2014b; Owen et al., 2012; Taylor et al., 2020). Additional theory provides insight into mechanisms of harm that stem from trust, namely, that betrayal interferes with a victim’s frame of reference in terms of relationships, but betrayal has only been minimally explored the context of different trust relationships (Brennan, 2021; Gartner, 1999). Betrayal is tied to emotional or psychic injury stemming from a trust relationship (Brennan, 2021; Gartner, 1999). Interpersonal trust betrayal has been discussed as sexual betrayal, a form of betrayal that encompasses sexual assault, incest, and sexual trauma (Gartner, 1999). Sexual betrayal violates either implicit or explicit trust when seemingly unbreakable bonds are broken (Gartner, 1999). Importantly, this betrayal violates “the sanctity of interpersonal relationships” by causing a “break in the interpersonal frame of reference from which to view the world of interpersonal relationships” (Gartner, 1999, p. 14) and can therefore have wide reaching impacts on future interpersonal relationships. Because the betrayal itself carries a change in frame of reference in regard to relationships, “the betrayal of a trusted relationship is frequently more traumatic than the sexual acts themselves” (Gartner, 1999, p. 40). The psychic pain in detecting a betrayal is theorized to relate to a behavior change (DePrince & Freyd, 2002), namely, blindness or inability to label the behavior as abusive. Research suggests that betrayal traumas might impact social capacity including the ability to assess trustworthiness and increases risk of revictimization (Gobin & Freyd, 2009, 2014). 109 A separate mechanism that can help explain emergent harm linked to trust is moral injury. Moral injury also is tied to emotional harm, but through damage to identity instead of relationships. In general, moral injury refers to “profound and persistent psychological distress that people may develop when their moral expectations and beliefs are violated by their own or other people’s actions” (Molendijk et al., 2022). Brennan expanded the concept of moral injury in the context of trust, describing it as a type of betrayal that occurs when an expectation of a fundamental regard (i.e., being treated with decency by others) is disregarded by another person (Brennan, 2021). Moral injury stems from a breach in this recognition trust, which is an expectation that others respect our moral status as a person (Brennan, 2021). When this trust is betrayed, there are three interrelated levels of harm: first, there is innate harm in the violence or betrayal itself. Second, there is harm that relates to “ the disabused notion of one’s safety in the world” (including a newly recognized loss of connection to others) (Brennan, 2021, p. 3804). Finally, the third level is related to a loss of self, in part linked to realizing you are not in control of your identity. This harm may be tied to reactions like guilt, shame or anger for not in some way stopping this betrayal (Brennan, 2021). In other words, betrayal of recognition trust causes moral injury in that it degrades the victim’s sense of self and value, making them realize just how much they depended on others to recognize their personhood (Brenna, 2021). This stems from breaches of interpersonal trust, which then has serious impacts on a victim’s identity and self- image. Mechanisms of harm related to trust betrayal have also been explored in the institutional context. Research on workplace betrayal suggests a betrayal of the “psychological contract” that an employee believed they had with their employer will relate to psychological harms (Fitness, 2012). This contract signifies notions of fairness and the right to be treated with respect by the 110 institution in which they work, and when this is violated or betrayed they may respond with outrage, decreased trust in the organization, and negative behaviors such as industrial sabotage (Morrison & Robinson, 1997). Even within an organizational setting, betrayal of trust has broader organizational impacts, such as reduction in trust, decreased job satisfaction, and reduced intentions to stay with the organization (Fitness, 2012; Morrison & Robinson, 1997). However, there is a dearth of research on complexities of multilevel betrayal in institutional contexts. The research on moral injury stemming from institutional trust is less developed. Moral injuries could potentially “also happen gradually, when structural social conditions impose themselves on individuals day in and day out.” (Brennan, 2021, p. 3803). The risk for moral injury can be compounded by organizational stressors such as role in the organization, supervisory relationships, or organizational structure and climate (Kamkar et al., 2019), which are related to institutional trust perceptions. This builds on a broad literature that considers moral injury in different institutions, including the military, policing, and healthcare (French et al., 2022; Kamkar et al., 2019; Molendijk et al., 2022; Williamson, Stevelink, & Greenberg, 2018). This research suggests that morally injurious experiences across setting relate to psychological difficulties (e.g., PTSD, depression) (Kamkar et al., 2019; Williamson et al., 2018). While victim responses might differ, moral injuries in institutional context “can shatter an individual’s core belief system whereby one’s views of self, others, the future and the world can fundamentally change” (Kamkar et al., 2019, p. 121). Similarly, emotions such as anxiety, guilt, shame, and fear might stem from traumatic incidents linked to institutional moral injury (Blumberg, Papazoglou, & Creighton, 2019; Kamkar et al., 2019). 111 Current Study The current research explores harm using the process of trust as a lens by drawing from the example of ISA perpetrated by Nassar against hundreds of victims. Some research suggests emotional harm stemming from betrayal and moral injury might emerge as a distinct harm in ISA, however, limited research has sought to understand how victims of institutional sexual assault speak to how harms stem from different trust relationships. Specifically, this research explores how the second stage of the trust process (i.e., trust and emergent vulnerability) are tied to harms within the trust relationship. This provides insight into harms that were perhaps not contemplated upon the initial decision to trust but were shaped by decisions within the trust relationship (e.g., reduced monitoring, ignoring red flags because trust had already been established). The VIS given by victims of abuse by Nassar provided an opportunity to examine those harms associated with risk taking in a trust relationship. To explore harms described by victims, I utilized constant comparative and grounded theory methods (Auerbach & Silverstein, 2003; Glaser & Strauss, 1967). Victims were compared within and between groups of those who discussed only interpersonal trust of Nassar (interpersonal trust) and those who described both interpersonal and institutional trust (multilevel trust). I examined 127 VIS given at Nassar’s criminal sentencing explore the following questions: (1) How do victims in the interpersonal group (versus in the multilevel group) describe how trust facilitates harm? (a) in what ways does harm differ for those who described trusting individuals (versus multilevel)? (b) in what ways is harm similar for those who described trusting individuals (versus multilevel)? 112 Method Data The same data from the previous chapter were used in this research. The data come from 127 VIS given as part of the sentencing proceedings from Nassar’s criminal plea to multiple charges of criminal sexual conduct in Ingham County Court, Michigan. The final sample was limited to statements that were either read or written by victims and for whom transcripts were available11 (N=143). The sample was limited only to those victims who mentioned trust (either used some form of the word “trust” (N=120) or described key elements of existing definitions of trust (e.g., benevolence, integrity, vulnerability, reputation) (N=7)).12 This resulted a total of 127 statements given by 127 victims. The previous chapter presented demographic information about the sample. Importantly, the sample included only girls or women, for a total of 118 adults and 9 minors. Study Design Victim impact statements are a rich narrative source that allow a search for meanings based on individuals’ interpretations in their social, cultural and personal context (Orbuch, 1997, p. 461). Analysis of the VIS provide insight into how people organize views of themselves, others, and their social worlds and are one of the few spaces in the legal system where victims may provide broad evidence of experienced harms (Brunson & Wade, 2019; Orbuch, 1997). Again, because victims were not explicitly asked about trust, trust and trustworthiness literature were used as sensitizing concepts (Bowen, 2006) and as a proxy for trust. 11 Five statements made over video were not transcribed in the legal record. 12 These indicators of trust were drawn from the theoretical frame of the dissertation. 113 This study investigates how victims understand harm that stems from trust and how these patterns vary by group (interpersonal versus multilevel). To make comparisons across these groups, I separated the VIS into: (1) those victims who endorsed trust and described harm that was associated with an individual trustee (e.g., Nassar) (N=51) and (2) participants who discussed trust and harm that was associated with both an individual trustee (e.g., Nassar) and an institutional trustee (e.g., an organization) (N=76). To identify these groups, each VIS for each victim was examined, and those who only mentioned trusting an individual were sorted into the “interpersonal” trust group. It is important to note that institutional setting may have been mentioned by this group (e.g., a brief reference to seeing Nassar at offices at MSU) but that this reference alone was not sufficient to indicate institutional trust. Participants were sorted into the “multilevel group” if they described trust in an individual (Nassar) and some level of trust (also indicated by reliance, blame, or harm) to an institution. Capturing references to institutions was broadly guided by Ostrom (2005)’s definition of institutions, which refer to prescriptions used by humans to organize social interactions. In practice, this meant that the multilevel group endorsed both references to broader institutions (.e.g., “the system”) as well as specific institutions or organizations (e.g., Twistars gymnastics). Data Analysis The victim impact statements were analyzed inductively using constant comparative and grounded theory methods to compare how those in the interpersonal group differed from the multilevel group in relation to harm perceptions (Glaser & Strauss, 1967). This type of qualitative research relies on subjective experiences to inductively develop a pattern of meaning (Cresswell, 2003). Essentially, grounded theory methods utilize systematic inductive guidelines for collecting and analyzing data to build middle range theoretical frameworks (Charmaz, 2000). 114 Traditional grounded theory methods require simultaneous data collection and analysis, however, there has increasingly been a recognition of relying on secondary sources, which was the case here (Glaser, 1963). Coding To strengthen the internal validity of the analysis, grounded theory methods that entailed a constant comparative approach and the engagement with deviant cases were used. Constant comparison is what generates theoretical properties of a given category (Charmaz, 2006). Additionally, this method enhances reliability and validity by ensuring the researcher systematically checks the data and monitors analyses and interpretation (Glaser, 1965; O’Neal & Hayes, 2020). In order to apply constant comparison, I adapted Auerbach and Silverstein’s (2003) step-by-step approach which first culled the victim impact statements into relevant text and then entailed a three-level categorization system analyzing the content (see Table 8 for an example) (Auerbach & Silverstein, 2003). Memo writing accompanied each stage of coding and analysis. The first step included reading each VIS line-by-line using the process of open coding, which captured similar words or phrases that depicted the same concept. If these were endorsed by two or more individuals, they were then extracted and labeled as repeating ideas (Auerbach & Silverstein, 2003). Separate files were created for the interpersonal and multilevel groups. These were then inspected separately for patterns surrounding harm. The second step involved creating themes by organizing the repeating ideas into larger groups. Systematic comparisons were made across the themes and between the groups to identify similarities and differences in victim experiences. The third step was to identify overarching 115 theoretical constructs from the themes. This method allowed me to move up in abstraction to better understand victim experiences of harm (Auerbach & Silverstein, 2003). While attempts were made ensure that the concepts provided typified the most common patterns in respondents’ accounts, an effort was also made to identify and explain deviant cases. Where deviant cases were identified, they are described in the results. To provide context, the percentage of individuals who endorsed a particular theme in each group are presented, along with the number of individuals within a group who endorsed a repeating idea (code). As with any qualitative research, while not generalizable, this study provides further insight into how participants experienced harm stemming from trust. Results Interpersonal Trust Group Victims described how vulnerability emerged within a trust relationship that was related to direct harms to their health (Table 9). A majority (67%) of the victims in this group described some impact on physical or mental health. Physical health impacts largely related to “iatrogenic” harms, or harms that stemmed directly from medical treatment and diagnoses (Behrman & Dan- Cooke, 2019) (N = 19). They linked these harms to Nassar’s behaviors within the trust relationship, and to their own reduced monitoring, increased dependence, and cooperation (risk taking in a relationship of a patient in their doctor). Specifically, these iatrogenic harms were caused by Nassar’s incorrect diagnoses, incompetence, or negligent treatment and impacted their physical well-being or caused further injury. Some negative cases arose, where individuals described how some of his medical treatment did help them heal. For example: 116 After years of suffering I felt this weight lifted off of me. You opened my window of dreams back up. I could live in the world of dance again after four years of miserable living (BR). However, even when there was some element of healing, they often also described worse pain because of the betrayal. As noted by KS: To say you did nothing for me would be a lie. You helped me through some of the toughest times of my life physically and mentally, but now you have caused more pain than I have ever endured. Because of what you did, you knocked down both of our towers. Additionally, as V48 explained, “What he did physically to me for three years is nothing in comparison to what I have emotionally and mentally endured. I still have medical issues that he was supposed to be treating me for but failed.” Harms stemming from the trust relationship were difficult to understand. In fact, KP lamented the “confusion” that “comes from both being helped and harmed at the same time” which then caused mental stress for victims. In relation to the ongoing relationship, victims described that they feared he was purposefully not treating real injuries (or misidentifying injuries) to enable continued access to them for abuse. “He found an injury that no other doctor found and the injury lasted longer than planned. I wonder if there was even something seriously wrong or if he did it for himself and found something that didn't exist? (JS).” If doubts arose, victims discounted them and continued to trust, opening themselves to harm within the trust relationship. As JB recalled: “I told myself I needed to trust him, too, so I convinced myself it was just a normal part of the procedure and kept my mouth shut.” Nassar took advantage of their established trust, corresponding 117 dependence, and reduced monitoring of his behavior to continue to abuse and cause distinct harm. For some, abuse continued for years in addition to their injuries not being addressed. A broad array of direct harms in the form of mental health-related impacts also emerged in participants’ experiences. There were separate mental health impacts associated with coming to terms with the betrayal of trust. I need to change 12 years worth of knowing a person and feeling -- and feeling that I could trust them completely. I have been in therapy every week for the past year plus now trying to learn coping skills for my fears and learning to trust myself and instincts again. In the past I have struggled with anxiety, depression, and suicidal thoughts with no explanation. This was prior to understanding what Nassar had been really doing to me. I still struggle with anxiety and depression, and although I have yet to completely accept all the events that have happened this past year, I do have these moments where reality hits me and it hits hard. I feel like I can't breathe. I can't think straight, and everything feels upside down (AY) Others described panic attacks, depression and self-harm which impeded their ability to live their lives; as AG plainly stated: “The anxiety has overtaken my life.” A troubling subset (N = 7) also discussed contemplating or attempting suicide and “fighting off suicidal thoughts (KJ).” These mental health impacts impacted sleep, related to flashbacks and nightmares, and led to extreme frustration. These impacts may have been tied to facing the betrayal, or as KP said, accepting the truth. In her words: “So many times I have contemplated ending my own life thinking that torturing myself would be better than accepting this truth because a least then I would have control over the pain.” 118 Multilevel Trust Group Negative health impacts that stemmed from trust emerged in even more victims than in the interpersonal group (83%). In some ways the confusing feelings around these harms were similar, where AM said “saying that you did nothing for me would be a lie. You helped to heal me. My fragile bones, that is.” However, she noted, “you also abused your power and my trust in you, and that is not okay.” As summarized by OV “Larry was someone we trusted and someone who was supposed to be taking pain away when, in fact, he was causing more pain than I ever thought I would feel.” Again, this pain was distinctly tied to breach of trust and failure to heal. An important difference between different individuals within this group had to do with the context in which they interacted with Nassar. For example, because he was considered “the gymnastics doctor,” it meant his diagnosis was trusted by coaches above the experienced pain felt by participants. For example, IH noted that Nassar’s diagnosis was the final word on her injury, and this led to a bad injury getting worse. I saw Larry and he examined it and he thought there was nothing wrong. Therefore, if Larry said nothing was wrong, then nothing was wrong, even if it hurt me to walk, and I was not allowed to get a second opinion. . . Then during a practice the week of nationals I could barely walk because my leg hurt so bad. I ended up getting screamed at and kicked out of the gym and I went to the ER immediately after. The x-ray showed a broken leg. It looked like an axe splitting a piece of wood, and every time that I tumbled the bone splintered more and more, so for over a month I practiced, competed, and made it to nationals on a broken leg because Larry Nassar said that there was nothing wrong, and we believed the child molesting doctor over the child who was the one experiencing pain. To this 119 day I wonder if I was ever getting accurate medical treatment…. Did he ignore what was wrong with my leg so he could -- so I would continue seeing him for longer? In a comparable situation, ML described how she was injured at a gymnastics camp but “Larry checked out my x-rays saying I was fine.” As a result: I was literally left crawling the rest of the camp or using a rolling office chair I found in the lounge to get around. He didn't even wrap my feet for the flight back home. When my doctor checked my x-rays, I had a fracture. I simply cannot even get myself to consider you as a real doctor. Your priorities should have been my health, yet your priority was solely to molest me. Both of these examples show that his institutional reputation as “the gymnastics doctor” meant they continued to see him and trust the gymnastics institutions they were a part of, even when his treatments (or mistreatments) were causing them physical pain. Their actions within the trust relationship differed from the interpersonal group. Their dependence on him was perhaps even greater, they were pressured to continue seeing (and trusting) him, and his institutional association contributed to unique iatrogenic harms. Similarly severe mental health impacts were described by victims in the multilevel group. They described paranoia, anxiety, depression, sexual issues, insomnia, panic attacks, dissociative episodes, eating disorders, flashbacks and PTSD. As MM summarized, Nassar “left scars on my psyche that may never go away.” Again, these mental health impacts related to wide-reaching damage on their lives, whether it was self-medicating with alcohol, a pervasive sense of fear, or struggles that “made a normal social life impossible” (KT). As LL retorted to the President of 120 Michigan State University (who was perceived to not have attended the hearing because of a scheduling conflict): I can assure you none of us had the time in our schedules for the past 20 years for Larry Nassar to abuse us, but we had no choice. Neither did we have the time in our schedule for therapy, tears, stress, anxiety, panic attacks, sleepless nights, guilt, or, for some, self harm. But we did anyways, because of you and the others who let Larry into our world. They also described suicidal thoughts or behaviors, but different harms arose around shame, such as feeling uncomfortable in their own body or embarrassed. For example, TC confessed: “there are days I hate being in my own skin.” A wide variety of far-reaching mental health impacts emerged from these trust relationships and were incredibly salient and impactful as described by victims in both groups. Interpersonal Mechanisms for Understanding Harm Participants in both groups also endorsed themes that mapped to different mechanisms that were tied to complex emotional harms. These provided important insight into additional harms that emerged tied to betrayal within the trust relationship. The theoretical constructs of betrayal and moral injury emerged for both as theoretical constructs and thus frame the results (Table 10, Table 11). Victims’ accounts of emergent harm mapped on to sexual betrayal, which causes a change to one’s frame of reference or worldview, thus impacting relationships (DePrince & Freyd, 2002; Gartner, 1999). Here, when victims’ trust was violated, a small number of participants in the interpersonal group (12%) described a shift in how they viewed themselves in relation to the world. Namely, they described feeling their voice (N = 3) and credibility (N = 3) 121 were damaged. Individuals in this group described how Nassar “took my voice away” V153, feeling silenced. The impact of the betrayal was that “you isolated me by fear because no one would believe me about what you were doing.” V153. A little more than three-fourths (76%) of the victims described widespread damage to their futures and or relationships. A shift in worldview was tied to a change in their relationships with their careers, sports, or even their dreams and goals. For example, “I didn't tell a [soul], and I try not to think about it anymore, but my dream of becoming a sports medicine doctor ended that day along with my happy and trusting self. He had broken me. JT” Additionally, they described damage to relationships with family, intimate relationships and difficulty being around men (including coworkers, male doctors, possible romantic partners). As KS explained: Larry Nassar wedged himself between myself and my family and used his leverage as my parents' trusted friend to pry us apart until we fractured. And fractured we did. My relationship with my mother is still marbled with pain, anger, and resentment, and for a long time I told people that I did not have a family. Some described pain in fearing that their parents might blame themselves It pains me to think about how my parents must feel. I'm sure to some extent they blame themselves for this happening to me. Yes, my dad took me to the appointment and, yes, he was in the room with me when the horrific crime took place. There was no way he could have ever imagined that this would happen. I know my mom has been beating herself up with who referred us to Mr. Nassar and was it one of her connections? I can start crying thinking about what my parents are going through. That's what hurts the most (V127). 122 Alternatively, victims described trusting parents to protect them from abuse, but Nassar’s betrayal consequently reframed (and harmed) these relationships. For example, “I felt abandoned. I lost faith and trust in her role as a mother (V48).” The betrayal of trust also related to discomfort around men and a direct impact on intimacy. “Almost every friend I've ever had has told me that they don't understand me, they don't understand why I want to live alone. I don't like being around other people when I'm vulnerable. KM” Finally, betrayal culminated in damage to trust in one’s self and others, as KS lamented: Because of you I now find it hard to trust not only myself but everyone around me. I am constantly questioning people's intentions about everything. You were one of the most well trusted people in my life. I thought you generally cared about my well-being and me. If you were able to do this to me, what would stop the next person? I would have never thought trust and loyalty would have been a bad quality to have until now. Understanding the mechanism of betrayal as tied to shift in worldview explains damage to numerous central relationships (oneself, family, intimate relationships). Aversion to being vulnerable or difficulty trusting also stemmed from the emotional impact of the betrayal, which aligns with research on impacts of betrayal trauma (Gobin & Freyd, 2014). Multilevel Mechanisms for Understanding Harm Victims in this group also endorsed a change in how they viewed themselves in relation to the world, but instead described betrayal in institutional context. Almost three times as many (35%) compared to the interpersonal group (12%) endorsed themes of voicelessness. They described a similar sense of isolation and lack of power as impacts of 123 betrayal. However, unlike the interpersonal group, some tried to reclaim power and voice. As JH said, The voices of all the little girls you so casually destroyed are rising up and now we are loud enough for everyone to hear. Because of the tenacious courage that your victims have shown, I now know we are strong and believe with all my heart that we will thrive despite your best efforts to violently harm us. The wide-ranging impact on relationships were described by many in this group as well (69%). As KL explained: From that point on my life was forever changed. Immediately following that appointment I no longer loved the sport of gymnastics and I didn't even know why. Then I realized it didn't matter because you already told me to quit for an injury I never had, and you knew how much gymnastics meant to me. Or, for some, embarrassment about the betrayal caused a wedge in relationships, as V10 expressed “shame, guilt, and disgust prevent me from trusting my family, will -- trusting that my family will support me unconditionally.” However, almost three-fourths of victims in this group described distinct harm in being institutionally disempowered (73%). Unlike the interpersonal group, they implicated betrayal by multiple trustees. For some, Nassar’s betrayal was seen as facilitated by other institutional actors, which compounded harm to their dreams and goals. As IH explained about her gymnastics coach who she feared allowed the abuse to occur in his gym: The dynamic duo that is Larry Nassar and John Geddert [gymnastics coach] had lasting effects on me that go beyond physical ones… What was once a sport that I 124 loved since I was five years old and would give up school and a social life for turned into my own personal prison. They often could not identify why, but their passions were tainted and no longer sources of joy. These responses also align with the theorized response to betrayal trauma, including “blindness to betrayal” as a coping mechanism when abuse occurs in a relationship of trust and dependence (Birrell & Freyd, 2013). Feeling disempowered by institutions then impacted their ability to trust institutions. As AR told Nassar: The effects of your actions are far-reaching. Abuse goes way beyond the moment, often haunting survivors for the rest of their lives making it difficult to trust and impacting their relationships It is all the more devastating when such abuse comes at the hand of such a highly regarded doctor, since it leaves survivors questioning the organizations and even the medical profession itself upon which so many rely. I am here to face you, Larry, so you can see I have regained my strength, that I am no longer a victim, I'm a survivor. I am no longer that little girl you met in Australia where you first began grooming and manipulating. For others, institutional actors silenced them in conjunction with Nassar. This may have been tied to norms in gymnastics culture not to question the decisions of coaches and trainers. As LL explained, “[John Geddart] brainwashed me and so did Larry. I couldn't speak up for myself because I was seen as disrespectful. I didn't have a voice, but now I do, so I hope you are ready.” Similarly, LB described how “You and Kathie [MSU gymnastics coach] silenced me.” They also described feeling like they didn’t have a voice in the institutional setting, perhaps a result of institutional culture. For example, an institution with an environment where they didn’t feel safe 125 to speak up, being told not to question authority as it was disrespectful, not being believed or listened to, or not taking reports seriously all caused harm and kept victims silent. LL explained: And, lastly, Michigan State University. Shame on you. I went public about my story back in January of 2016, and let me tell you, I was terrified. I was terrified because of what you would do to me. As a full ride athletic scholarship athlete on the gymnastics team, I was worried about my consequences and what they would be if you realized who I was and went public about my story. How messed up is that? You created the type of environment where victims were afraid to speak up. Little did I know you did this for years. I was afraid my scholarship would be taken that I worked so, so hard for. The scholarship that I went through years of abuse by multiple people to get…To USAG who I don't think is here today unfortunately, I wish they were, but I'm just going to be blunt and start by calling you out for paying millions of athletes millions of dollars to stay quiet about Larry Nassar. Multiple types of silencing contributed to harm in feeling disempowered for this group. Interestingly, only multilevel group endorsed a code specifically using the term betrayal. Some referred directly to betrayal by Nassar, like AA, who said “I was taught that it is not okay for anyone to touch you down there unless it's a doctor, and you were a world-renowned doctor. You betrayed my trust in the most vile way possible.” Or, betrayal may have implicated other institutional actors (e.g., coaches), or the very structures the expected to be in place that ultimately failed them caused distinct harm. As JW summarized: “My parents trusted USA Gymnastics and Larry Nassar to take care of me and we were betrayed by both.” Multilevel trust seems to map on to multilevel betrayal, and thus, multilevel impacts. It was not just institutional action, however, that caused harm to victims in the multilevel trust group. Many described how institutions turned a blind eye to abuse (N = 33) which referred 126 to institutions standing by or keeping silent. As IH poignantly said: “If you are not a part of the solution, then you are a part of the problem.” Victims described a shift in their worldview as relates to the ability (or motivation) or institutions to protect them. Institutional inaction was perceived as particularly hurtful because it was “deliberate.” Unfortunately, victims identified reasons why they believe this occurred. In part, they concluded that institutions were motivated to turn a blind eye in prioritizing money or reputation over the safety of institutional members: I will never fully understand the evil that motivates an adult to abuse an innocent child, but I do understand the evil that organizations like USA Gymnastics and Michigan State to turn a blind eye to this abuse. It is the evil that places money and medals above the welfare of children (JA). The institutions were perceived to have betrayed their trust in exchange for reputation, gold medals, or profit, at the cost of victims’ safety. Finally, participants in the multilevel group described how there was harm inherent in enabling Nassar to continue abuse. This was comparable to harm in turning a blind eye: “MSU and USA Gymnastics are to blame for employing and enabling this predator and for turning a blind eye (KT).” Additionally, there was a betrayal felt when institutions not only enabled him but promoted him or rewarded him. Some explicitly note that this enabling ended up “exacerbating” the harm that could have been prevented (CW). The enabling and harm associated with it is perhaps more upsetting to victims because multiple institutions were implicated. As MM pointed out, A question that has been asked over and over, how could Larry Nassar have been allowed to assault so many women and girls for more than two decades? The answer to that question lies in the failure of not one but three major institutions to 127 stop him; Michigan State University, United States Gymnastics Association, and the United States Olympic Committee. Going a step further, V136 notes that institutions enabling his behavior, covering up what he did, or ignoring reports entailed “shameful negligence” and therefore seeks “to hold universities, medical establishments, and those that practice medicine accountable.” In the words of AR, Over those 30 years when survivors came forward, adult after adult, many in positions of authority, protected you telling each survivor it was okay, that you weren't abusing them. In fact, many adults had convinced the survivors that they were being dramatic or had been mistaken. This is like being violated all over again. Realizing that institutions failed them or actively betrayed them in multiple ways mapped on to a shift in how institutions were viewed and was linked to multiple harms distinct from the interpersonal trust group. Violations in what were viewed as sacred interpersonal or institutional trust relationships explain the mechanism of betrayal and impacted multiple types of relationships after the betrayal. Interpersonal Group Broadly, many victims (65%) in this group endorsed themes that indicated damage to identity which map on to the theory of moral injury stemming from a breach in recognition trust (Brennan, 2021). The mechanism underlying harm stemming from moral injury is theorized to have three distinct levels of harm. The first level of harm (in the betrayal itself) emerged in codes about victims grappling with realizing they were abused. As MJ explained, there was harm in coming to terms with the identity of “victim” because she feared it would change her life. “I now know I was struggling so hard to decide if it was abuse or not because I knew if I admitted to 128 myself that it was, it would change my life so much, and it did (MJ).” While all labeled the behavior as abusive by the time they gave victim impact statements, not all individuals realized they were abused (or that trust was betrayed) when it occurred. For some, they only realized they had been victimized when the abuse was discussed in the 2016 Indystar article. GA said “I remember that exact moment I realized I had been molested by somebody I trusted, that I was one of the gymnasts he had abused, and my life was never going to be the same, and that it was true. I realized I was one of the gymnasts he had abused and my life was never going to be the same”. The damage was hard to describe, as JD explained, The moment I heard about the questionable procedure my heart sank. I was a victim. Over the next couple of months after it first came out, realizing that I was a victim of sexual abuse by someone who I trusted is a feeling that is impossible to explain…. I have felt sick to my stomach every day since realizing I have been a victim of his over ten times for his own sexual pleasure. It was associated with related harms that overlap with mental health and iatrogenic harms. CB explained: “The injury I suffered in 2000 still gives me pain to this day. That pain is physical, but the information I learned about you in 2017, that information brought new injury, emotional injury.” The next level of harm from moral injury is theorized to impact one’s notion safety in the world and (Brennan, 2021), and many victims (39%) described wide ranging (ripple) effects stemming from realizing they had not been safe with Nassar. For V195, this was an unanticipated price: “I gag at the thought of ever thinking of you as a saint, for sacrificing time for us. It was not a volunteered sacrifice for you. It had a price we were unaware we were paying (V195).” As AE summarized, “I have to rethink everything.” Some described damage that was 129 “damage was internal, unseen” (KS) and corresponding “dehumanization” (MK). For others, they no longer felt safety in their hometowns or cheering for university sports. As AC explained, “I have also lost a lot of the love I had for coming back to my hometown and the passion I always held for being a Spartan.” Reflecting the fact that multiple institutional contexts were implicated in this betrayal, V142 said “I'm so sad that young girls' Olympic success will always be shadowed by the horrific choices you made. I am sad that you have taken away the school I grew up cheering for.” What seemed like havens of safety (hometowns), or success (Olympic success) were no longer safe spaces, even despite this group not describing trust in institutions. The final level implicates a loss of sense of self and lack of control in identity (Brennan, 2021), where victims described that something was deficient in themselves (e.g., questioned their sanity), that innocence was stolen from them, or experienced negative emotions. Their innocence was considered a central part of their identity, where, as AG stated: “Larry stole my childhood, my innocence, my virginity, and my self worth.” This breach impacted how others viewed them and how they viewed themselves and the world. “Larry Nassar, along with stealing my innocence, left me with a complex feeling of being misunderstood; misunderstood by my parents, by my doctors, by my friends. Most of all, I was misunderstood by myself (WB).” A loss of sense of self resulted from feeling robbed of “their wholeness” (V10) and accompanied shame, disgust, self-hatred, self-blame, or that something was wrong with them in concordance with the theorized impacts of moral injury (Brennan, 2021). Multilevel Group Similarly, 68% of victims in the multilevel group indicated damage to identity. Again, there was harm as associated with the first level in realizing they were a victim of sexual abuse by someone they trusted. They did however uniquely describe how this had even broader 130 impacts, where “everything about me changed (IH).” Unlike the interpersonal group, this change to their life trajectory explicitly stemmed from both Nassar and institutions: “When Larry Nassar sexually assaulted me and MSU covered for him, they altered the entire course of my life. From my career path to just the way I navigate through crowded rooms, everything has changed (AT).” However, when it came to the second level of harm (loss of safety), the multilevel trust group implicated institutional responses more overtly and described damaged perceptions of safety that had been found in institutions. In fact, 56% of the victims described damaged views of institutions as tied to loss of safety. Victims uniquely experienced harm in the way abuse was handled by institutions. For example, victims were hurt by institutions not reaching out to or talking with them, perceived non-genuine apologies, or not believing them (e.g., police taking Nassar’s word). As NR summarized: “MSU's response or lack of response has compounded my pain.” Similarly, AR explained about USAG: It's easy to put out statements talking about how athlete care is the highest priority, but they've been saying that for years and all the while this nightmare was happening. False assurances [of safety] from organizations are dangerous, especially when people want so badly to believe them. Finally, victims perceived institutional responses as blaming victims and causing distinct harm (such as slandering victims in the media). For example, LL tapped into how ineffective responses contributed loss of safety with this response: Your broad spectrum of emails that are sent to everyone, not specific survivors, that states that you are sorry and working to fix this, please stop and save yourself the pity party. Does that seriously make you feel better? We don't care. It's too late for your sorrys and trying to make sure that we are safe. 131 While both groups experienced emotional responses tied to loss of self, fear was a more salient threat in the multilevel group. JW emphasized “current and future gymnasts do not deserve to live in anxiety, fear, or be unprotected like I was.” Similarly, fear interfered with their decisions and ability to live their lives as a result of the trust breach: “While I have the same concerns as most parents, I am paralyzed by the fear that I could place my daughter in the hands of someone who could repeatedly molest her under the guise of legitimacy (MS).” For some, institutional responses incited fear and compounded loss of safety. As if dealing with the intense publicity and pain around Nassar wasn't enough, you have added a heartless, depraved level of denial and victim shaming to the mix. You have said terrible things. For example, that victims didn't understand the nuanced difference between sexual abuse and medical treatment and it is virtually impossible to stop a determined sexual predator on your campus. This is disgusting. Do you know how the statements make us feel? Do you understand the impacts of our words? Do you know how these statements put other victims in the state of fear so deep that they may never come forward? (MH) In addition, an institution that doesn’t comply with institutional safeguards that victims had relied upon in willingness to be vulnerable is now viewed as an unsafe one. AA explained that I saw Larry from 2010 to 2016. In 2014 there was a Title IX investigation into the treatment he was performing. During this investigation he was cleared of all wrongdoing and was supposed to have someone in the room during all appointments, and he was supposed to limit skin to skin contact in sensitive areas. Had my parents been informed of this in 2014 I would not have gone back to see him. 132 Ultimately, a betrayal of trust by the institution was devastating to their overall sense of safety: “MSU, you need to realize that you are greatly compounding the damage done to these abuse victims by the way you are responding. This, what it took to get here, what we had to go through for our voices to be heard because of the responses of the adults in authority, has greatly compounded the damage we suffer. And it matters (RD).” A few others (N = 5) specifically described damage to the sport of gymnastics, intersecting with loss of safety in this sport and culture. For CS, this also damaged her identity. As a result, she is “embarrassed to tell people I was a gymnast. I am embarrassed to tell people that I went to Great Lakes and Twistars. Now that doesn't mean what it used to.” As BL said, “To USAG, I'm utterly saddened that defendant has caused the sport of gymnastics to be tied to the weight of sexual abuse and child molestation. I am more disgusted that you knew of this and failed to act when you should have. You have failed to keep the sport of gymnastics safe and with the name it deserves.” Finally, as to the third level of harm, like the interpersonal group, damage to sense of self was tied to having their innocence taken (something viewed as essential to their identity). It had slightly further-reaching impacts for some, including damage to their memories of institutions: “Larry Nassar destroyed my childhood innocence and shattered any positive experiences I had had in the gymnastics world (NW).” A novel theme that emerged in the multilevel conflated loss sense of self with guilt (N = 20). Here, victims described feeling guilty around not protecting others later from experiencing the abuse they experienced, feeling like a “test case” that allowed abuse to continue. As RM explained, 133 It sickens me to know that I was on the early side of his abuse. He used those early appointments with people like my mom and I as a test case to see how far he could push the patient/doctor trust for his own instant gratification, and it makes me feel so disgusted and weak that I was part of his learning process….really guilty for not contributing in some way to stop it sooner. They also grappled with self-blame, where NS wondered “sometimes I feel a sense of guilt because somehow I wonder if it was my fault that all of this happened?” While guilt, blame, or anger for not stopping the betrayal (here, for themselves or others) are associated with moral injury (Brennan, 2021), guilt only emerged in the multilevel trust group. In terms of source, V125 directly links the shame and guilt around misplaced trust in Nassar, but also mentions his institutional roles: Had I not been sucked into your pretend care and pretend love, had I not trusted you, had I actually understood that having my vagina and anus penetrated by a grown man that I loved and trusted when I was 12, 13, 14, and beyond was not okay and that was not your love or your care for me or your duty as my athletic trainer or Michigan State medical school student or Michigan State doctor or Olympic doctor, then I could have spared every girl who came after me the suffering that only someone who has experienced this will understand. Unfortunately, this guilt emerged even when they did report the abuse. Succinctly stated by V55: I did, however, deal with a sense of guilt only due to the fact that I previously thought that if I would have spoken up more or louder then somebody would have listened at Michigan State University. But I did tell a coach while on my gymnastics 134 team at MSU and Kathie chose to put Larry's name, her identity, and the university's reputation above the choice to protect me. Failures of an institution spilled over into their own guilt, perhaps because they identified with the institution and saw it as an extension of themselves, or because they (wrongly) trusted in institutions to protect them or respond appropriately to abuse. Finally, moral injury related to damage to institutional identities (N = 18). For example, victims described harm to their identity as wrapped up in an institution (e.g., alumni, student athlete, USAGymnast, Spartan). Even when the institutional response failed (or harmed) them, they still viewed the institution as a space that should have protected them. They noted that, realistically, if they had spoken up “more” or “louder” it may not have mattered, but they still felt guilt surrounding their own action. Additionally, they described the visceral impact tied to the institution’s breach of trust, which damaged its reputation. For example, AG described her reaction: “If I see anyone with an MSU shirt or hoodie, I get sick. This could have stopped decades ago but you kept doing it and never got caught. It is absolutely sickening.” In fact, 52[ X] characterized her pride as separately harmed: “And last, but certainly not least, my Spartan pride is a victim.” Their positions as alumni, or fans, ex-athletes were all tainted. After being a proud Spartan alum for the past six years, I now feel ashamed to have ever helped a school that will not take accountability. Two of us cannot bring ourselves to hang up our diplomas that we made sacrifices for as student athletes and worked so hard to achieve because of now what it represents. (AM) However, some individuals did reclaim their institutional identity, like EAM: 135 I don't have to be ashamed or anti-MSU. I can be a Spartan if I choose. I can yell go green. Go green. Let me say that again, judge, because I don't think the crowd heard me. Go green. UNIDENTIFIED SPEAKER: Go white. These responses align with literature on loyalty, where stakeholders have varied responses to institutional betrayals of trust including partial identification with the organization (Carvalho & de Oliveira Mota, 2010) Discussion Direct Harms The results provided important insight into emergent vulnerability and decisions within a trust relationship that contribute to our understanding of harms stemming from the trust process. Emergent vulnerability was first tied to direct “iatrogenic harm” that stemmed from Nassar’s treatment. These harms were closely tied to the specific trust relationship and were rooted in victim behaviors within the doctor/patient trust relationship (e.g., reduced monitoring, increased dependence on Nassar to heal their injuries). Once they had established trust with Nassar and the institutions, they took steps that made them additionally vulnerable, 13 such as entering his offices (with or without parental supervision), returning for multiple visits, and enduring painful and abusive procedures. For example, some described how the abuse couched as medical procedure might have raised red flags, but they looked to his behavior and the context of the situation (acting calm, a parent being in the room, the physical space) to continue to trust once 13 An important note is that while emergent vulnerability is tied to “risk taking in a relationship” within the trust process, this language runs the risk of appearing to blame a victim for their behavior (Mayer, Davis, Schoorman, 1995). However, a victim’s decisions in a trust relationship (especially one bolstered by so many reasons to trust both Nassar and the institutions) merely reflects logical decision-making predicted and risk taking is continually lined to trust in the research (Colquitt et al., 2007; Ferrin et al., 2008; Schoorman et al., 2007). Thus, the results add important knowledge to dynamic experiences with trust over the course of a relationship which might relate to distinct harms than those originally contemplated through no blame to the victims. 136 they had entered the trust relationship. The way their behaviors tied to emergent vulnerability align with theory conceptualizing trust as a leap of faith (Möllering, 2001). This leap is central to overcoming vulnerability in deciding whether to enter a trust relationship and is what differentiates trust from mere cooperation (Möllering, 2001). However, understanding a leap of faith outside the initial decision to trust has been under-explored. These results suggest that victims grappled with whether to continue to trust, even when the behavior felt uncomfortable or harmful, thus creating the potential for emergent harm in continually being exposed to abuse, resulting in the myriad emergent physical, mental health, and emotional harms described here. Direct harms were also identified in terms of wide-ranging and serious mental health impacts for victims in both groups that aligned with existing theory and research on betrayal trauma and institutional betrayal (C. P. Smith & Freyd, 2013; Tang & Freyd, 2012). Victims and prior research have identified shame, PTSD, dissociation, suicidality and substance use among victims (Delker & Freyd, 2014b; Edwards, Freyd, Dube, Anda, & Felitti, 2012; Monteith et al., 2016; Platt & Freyd, 2015; C. P. Smith et al., 2016; C. P. Smith & Freyd, 2017). These results both add to that list (e.g., general trauma, discomfort with body, voicelessness, ruining dreams) and more concretely locate the harms as tied to emergent vulnerability in the trust process. The types of direct harms that emerged lead to several policy recommendations. Victims stressed the importance of institutions actually implementing policies intended to protect them and to monitor and punish abusive behavior of institutional actors. They identified multiple institutional levels at which this failed (e.g., coaches, leaders, oversight boards, police). Thus, multilevel sources of harm align with recent research on understanding and improving the climate of institutions at multiple levels to address sexual assault (McMahon, Karp, & Mulhern, 2019; Moylan et al., 2021). However, the unchecked abuse identified here was linked to wide- 137 ranging harms despite mechanisms already being in place (at least in theory) to prevent abuse. So, perhaps current methods of addressing and preventing abuse in institutional settings is not sufficient and a more radical “transformative justice” approach that some feminist scholars are increasingly endorsing (Mendez, 2020). Mechanisms for Understanding Harm Betrayal theory provides a framework to understand emotional harm described by victims stemming from the abuse as breach of trust. Betrayal violates “the sanctity of interpersonal relationships” by causing a “break in the interpersonal frame of reference from which to view the world of interpersonal relationships” (Gartner, 1999, p. 14) and can therefore have wide reaching impacts on future relationships. The harms that both groups of victims described paint a picture of isolation, feeling disempowered, and damage in the very way they view the world and interact with other individuals. There were some important similarities between the groups in harms linked to betrayal. If, at its heart, betrayal impacts relationships, it makes sense that both groups of victims endorsed codes around damaged trust (in men, in romantic partners, in family members, in institutions) and impact on dreams, goals and careers. These results align with research on betrayal trauma which is negatively associated with future trust in another person (Gobin & Freyd, 2014) but provide additional insight into the mechanism. By impacting their worldview, betrayal shaped their trust in themselves and their judgment to enter in relationships. Betrayal for both groups was tied to voicelessness in ways that impacted their relationships with institutions. The multilevel group described emotional impacts from being institutionally disempowered and silenced by numerous institutional actors (e.g., coaches) and institutions (not investigating reports). Victims in both groups identified multiple actors, roles, systems, and physical spaces that were be implicated in institutional trust (and silencing) (Kramer, 1999) and 138 emotional impacts tied to the betrayal map on to research on secondary victimization where there is hurt from being silenced or not believed (Ahrens, 2006; Campbell & Raja, 1999; Roskin- Frazee, 2022). A previously unrecognized impact of this harm for the multilevel group was on their ability to trust institutions. Here, victims described questioning the organizations and medical profession that they originally relied upon in assessing trustworthiness. The impact of institutional betrayal has never been explored on tendency to trust (either individuals or institutions) or on the specific harm in being silenced that stems from trust betrayal, but the results suggest both are important areas for future research. Finally, for the multilevel group, harm occurred when institutional safeguards or institutional actors who were supposed to protect them from failed (e.g., institutional blindness, inadequate apologies, behavior that appeared to put institutional wealth or reputation above the safety of its members). By failure, negligence, or deliberate inaction, institutions were perceived to have not only permitted Nassar to carry out abuse, but in fact created or perpetuated an environment where he could thrive. Institutions breached the psychosocial contract they had with the institution to keep them safe in exchange for their membership (Fitness, 2012; Morrison & Robinson, 1997). These perceived intentional failures in oversight and institutional safeguards that they trusted to keep them safe caused emotional harm. Their accounts don’t only reflect a change in worldview, damage to relationships, and a break in the interpersonal frame of reference, there is a break in the frame of reference about the role and ability to institutions to protect them. Thus, those with multilevel trust experienced multilevel betrayals. The major implication is that institutions that seek to repair relationships with victims must acknowledge multiple levels of betrayal and harm (even, surprisingly, for those who endorsed only interpersonal trust). 139 Moral Injury While the mechanism of betrayal largely stemmed from changes in worldview and impacted relationships (with oneself, with others, with institutions), moral injury related to damage to identity stemming from a breach in recognition trust (Brennan, 2021). Additionally, victims in both groups described guilt in accordance with the theory (where loss of self is associated with guilt, blame, or anger for not stopping a betrayal) (Brennan, 2021, p. 3804). However, here, victims’ guilt was around a perceived inability to stop betrayals of other victims (even when they did report the abuse and were ignored). Both groups also described harm associated with discovering abuse (or shift in identity to “victim”) which challenged the identity they thought they held. Alternatively, some victims described how this identity shift helped them feel validated or “not crazy.” As ISA continues to come to light and victims increasingly learn about the abuse from others who come forward publicly (for example, recent news coverage of abuses at University of Michigan, Ohio State University) advocates should acknowledge the harm that may be associated with viewing oneself as a victim for the first time. So, resources should be tailored to address blame, loss of identity, fear and guilt for possible victims who may come forward after discovering they were abused. Additionally, the therapeutic nature of victim impact statements or litigation bringing similarly-situated victims together can perhaps help assuage the guilt tied to damage to identity that was prevalent here (Parsons & Bergin, 2010; Sheley, 2012). While some harms identified by the multilevel group aligned with behaviors identified by the theory of institutional betrayal (C. P. Smith & Freyd, 2014) (e.g., institutions perceived to create a culture where abuse is common; not taking reports seriously), harm to institutional identity emerged as a previously-unrecognized harm described here. For example, for some, gymnastics was central to their identity, so they felt an acute loss of safety when they discovered the abuse 140 facilitated by institutional inaction of gymnastics clubs or organizations. Similarly, those who had strong institutional loyalty before the abuse might have been uniquely harmed compared to those who did not. The results align with empirical research on loyalty which is closely tied to institutional trust (Carvalho & de Oliveira Mota, 2010), suggesting damage to institutional loyalty and identity may be a distinct harm associated with breach of institutional trust. Additionally, while emotions were prevalent in both groups (including sadness, embarrassment, and anger), damage to identity implicated fear in the multilevel group (but not in the interpersonal trust group). Perhaps fear was related to the loss of safety they experienced in realizing institutions could not (or chose not to) protect them. Importantly, the results here add nuance to the harms associated with IBT by linking it to a mechanism, namely, moral injury. There is precedent for exploring both moral injury and institutional betrayal in the realm of military sexual assault (Holliday & Monteith, 2019; Molendijk et al., 2022; Monteith et al., 2016), but they have not been considered together, making this a logical avenue for future research. These results suggest that linking institutional betrayal responses to moral injury but could be useful in anticipating harms that are tied to damage to identity. In terms of practical implications, currently, the Institutional Betrayal Questionnaire captures the extent to which a victim had an association with an institution. However, perhaps measuring loyalty, trust, or institutional identity might shed additional insight into differential harm experiences. Conclusion This research filled a gap in the literature on harm related to institutional sexual assault by exploring victim perceptions of how trust relationships (whether interpersonal or multilevel) related to distinct harms for victims. Understanding harm that stemmed from those different trust 141 relationships required a closer look at latter stage of the trust process, namely, mechanisms underlying different harms experienced once these groups had established trust. While there were some similarities, results suggested that harm within a trust relationship does differ for victims who described interpersonal versus multilevel trust. Victims in both groups described physical (iatrogenic) and mental health harms, but even the interpersonal group described ripple effects which included institutional damage (e.g., damage to institutional reputation). The mechanisms of betrayal and moral injury provided a framework to understand experienced emotional harm by teasing apart how damage to relationships and identities (respectively) were identified by victims. Multilevel group members identified harms that aligned with institutional betrayal (e.g., not taking reporting seriously, creating a culture where abuse was common), but also identified new harms such as damage to institutional trust, institutional identity, and institutional voicelessness. Resources for victims must account for differences in emotion (e.g., fear, guilt) and the extent to which institutional loyalty might factor into unique harm for those who had experience multilevel trust. 142 CHAPTER 5: DISCUSSION AND IMPLICATIONS Introduction Increasingly, instances of ISA have dominated the news. For instance, ISA has been identified in higher education, religious institutions, policing, and the military (Austen, 2022; Cave, 2022; Gerson, 2022; Svrluga, 2022). The perpetrators range from colleagues and peers to the highest leaders or figureheads, and victims are of varied ages, sexualities, gender identities, and institutional roles. These assaults cause distinct harm to victims compared to sexual assault more generally because trust is at the heart of many of these relationships; but this is harm that we don’t fully understand. Institutional betrayal and betrayal trauma theory theorize that harms are distinct or exacerbated because ISA may implicate different trust relationships: interpersonal and institutional trust(Birrell & Freyd, 2013; C. P. Smith & Freyd, 2013). The research bears out that abuse occurring in these trust relationships has similar wide-reaching impacts as harm stemming directly from sexual assault (e.g., PTSD, suicidality, difficulty trusting), but to fully understand the impact on victims and provide them with resources or avenues to justice, more must be done to explore what trust looks like, distinct vulnerabilities to harms, reasons why one might accept those vulnerabilities, and mechanisms to understand these distinct harms. The first chapter set out the trust process framework to help illuminate harms rooted in trust and the ways they might differ by trustee. While trust is at the core of betrayal trauma and institutional betrayal theories, it isn’t defined or explored in any depth, so I rely on the common conceptualization from trust literature as “the willingness of a party to be vulnerable to the actions of another party” (Mayer et al., 1995). Central to this definition is vulnerability, or the general state of being susceptible to damage or harm, which is theorized to be tied to essential elements of human relationships (Misztal, 2012). Additionally, trustworthiness was identified as 143 central to understanding why one would accept that vulnerability to a trustee, and existing research suggests that both vulnerability and trustworthiness would differ for an interpersonal versus institutional trustee (Bachmann & Inkpen, 2011; Colquitt et al., 2007; Tan & Tan, 2000). The results here discuss how each empirical chapter helped inform understanding of the process of trust presented in Chapter . In presenting this trust process, I engaged in theory building by importing the conceptualization of emergent vulnerability into the trust process. This novel conceptualization of vulnerability accounts for behaviors in trust theory (including risk taking, dependence, decreased monitoring) within a trust relationship and is grounded in ISA research, which suggests that victims might experience abuse and betrayal over time. The overarching framework from this chapter gave reason to think that different harms may be associated with different stages in the trust process, different mechanisms and types of harm may be implicated, and vulnerabilities may be different with and have potential to change or emerge. So, this discussion chapter explores the ways each distinct methodology helps illuminate (or not) the trust process and the ways they help inform our understanding of harm. The second chapter applied a scoping review to explore the ways that the existing literature discussed harms stemming from trust. The trust process shaped the data collection and methodologies to explore (1) how the literature spoke to harms stemming from interpersonal versus institutional trust and (2) how harm was discussed in relation to the stages of trustworthiness, (focal, imposed, and emergent) vulnerability, and (concrete versus amorphous) harms. The results from the scoping review illuminated how this methodology provided insight or completely omitted some pieces of the frameworks in surprising ways. These results provided important insight into differences between concrete versus amorphous harms. Within concrete harms, for both interpersonal and institutional trustees, certain harms emerged (e.g., abuse, health 144 outcomes, exchange, financial) and multiple concrete types of harms could be considered at the same time. Concrete harms were associated with identity and context or situation in ways that has not been explored. On the other hand, amorphous harms also emerged in both types of articles but were more prevalent than concrete harms in the institutional articles. Two particularly relevant themes in interpersonal amorphous harms were emotional and relational harm. Additionally, articles that described amorphous emotional harms uncovered two mechanisms for understanding these types of harm: betrayal and moral injury, and while betrayal emerged for both types of trustee, only moral injury was identified with interpersonal trustees. Finally, concrete and amorphous harms differed in the extent to which they were focal to those doing the trusting, but this concept was under-explored. This adds complexity to the ways vulnerability is typically discussed The third and fourth chapters were designed to explore how victims spoke to harms stemming from interpersonal versus multilevel trust. Chapter 3 explored one stage of the trust process, focusing on vulnerability and the reasons why a victim might accept that vulnerability (trustworthiness) using the VIS from the sentencing trial of Dr. Larry Nassar. The abuse spanned decades and implicated multiple institutions as identified by victims (including educational institutions, medical institutions, and local and national sports institutions). Chapter 4 explored a different stage of the trust process and was designed to delve directly into the harms identified by victims once they entered a trust relationship. The two mechanisms identified in the scoping review (betrayal and moral injury) also emerged in the VIS chapter for emergent vulnerability and harms identified by victims. They identified physical health harms (including iatrogenic harms) and a range of serious mental health harms (e.g., PTSD, depression, suicide ideation). Additionally, they identified damage to a range of 145 relationships with families, intimate partners, and even their dreams and goals (through the mechanism of betrayal) as well as damage to their identities as innocent, as non-victims, and damage to the trust they had in themselves (through the mechanism of moral injury). Ultimately, the VIS results were a more useful methodology to illuminate differences in imposed vulnerability, emergent vulnerability, and trustworthiness in important ways that differed by group. While each individual methodology did not fill all the gaps in the larger picture, together they help provide important insight into how and when harm might differ within the trust process. For example, the amorphous harms in the scoping review were not identified in the literature as emergent harms but were identified as such by the victims in the VIS. This supports the choice to use two different methodologies to answer questions around how harm might stem from different trust relationships. The overall results indicated that that victim needs could differ as a function of different trustee, different stage of the trust process, and different harms experienced. Thus, the final portion of the discussion chapter explores insights and implications from the results across the dissertation. When we better understand the harm and sources of harm victims experience, we can better prevent, educate, and remedy ISA and serve the growing numbers of victims that are coming forward. However, the results also suggest that much work is left to be done to help clarify where harms might fit in this framework, with specific attention to theoretically and empirically building on different vulnerabilities and harms that might map on to the trust process. 146 Insights from Empirical Chapters The empirical chapters provide insight into harms that could stem from the trust process. As a reminder, the trust process suggested that different harms may be associated with different stages in the trust process, different mechanisms may be implicated, and vulnerabilities may be different to begin and may change. Vulnerability A first takeaway has to do with insights into ways to conceptualize vulnerability throughout the trust process. The scoping review provided some limited insight into imposed vulnerability. The results suggested that intersectional identity is important in exploring how people understand their vulnerability (e.g., female sex workers, relationship status as tied to negative health outcomes). The chapters engaging with victims’ accounts provided more insight into imposed vulnerability and flagged the possible intersecting vulnerabilities a victim could experience (e.g., age, innocence). Both the results of both the scoping review and victims’ accounts added nuance to imposed vulnerability by also considering the specific context of the trust relationship (e.g., risky professions, family situations), Focal vulnerability was more difficult to disentangle from imposed and emergent vulnerability in both the scoping review and victims’ accounts. In the scoping review, some articles fell into a theme of exchange, which suggested that vulnerability to certain specific harms were and explicitly accepted in return for some benefit within the relationship. For example, vulnerability could relate to the misuse of evidence of some past deviance that was offered to gain trust and entrance to a group of peer adolescent deviants. Alternatively, a theme arose around injuries to things we value that could have been at the forefront of a trustor’s mind in thinking about a given trust relationship (e.g., injuries to life, health, reputation, shared goods). 147 These results suggest that focal vulnerability could be very context specific and could implicate a variety of harms. Alternatively, results from victims’ accounts suggested that, for some, sexual assault was specifically not contemplated or considered focal in deciding to trust a person or an institution (e.g, not realizing a doctor could perpetrate such an action). Adding complexity, the harm that appeared to be focal for victims in entering a relationship of trust with a doctor was captured in the results as direct iatrogenic harm (harm stemming from Nassar’s treatments). Some type of harm that could stem from medical treatment is arguably at the forefront of any patient’s mind when they trust a physician to perform treatment—they accept some risk that the treatment will not work or could exacerbate existing injuries. However, this was not clearly explained by victims, and the extent to which this harm was emergent or focal is certainly worthy of future research. Trustworthiness The second takeaway has to do with the role of trustworthiness, where the trust process indicated that trustworthiness perceptions help inform a person’s willingness to accept their vulnerability to another. However, the link between trustworthiness and harm was somewhat surprisingly not engaged with an in any depth by the articles in the scoping review. The concepts of vulnerability and trustworthiness both shaped the search strategy (both terms were used in searches) so it was expected that results would fall into those categories and help illuminate harm at different stages of the trust process, but that was not the case. Instead, articles that grappled with trustworthiness in the search were not linked to harm, suggesting that it is an overlooked step of the process in existing research that explores trust-based harm. As opposed to the scoping review results, the victims did find trustworthiness relevant to harm, in fact, both groups identified many multilevel and overlapping reasons to trust both 148 Nassar and relevant institution(s). Despite the groups being divided into interpersonal versus multilevel trust, both groups identified the importance of institutional context in shaping trustworthiness. Both groups identified how trustworthiness was manipulatively signaled by Nassar, including leveraging his institutional affiliations to signal ability and benevolence. Similarly, both groups also described the important role of physical space, suggesting that when physical space is in an institutional setting it can be leveraged by perpetrators to signal trustworthiness. This has important policy implications that draw from Hirsch and Kahn’s (2019) work on sexual geographies, namely, by flagging how institutionally owned space can be used manipulatively by abusers to signal trustworthiness, even if it is not in the institution’s full control (e.g., Nassar decorated the walls with photos of athletes). Importantly, victims wanted the institutions held accountable for the way their physical space was abused, suggesting that increased liability related to space might be an avenue for policy change grounded in the harms tied to trustworthiness victims identified here. Trustworthiness’ emergence as a multilevel construct and implicating institutional space for both groups makes sense as institutional sexual assault doesn’t occur in a vacuum, but necessarily implicates institutional context(s). This gives insight into multilevel grooming and also supports the trend in trust research that explores institutional trust as multilevel trust or, at the very least, differentiates between type of trustee in an institutional context (e.g., supervisor, co-worker, institution as a whole) (Fulmer, 2018; Kramer, 1999; Tan & Tan, 2000). Emergent Harm The results of the scoping review did not provide explicit insight into emergent harm (as compared to vulnerability to that harm). The results did suggest that more research contemplated harms that stemmed from interpersonal trust than institutional trust. Additionally, harms in both 149 groups fell into categories of concrete versus amorphous harms. Within amorphous emotional harms, however, the scoping review identified two mechanisms that help understand this harm: betrayal and moral injury. These emerged differently by type of trustee, where moral injury only emerged in interpersonal trust as a salient mechanism. While this type of harm was not discussed specifically as emergent harm, the results provided some important initial insight into the different mechanisms that might be tied to different type of harms stemming from trust. In the victims’ accounts, emergent vulnerability differed from imposed vulnerability (e.g., age, innocence, situation). For example, both groups identified some novel types of emergent vulnerability that were closely tied to the relationship (doctor/patient) and institutional context (institutions didn’t take reports seriously, put money over individuals). Emergent vulnerability for victims was also tied to emotional harm, mapping on to betrayal and moral injury as mechanisms. Interestingly, these did not differ in important ways for the interpersonal versus multilevel group, even though it did differ in the scoping review. This may be tied to the nature of the data, where solely institutional trust was impossible to disentangle in this sample of victims, so when moral injury was identified in the multilevel group it may have stemmed from interpersonal trust (because that type of trust could have been present as well). Future research that explicitly tries to disentangle institutional trust from interpersonal trust in a sample of ISA victims can provide essential insight into differences in mechanisms for harm. Trust Relationship The final takeaway relates to the differences in harm linked to the trustee, or target of trust. Prior research flagged important differences between interpersonal, institutional, and multilevel trustees in terms of vulnerability, trustworthiness perceptions, and type of harm that may stem from the relationship. the results raised an issue of whether institutional trust and harm 150 can be measured independently from interpersonal trust and harm in the context of ISA. On one hand, results of the scoping review presented instances where institutional trust and betrayal had been successfully brought outside of the sexual assault (and higher education) context in which it was originally introduced. These articles identified betrayal as the mechanism to explain amorphous harm stemming specifically from institutional trust, successfully examining institutional trust and harm as a separate concept. These results therefore provide some evidence that institution-based trust betrayals can be distinctly measured and linked to harms (e.g., ethical distress in nursing). However, these articles did suggest that even this harm might implicate multiple trustees (e.g., physicians, doctors in general, or healthcare organizations). In victims’ accounts, it was impossible to disassociate trust in an individual perpetrator from trust in an institution; even for the group that only endorsed interpersonal trust, institutional context mattered. This insight into the complicated nature of trustees implicated in ISA ultimately challenged the proposed flow of the trust process. Victims’ accounts suggested that the role of third parties complicates our understanding of trust stemming from one trustee to one trustor. Instead, for victims, the perpetrator might stand as a representative of certain institutions because of their position of power, celebrity, or authority in those institutions. Alternatively, victims described some level of role-based trust that was innately tied to an institution (e.g., an employer, a system of training). A final complication that arose was that victims trust may have been filtered through trust in their parents or coaches, especially if they were minors when they sought treatment. For example, a victim might have trusted a parent to find a doctor, seek a referral, or make an appointment on their behalf. This means that victims’ perceptions were in part based on a parent’s assessment of trustworthiness of the person or institution. In other words, victims might have assessed the trustworthiness of the trustee(s) through the lens of their 151 parents’ trust (i.e., my trusted parent trusts this person). These examples suggest that while victims described trustworthiness and trust in their accounts, their agency and the process by which they actually entered a trust relationship might have been more complicated than current trust process acknowledges, which posits that it only flows from one trustee to one trustor. The implications of identifying these complex trustees that interfered with the theorized flow of the trust process suggests that multilevel trust is likely the best direction in which to take research on harm stemming from trust in the context of ISA. In the scoping review there were very few institutional trust related harms, and they primarily mapped on to amorphous harms. However, a theme around multilevel harm emerged, despite the restrictions on search strategy limiting the literature to interpersonal and institutional trust. Again, this aligns with the trend in trust literature suggesting that trust is more complex and multilevel in institutional settings than perhaps previously considered (Bachmann, 2011; Lumineau & Schilke, 2018). Similarly, victims were split into interpersonal and multilevel groups, as shaped both by the sample and the nature of ISA (where institutional context and an interpersonal abuser were both relevant). However, there were qualitative differences in the experiences of vulnerability, trustworthiness, and harms experienced by the interpersonal versus the multilevel group. Individuals in the multilevel group were very clear in explaining why and how they attributed harm to institutional trustees. They experienced a wide range of harms from failures of institutions, being silenced, and disempowered. Both groups also identified a wide array of very serious mental health impacts, that impeded their ability to sleep, have relationships, keep jobs, and function which add to the growing research on both institutional and interpersonal betrayal (Delker & Freyd, 2014b; Edwards et al., 2012; Monteith et al., 2016; Platt & Freyd, 2015; C. P. Smith et al., 2016; C. P. Smith & Freyd, 2017). However, the results suggest that multilevel 152 betrayal should be more explicitly contemplated, and future research should consider multilevel harms that emerge. Conclusion ISA is widespread and currently linked to numerous relationships and harms experienced by victims. The dissertation sought to contribute to the understanding of harm experienced by victims of ISA using two distinct data sources and methodologies to tease apart the trust process and exploring different trust relationships. The scoping review brought together strains of literature that have not been considered together in any meaningful way before (trust, harm, vulnerability) to aid in understanding harm. The victims’ accounts explored how descriptions of vulnerability and trustworthiness of different targets differentially facilitated harm in their experiences of institutional sexual assault. The distinct methodologies allowed for a novel comparison of the stages of the trust process and harms that stemmed from different relationships as identified by victims and the literature. Much more is needed in research that acknowledges the limits on our current understanding of harms that stem from the trust process, especially as it relates to the complex trustees implicated in ISA. 153 APPENDICES 154 APPENDIX A: Figures for Chapter 1 Figure 1: A Visual Representation of Harm within Institutional Sexual Assault Experiences Institutional Sexual Assault Actor Individuals Institutions Violation of Violation of Mechanism Sexual Assault Interpersonal Trust Institutional Trust Harm Direct Harm Betrayal Trauma Institutional Betrayal Figure 2: Trust Process Imposed Emergent Focal vulnerablity Trustworthiness Trust Vulnerability Vulnerablity Potential for Harm 155 APPENDIX B: Figures and Tables for Chapter 2 Figure 3: PRISMA Flowchart for Scoping Review Trust and Harm 156 Figure 4: Database Search for Scoping Review Database 1: Psychinfo Incl Psych Articles Date: 8/19/21 Keyword/control language: Trust(social behavior) Final search: (noft(trust) OR MAINSUBJECT.EXACT("Trust (Social Behavior)")) AND noft(Harm OR damage OR risk OR "potential negative outcome" OR "adverse effect" OR "adversely effect") AND (noft(trustworth*) OR noft(vulnerab*)) Search # Keyword or Control Language # Results 1 Trust 69,011 2 MAINSUBJECT.EXACT("Trust 12,923 (Social Behavior)") 3 1 OR 2 69,011 4 Noft(Harm OR damage OR risk or 696,371 “potential negative outcome” or “adverse effect” or “adversely effect”) 5 Trustworth* OR vulnerab* 86,380 6 3 and 4 and 5 964 Database 2: Sociological Abstracts Date:8/19/21 Keyword/control language: MAINSUBJECT.EXACT("Trust") *in NOFT Search for harm in thesaurus: not relevant FINAL SEARCH: (noft(trust) OR MAINSUBJECT.EXACT("Trust")) AND noft(harm OR damage OR risk OR "potential negative outcome" OR "adverse effect" OR "adversely affect") AND (noft(trustworth*) OR noft(vulnerab*)) 157 Figure 4 (cont’d) Search # Keyword or Control Language # Results 1 Noft(Trust) 17,239 2 MAINSUBJECT.EXACT("Trust") 6,881 3 1 OR 2 17,239 4 noft(harm OR damage OR risk OR 91,434 "potential negative outcome" OR "adverse effect" OR "adversely affect") 5 noft(trustworth* or vulnerab*) 23,233 6 3 AND 4 AND 5 190 Database 3: Criminal Justice Database Date: 8/19/21 Keyword/control language: MAINSUBJECT.EXACT("Trust") Final Search: (noft(trust) OR MAINSUBJECT.EXACT("Trust")) AND noft(harm OR damage OR risk OR "potential negative outcome" OR "adverse effect" OR "adversely affect") AND (noft(trustworth*) OR noft(vulnerab*)) Search # Keyword or Control Language # Results 1 noft(trust) 7, 395 2 MAINSUBJECT.EXACT("Trust") 1,054 3 1 OR 2 7,395 4 noft(harm OR damage OR risk OR 90,687 "potential negative outcome" OR "adverse effect" OR "adversely affect") 5 noft(trustworth* OR vulnerab*) 10,320 6 3 AND 4 AND 5 99 158 Figure 4 (cont’d) Database 4: Web of Science Core Collection Steps: https://www-webofscience-com.proxy1.cl.msu.edu/wos/woscc/basic-search àselect web of science core collection Date: 8/19/21 Keyword/control language : Trust (no control language) Search # Keyword or Control # Results Language 1 Trust (title) or trust (abstract) 149,675 2 Trustworth*(title) or 20,267 trustworth* (abstract) 3 Vulnerab* (title) or vulnerab* 299,621 (abstract) 4 2 OR 3 319,206 5 harm OR damage OR risk OR 4,034,309 "potential negative outcome" OR "adverse effect" OR "adversely affect" (title) or harm OR damage OR risk OR "potential negative outcome" OR "adverse effect" OR "adversely affect" (abstract) 6 #1 AND #4 AND #5 1,706 Database 5: Proquest Dissertation/Theses Date: 8/19/21 Keyword: trust (no control language) Final search: noft(trust) AND noft(harm OR damage OR risk OR "potential negative outcome" OR "adverse effect" OR "adversely affect") AND noft(trustworth* OR vulnerab*) Search # Keyword or Control # Results Language 1 noft(trust) 30,677 159 Figure 4 (cont’d) 2 noft(harm OR damage OR 297,209 risk OR "potential negative outcome" OR "adverse effect" OR "adversely affect") 3 noft(trustworth* OR 42,970 vulnerab*) 4 1 AND 2 AND 3 456 160 Figure 5: Template for Extraction Review General information Study ID: (First author, year) Title: Title of paper / abstract / report that data are extracted from Authors: list all authors Year: publication year Country in which the study conducted United States UK Canada Australia Other Trustor Individual Institution Other Trustee Individual Institution Other Specific Trustee Interaction with harm Type of Harm Emotional Harm/Vulnerability Exploitation Theoretical Health outcome self/others Abuse Financial Harm Fraud Security/Privacy Institution Characteristics of included studies Methods: (qualitative, quantitative, mixed, theoretical) Aim of study: (authors’ stated aim) Study design 161 Figure 5 (cont’d) Randomised controlled trial Non-randomised experimental study Cross sectional study Systematic review Qualitative research Theoretical discussion Economic evaluation Other Participants Population description (characteristics of population) Total number of participants (where relevant) Trust Measure (if not a quantitative study then which definition(s) of trust do they rely on) Harm Measure (or, how harm is defined/discussed) Relationship of harm and trust (how are they measured or discussed in relation to one another) Conclusion (overall summary of the findings and takeaways) Other notes 162 Table 1: Understanding Harm Stemming from Trust Relationships Interpersonal Multilevel Institutional Amorphous Amorphous Interpersonal Amorphous Multilevel Amorphous Institutional Harm Concrete Concrete Interpersonal Concrete Multilevel Concrete Institutional Harm 163 Table 2: Studies Included in Scoping Review Author Name(s) Title Study Setting Specific Trustee Study and Year Design Ahmad, Ferlie, How Trustworthiness is Pakistan Healthcare provider Qualitative & Atun, 2013 Assessed in Health Care: research A Sensemaking Perspective Amundsen, Hetero-sexting as UK Male partners/sexual interests Qualitative 2020 mediated intimacy work: research “Putting something on the line” Bachmann & Can trust flourish where NA Two-tiered governance Mixed Hanappi- institutional distrust systems of incorporated Theory Egger, 2014 reigns? companies in Austria and AND Germany empirical Baier, 1986 Trust and Antitrust United States Many others (intimates, Theoretical strangers, enemies) discussion P.E. Bailey et Age-related Similarities Australia Trustworthy/untrustworthy Quantitative al., 2017 and Differences in First individual methods Impressions of Trustworthiness Bailey & Agency, lapse in condom Jamaica Client Qualitative Figueroa, 2017 use and relationship research intimacy among female sex workers in Jamaica Bearman & Intellectual streaking: The Australia Student Theoretical Molloy, 2017 value of teachers exposing discussion minds (and hearts) Bernstein, 2011 Trust: On the Real but NA Another person Theoretical Almost Always discussion Unnoticed, Ever- Changing Foundation of Ethical Life Bhana & Pillay, Negotiating femininities South Africa Romantic partner Qualitative 2018 on campus: Sexuality, research gender and risk in an HIV environment Bianchi, 2015 Willingness to believe and NA Seller/buyer Theoretical betrayal aversion: the discussion special role of trust in art exchanges Brennan, 2021 Recognition trust NA NA Theoretical discussion Brewer, 2021 Institutional betrayal in United States Nurse and employer Systematic nursing: A concept review analysis Brockway, 1993 Limited Paternalism and United States Salesperson Theoretical the Salesperson: A discussion Reconsideration 164 Table 2 (cont’d) Buchman & What’s trust got to do North America Doctor/patient Other: Ho, 2014 with it? Revisiting Literature opioid contracts Review Butler The Right Amount of Multinational "Most people' Quantitative Giuliano, & Trust methods Guiso, 2016 Carr & Developing the United States Patient/analyst (and vice Qualitative Sandmeyer, vicissitudes of safety and versa) research 2018 danger in psychoanalysis: developing trust through mutual engagement Castle et al., Neural and behavioral United States Another person Quantitative 2012 bases of age differences in methods perceptions of trust Chan, 2009 “Why Did You Hurt NA Another person Theoretical Me?” Victim’s discussion Interpersonal Betrayal Attribution and Trust Implications Chen et al., To Disclose or To Falsify: United States Hospitality organization Mixed 2021 The Effects of Cognitive Methods Trust and Affective Trust on Customer Cooperation in Contact Tracing Damodaran Trust and risk pitfalls in Australia Clinical teachers trusting Qualitative Jones, & medical education: A students research Shulruf, 2021 qualitative study of clinical teachers DeAguiar & Romantic Relationships, Brazil Partner Quantitative Camargo, 2014 Adolescence and HIV: methods Love as an Element of Vulnerability Dupont & The Trust Triangle: Laws, United States Multi-dimensional (personal, Theoretical Karpoff, 2020 Reputation, and Culture in relational, institutional) discussion Empirical Finance Research Flashman & Thick as thieves: United States Other deviants Theoretical Gambetta, 2014 Homophily and trust discussion among deviants French Hanna, “If I Die, They Do Not UK Individual and/or institution Qualitative & Huckle, 2022 Care”: U.K. National within NHS research Health Service Staff Experiences of Betrayal- Based Moral Injury During COVID-19 Gobin, 2012 Trauma, Trust and United States Trusted person who Quantitative Betrayal Awareness perpetrated the betrayal methods 165 Table 2 (cont’d) Gollwitzer, When The Need to Trust NA NA Theoretical Rothmund, & Results in Unethical discussion De Cremer, Behavior: The 2009 Sensitivity to Mean Intentions (SeMI) Mode Gu, Liu, & Locality Stereotype, CEO Other: China CEO Quantitative Peng, 2022 Trustworthiness and methods Stock Price Crash Risk: Evidence from China Hamm, Searle, Public Vulnerability to the United States Police Quantitative Carr, & Rivers, Police: A Quantitative methods 2021) Inquiry Han & Elder abuse in the Covid- United States "Trusted other", caregiver, Theoretical Mosqueda, 19 era family, neighbor, financial discussion 2020 advisor Houde, The Four Faces of Trust: United States NA Mixed Sherman, An Empirical Study of the methods White, & Nature Sheppard, of Trust in Relational 2004) Forms Huddy, 2016 Vulnerability in the United States Business faculty Mixed classroom: How methods undergraduate business instructors' ability to build trust impacts the students' learning experience Judges, Role of cognition, Canada Scammers Quantitative Gallant, Yang, personality, and trust in methods & Lee, 2017 fraud victimization in older adults Kerler & The Effects of United States "Auditee" Quantitative Killough, 2008 Satisfaction with a methods Client’s Management During a Prior Audit Engagement, Trust, and Moral Reasoning on Auditors’ Perceived Risk of Management Fraud Kirsner, 2011 Trust and the global NA Banking sector Theoretical financial crisis discussion Koehler & Betrayal aversion: When United States Asymmetric protective trusts. Quantitative Gershoff, 2003 agents of protection methods become agents of harm Kraft et al., beyond consent: building United States Research institutions Qualitative 2018 trusting relationships with research diverse populations in precision medicine research 166 Table 2 (cont’d) Kramer, 2012 Moving between United States Student or patient Qualitative laboratory and fi eld: a research multi- method approach for studying trust judgments Kutsyuruba & The Destructive Effects of United States Leaders Theoretical Walker, 2016 Distrust: Leaders and discussion Brokers of Trust in Organizations Landa, Zhou, Interrogating the Role of Zimbabwe Clergy member Qualitative & Tshotsho, Language in Clergy research 2019 Sexual Abuse of Women and Girls in Zimbabwe Luchters et al., The contribution of Kenya Emotional partner, client Quantitative 2013 emotional partners to methods sexual risk taking and violence among female sex workers in Mombasa, Kenya: A cohort study Maia, Guilhem, Vulnerability to Other: Brazil Common law partner Quantitative & Freitas, HIV/AIDS in married methods 2008) heterosexual people or people in a common-law marriage Novack, 2018 The Analyst's experience United States Patient Theoretical of trust and mistrust discussion Palmer, 2005 Mummers and Moshers: Newfoundland Other participants of Theoretical Two Rituals of Trust in mumming or moshing discussion Changing Social Environments Pratt, Lepisto, The Hidden Side of Trust: United States Other firefighters mixed Dane, & Jones, Supporting and Sustaining method 2019) Leaps of Faith among Firefighters Rotenberg, The conceptualization of UK Parent Theoretical 2010 interpersonal trust: A discussion basis, domain, and target framework Sapiro, 2020 Assessing trustworthiness: United States Mental Health professional Qualitative Marginalized youth and research the central relational paradox in treatment Smith & Freyd, Dangerous Safe Havens: United States Institution Quantitative 2013 Institutional Betrayal methods Exacerbates Sexual Trauma Smith & Freyd, Institutional Betrayal United States Institution Theoretical 2014 discussion 167 Table 2 (cont’d) Smith. 2016 First, Do No Harm: United States Individual physician and in Quantitative Institutional Betrayal in a larger healthcare methods Healthcare institution Wolfensberger, A conceptual analysis of UK Medicine Theoretical 2016 trust in medicine: its discussion definition, decline, and significance 168 Table 3: Themes for Scoping Review Articles Type of Trustee Meta theme Theme Sub Citation Theme Interpersonal Concrete Abuse Abusive • (Amundsen, 2022) (N = 23) Exchange • (Flashman & Gambetta, 2014) • (Palmer, 2005) Physical • (Landa et al., 2019) risk • (Pratt et al., 2019) Health • (Ahmad, Ferlie, & Atun, 2013) • (A. E. Bailey & Figueroa, 2018) • (Bhana & Pillay, 2018) • (De Aguiar & Camargo, 2014) • (Luchters et al., 2013) • (Maia et al., 2008) Financial • (. E. Bailey et al., 2016) • (Brockway, 1993) • (Butler et al., 2016) • (Castle et al., 2012) • (Gu et al., 2022) • (Judges et al., 2017) Multiple Multiple • (M. L. E. Chan, 2009) harms • (Damodaran et al., 2021) • (Han & Mosqueda, 2020) • (Houde, Sherman, White, & Sheppard, 2004) Being • (Baier, 1986) exploited • (Gollwitzer, Rothmund, & De Cremer, 2009) Amorphous Emotional Betrayal • (Bearman & Molloy, 2017) (N = 14) • (Gobin, 2012) • (Huddy, 2015) • (Koehler & Gershoff, 2003) Moral • (Brennan, 2021) injury • (Bernstein, 2011) • (Kutsyuruba & Walker, 2016) • (Rotenberg, 2010) Professional • (Kramer, 2011) • (Kerler & Killough, 2009) Security • (Chen et al., 2021) /Privacy • (Sapiro, 2020) Relational • (E. M. Carr & Sandmeyer, 2018) • (Buchman & Ho, 2014) • (Bianchi, 2015) • (Novack, 2018) Multilevel (N = 2) Interpersonal • (Hamm et al., 2021a) & • (Wolfensberger, 2016) Institutional 169 Table 3 (cont’d) Institutional Concrete Financial • (Kirsner, 2011) (N = 3) • (Dupont & Karpoff, 2020) • (Bachmann & Hanappi-Egger, 2012) Amorphous Institutional • (Brewer, 2021) (N = 6) Betrayal • (French et al., 2022) • (Kraft et al., 2018) • (C. P. Smith & Freyd, 2013) • (C. P. Smith & Freyd, 2014) • (Carly P. Smith, 2016) 170 APPENDIX C: Figures and Tables for Chapter 3 171 Table 4: Demographic Characteristics Demographic Characteristics Study Sample (n=127) Sex Female 100% (n=127) Male 0% (n=0) Current Age Minor 7% (n=9) Adult 92% (n=118) Interaction with Nassar Gymnastics 48% (n=62) Cross Country/Track 3% (n=5) Dance 4% (n=6) Other14 42% (n=54) 14 Where fewer than 2 participants endorsed a certain type of interaction with Nassar or if they did not describe a reason for their initial interaction it was included in other (e.g., soccer, volleyball, family friend, general pain, car accident, referral from pediatrician). 172 Table 5: An Example of the Three-Tiered Coding Framework Used for Vulnerability Interpersonal Trust Multilevel Trust (1) VULNERABILITY (1) VULNERABILITY **Imposed vulnerability **Imposed vulnerability - Vulnerable situation -“At a very vulnerable time in our lives” - Young, innocent, naïve - Young, innocent, naïve - Desperation -“Only person who could help” - Because he was the adult - Didn’t know much about medicine -Uncomfortable with body - Culture of gymnastics - John Geddart Key: Different stages of the qualitative study are indicated by different symbols (1) THEORETICAL CONSTRUCT: an abstract idea that arose from groups of themes **Theme: an organizing concept that built on repeating ideas - Repeating idea: similar words or phrases used by participants, often in participant’s exact words 173 Table 6: Theoretical Constructs, Themes, Codes for Vulnerability Differences between groups highlighted in italics Interpersonal Multilevel THEORETICAL Theme Repeating Idea THEORETICAL Theme Repeating Idea CONSTRUCT CONSTRUCT VULNERABILITY Imposed ● Vulnerable VULNERABILIT Imposed ● “At a very vulnerability situation Y vulnerability vulnerable ● Young, time in our innocent, lives” naïve ● Young, ● Desperation innocent, ● Because he naïve was the adult ● “Only person who could help” ● Didn’t know much about medicine ● Culture of gymnastics ● John Geddart Institutional ● Because parent Institutional ● Institutional and non- was in the safeguards safety institutional room ● “Raised to safeguards ● Position of respect trust and authority” power ● Elements of ● Elements of physical physical space space 174 Table 7: Theoretical Constructs, Themes, Codes for Trustworthiness Differences between groups highlighted in italics Interpersonal Multilevel THEORETICAL Theme Repeating Idea THEORETICAL Theme Repeating Idea CONSTRUCT CONSTRUCT GROOMING Benevolence ● “Doing what GROOMING Benevolence ● “My friend, was best for my doctor, me” someone I ● Trusted friend thought ● Actions to gain cared about trust me” ● Behavior was ● Gifts normalized as ● Actions to non abuse gain trust ● Others experienced the same thing INSTITUTIONAL Ability ● Because he INSTITUTIONAL Reputation in ● “A man I CONTEXT was a doctor LOYALTY context was told ● Referrals deserved my trust” ● Others trusted him Reputational ● Reputation ● “So well trust ● “The best trusted and gymnastics so well doctor” known” ● Had trust of my ● Gymnastics family doctor ● Association with an institution 175 APPENDIX D: Figures and Tables for Chapter 4 176 Table 8: An Example of the Three-Tiered Coding Framework Used for Harm Interpersonal Trust Multilevel (1) DIRECT HARMS (1) DIRECT HARMS ** Health Impacts **Health Impacts- - PTSD -PTSD - Mental Health -Mental health - Suicide -Suicide - Iatrogenic harm -Iatrogenic harm -Uncomfortable with body -Embarrassing -General trauma Key: Different stages of the qualitative study are indicated by different symbols (1) Theoretical construct: an abstract idea that arose from groups of themes **Theme: an organizing concept that built on repeating ideas - Repeating idea: similar words or phrases used by participants, often in participant’s exact words 177 Table 9: Theoretical Constructs, Themes, Codes for Direct Harm Differences between groups highlighted in italics Interpersonal Institution DIRECT Health impacts • Iatrogenic DIRECT Health impacts • Iatrogenic harm HARMS harm HARMS • PTSD • PTSD • Mental Health • Mental • Suicide health • Uncomfortable • Suicide with body • Embarrassing • General Trauma 178 Table 10: Theoretical Constructs, Themes, Codes for Betrayal Differences between groups highlighted in italics Interpersonal Institution SEXUAL Voicelessness • Took my BETRAYAL IN Voicelessness • Took my voice BETRAYAL voice away CONTEXT away • Harm to • Betrayal credibility Impact on • Damaged Impact on • “Afraid to trust” future and trust future and • Ruining sports, relationships • Damaged relationships careers, dreams dreams/goals • Impact on • Damage to family family • Struggle with • Impact on intimacy intimate • Can’t be around relationships men • Discomfort around men Institutionally • Institutional disempowered voicelessness • Institutions turned a blind eye • Institutional enablers • Organization put themselves over safety of members 179 Table 11: Theoretical Constructs, Themes, Codes for Moral Injury Differences between groups highlighted in italics Interpersonal Institution MORAL Identity • Something is MORAL Identity • “Changed who I INJURY wrong with INJURY was as a person” me • Robbed of • Stolen innocence Innocence • Realizing was • Realizing I abused was a victim • Institutional of sexual Identity abuse • Fear • Sanity • Negative Emotions Ripple Effects • Damage to Damaged view • Institutional institutional of institutions response reputation • Failure of • Amorphous leadership damage/harm • Damage to gymnastics world • Guilt 180 APPENDIX E: Figures for Chapter 5 Figure 8: Trust Process Imposed Emergent Focal vulnerablity Trustworthiness Trust Vulnerability Vulnerablity Potential for Harm 181 REFERENCES 182 REFERENCES Abrahams, N., Devries, K., Watts, C., Pallitto, C., Petzold, M., Shamu, S., & García-Moreno, C. (2014). Worldwide prevalence of non-partner sexual violence: A systematic review. The Lancet, 383(9929), 1648–1654. https://doi.org/10.1016/S0140-6736(13)62243-6. Abrams, J. R., & Potts, A. (2020). The language of harm: What the nassar victim impact statements reveal about abuse and accountability. University of Pittsburgh Law Review, 82(1), 71–134. https://doi.org/10.5195/LAWREVIEW.2020.775 Ahmad, R., Ferlie, E., & Atun, R. (2013). How Trustworthiness is Assessed in Health Care: A Sensemaking Perspective. Journal of Change Management, 13(2), 159–178. https://doi.org/10.1080/14697017.2012.700525 Ahrens, C. E. (2006). Being silenced: The impact of negative social reactions on the disclosure of rape. American Journal of Community Psychology, 38(3–4), 263–274. https://doi.org/10.1007/s10464-006-9069-9 Amundsen, R. (2022). Hetero-sexting as mediated intimacy work: ‘Putting something on the line.’ New Media and Society, 24(1), 122–137. https://doi.org/10.1177/1461444820962452 Anderson, S., Allen, P., Peckham, S., & Goodwin, N. (2008, July 9). Asking the right questions: Scoping studies in the commissioning of research on the organisation and delivery of health services. Health Research Policy and Systems, Vol. 6, p. 7. https://doi.org/10.1186/1478- 4505-6-7 Arksey, H., & O’Malley, L. (2005). Scoping studies: Towards a methodological framework. International Journal of Social Research Methodology: Theory and Practice, 8(1), 19–32. https://doi.org/10.1080/1364557032000119616 Armstrong, E. A., Hamilton, L., & Sweeney, B. (2006). Sexual assault on campus: A multilevel, integrative approach to party rape. Social Problems, 53(4), 483–499. https://doi.org/10.1525/sp.2006.53.4.483 Auerbach, C. F., & Silverstein, L. B. (2003). Qualitative Data: An Introduction to Coding and Analysis (Qualitative Studies in Psychology). In Qualitative data An introduction to coding and analysis. NYU Press. Austen, I. (2022, May 30). Canada’s Military, Where Sexual Misconduct Went to the Top, Looks for a New Path. The New York Times. Retrieved from https://www.nytimes.com/2022/05/30/canada-military-sexual-misconduct.html Babcock, R. L., & DePrince, A. P. (2012). Childhood Betrayal Trauma and Self-Blame Appraisals Among Survivors of Intimate Partner Abuse. Journal of Trauma and 183 Dissociation, 13(5), 526–538. https://doi.org/10.1080/15299732.2012.694842 Bachmann, R. (2001). Trust, power and control in trans-organizational relations. Organization Studies, 22(2), 337–365. https://doi.org/10.1177/0170840601222007 Bachmann, R. (2011). At the crossroads: Future directions in trust research. Journal of Trust Research, 1(2), 203–213. https://doi.org/10.1080/21515581.2011.603513 Bachmann, R. (2018). Institutions and Trust. In The Routledge Companion to Trust (pp. 218– 227). https://doi.org/10.4324/9781315745572-15 Bachmann, R., & Hanappi-Egger, E. (2012). Can trust flourish where institutionalized distrust reigns? In J. Harris, B. Moriarty, & A. Wicks (Eds.), Public Trust in Business (pp. 266– 289). Bachmann, R., & Inkpen, A. C. (2011). Understanding institutional-based trust building processes in inter-organizational relationships. Organization Studies, 32(2), 281–301. https://doi.org/10.1177/0170840610397477 Bachmann, R., & Zaheer, A. (2013). Handbook of Advances in Trust Research Introduction. In R Bachmann & A. Zaheer (Eds.), Handbook of Advances in Trust Research (pp. 1–6). Univ Surrey, Guildford GU2 5XH, Surrey, England. Baghramian, M., Petherbridge, D., & Stout, R. (2020). Vulnerability and Trust: An Introduction. International Journal of Philosophical Studies, Vol. 28, pp. 575–582. https://doi.org/10.1080/09672559.2020.1855814 Baier, A. (1986). Trust and Antitrust. Ethics, 96(2), 231–260. Bailey, A. E., & Figueroa, J. P. (2018). Agency, lapse in condom use and relationship intimacy among female sex workers in Jamaica. Culture, Health and Sexuality, 20(5), 531–544. https://doi.org/10.1080/13691058.2017.1360944 Bailey, P. E., Szczap, P., McLennan, S. N., Slessor, G., Ruffman, T., & Rendell, P. G. (2016). Age-related similarities and differences in first impressions of trustworthiness. Cognition and Emotion, 30(5), 1017–1026. https://doi.org/10.1080/02699931.2015.1039493 Baker, R. W., McNeil, O. V., & Siryk, B. (1985). Expectation and Reality in Freshman Adjustment to College. Journal of Counseling Psychology, 32(1), 94–103. https://doi.org/10.1037/0022-0167.32.1.94 Baker, S. M., Gentry, J. W., & Rittenburg, T. L. (2005). Building understanding of the domain of consumer vulnerability. Journal of Macromarketing, 25(2), 128–139. https://doi.org/10.1177/0276146705280622 Banner, F. (2014). Institutional Sexual Assault and the Rights/Trust Dilemma. Cardozo Public 184 Law, Policy and Ethics Journal, 13, 97. https://doi.org/10.2139/ssrn.2489185 Banta, M. (2019, December 29). Where do cases against former MSU employees with ties to Nassar stand? Lansing State Journal. Retrieved from https://www.lansingstatejournal.com/story/news/local/2019/12/30/criminal-cases-msu-lou- anna-simon-william-strampel-kathie-klages-2019/2662203001/ Barter, C. (1997). Who’s to blame: Conceptualising institutional abuse by children. Early Child Development and Care, 133(1), 101–114. https://doi.org/10.1080/0300443971330108 Baumeister, R. F., & Leary, M. R. (1997). Writing Narrative Literature Reviews. Review of General Psychology, 1(3), 311–320. https://doi.org/10.1037/1089-2680.1.3.311 Bearman, M., & Molloy, E. (2017, December 2). Intellectual streaking: The value of teachers exposing minds (and hearts). Medical Teacher, Vol. 39, pp. 1284–1285. https://doi.org/10.1080/0142159X.2017.1308475 Beck, A., Berzofsky, M., Caspar, R., & Krebs, C. (2013). Sexual Victimization in Prisons and Jails Reported by Inmates, 2011–12. Beck, A., Rantala, R. R., & Rexroat, J. (2014). Sexual Victimization Reported by Adult Correctional Authorities, 2009-11. Behrman, S., & Dan-Cooke, D. (2019). Assessment and management of vulnerable patients. In A. Barrera, C. Attard, & R. Chaplin (Eds.), Oxford textbooks in psychiatry. https://doi.org/http://dx.doi.org/10.1093/med/9780198794257.003.0016 Ben-Yehuda, N. (2001). Betrayals and treason: Violations of trust and loyalty. In Betrayals and Treason: Violations of Trust and Loyalty. https://doi.org/10.4324/9780429502071/BETRAYAL-TREASON-NACHMAN-BEN- YEHUDA Bernstein, J. M. (2011). Trust: On the real but almost always unnoticed, ever-changing foundation of ethical life. Metaphilosophy, 42(4), 395–416. https://doi.org/10.1111/j.1467- 9973.2011.01709.x Beteille, A. (2012). Can Rights Undermine Trust: How Institutions Work and Why They Fail. NUJS Law Review, 5. Bhana, D., & Pillay, J. (2018). Negotiating femininities on campus: Sexuality, gender and risk in an HIV environment. Health Education Journal, 77(8), 915–926. https://doi.org/10.1177/0017896918784693 Bianchi, M. (2015). Willingness to believe and betrayal aversion: the special role of trust in art exchanges. Journal of Cultural Economics, 39(2), 133–151. https://doi.org/10.1007/s10824- 014-9224-3 185 Birrell, P. J., & Freyd, J. J. (2013). Blind to Betrayal: Why we fool ourselves we aren’t being fooled. Wiley. Blakemore, T., Herbert, J. L., Arney, F., & Parkinson, S. (2017). The impacts of institutional child sexual abuse: A rapid review of the evidence. Child Abuse and Neglect, 74, 35–48. https://doi.org/10.1016/j.chiabu.2017.08.006 Blumberg, D. M., Papazoglou, K., & Creighton, S. (2019). The moral risks of policing. In POWER: Police Officer Wellness, Ethics, and Resilience (pp. 49–75). https://doi.org/10.1016/B978-0-12-817872-0.00005-7 Bowen, G. A. (2006). Grounded Theory and Sensitizing Concepts. International Journal of Qualitative Methods, 5(3), 12–23. https://doi.org/10.1177/160940690600500304 Brackenridge, C. (1997). “He owned me basically...”: Women’s experience of sexual abuse in sport. International Review for the Sociology of Sport, 32(2), 115–130. https://doi.org/10.1177/101269097032002001 Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77–101. https://doi.org/10.1191/1478088706qp063oa Brennan, J. (2021). Recognition trust. Philosophical Studies, 178(11), 3799–3818. https://doi.org/10.1007/s11098-021-01626-z Brewer, K. C. (2021). Institutional betrayal in nursing: A concept analysis. Nursing Ethics, 28(6), 1081–1089. https://doi.org/10.1177/0969733021992448 Brockway, G. (1993). Limited paternalism and the salesperson: A reconsideration. Journal of Business Ethics, 12(4), 275–279. https://doi.org/10.1007/BF01666531 Brownmiller, S. (1975). Against Our Will: Men, Women, and Rape. New York: Random House. Brunson, R. K. (2007). “Police Don’t Like Black People”: African- American Young Men’s Accumulated Police Experiences. Criminology & Public Policy, 6(1), 71–101. https://doi.org/10.1111/j.1745-9133.2007.00423.x Brunson, R. K., & Wade, B. A. (2019). “Oh hell no, we don’t talk to police”: Insights on the lack of cooperation in police investigations of urban gun violence. Criminology and Public Policy, 18(3), 623–648. https://doi.org/10.1111/1745-9133.12448 Bublick, E. (2014). Who is Responsible for Child Sexual Abuse? A View from the Penn State Scandal. The Journal of Gender, Race, and Justice, 17(2), 297–311. Bublick, E., & Mindlin, J. (2009). Civil Tort Actions Filed by Victims of Sexual Assault: Promise and Perils. 186 Buchman, D. Z., & Ho, A. (2014). What’s trust got to do with it? Revisiting opioid contracts. Journal of Medical Ethics, 40(10), 673–677. https://doi.org/10.1136/medethics-2013- 101320 Butler, J. V., Giuliano, P., & Guiso, L. (2016). THE RIGHT AMOUNT OF TRUST. Journal of the European Economic Association, 14(5), 1155–1180. https://doi.org/10.1111/jeea.12178 Callan, A., Corbally, M., & McElvaney, R. (2021, April 1). A Scoping Review of Intimate Partner Violence as It Relates to the Experiences of Gay and Bisexual Men. Trauma, Violence, and Abuse, Vol. 22, pp. 233–248. https://doi.org/10.1177/1524838020970898 Campbell, R., Ahrens, C. E., Sefl, T., Wasco, S. M., & Barnes, H. E. (2001). Social reactions to rape victims: Healing and hurtful effects on psychological and physical health outcomes. Violence and Victims, 16(3), 287–302. https://doi.org/10.1891/0886-6708.16.3.287 Campbell, R., Dworkin, E., & Cabral, G. (2009, July). An ecological model of the impact of sexual assault on women’s mental health. Trauma, Violence, and Abuse, Vol. 10, pp. 225– 246. https://doi.org/10.1177/1524838009334456 Campbell, R., & Raja, S. (1999). Secondary victimization of rape victims: Insights from mental health professionals who treat survivors of violence. Violence and Victims, 14(3), 261–275. https://doi.org/10.1891/0886-6708.14.3.261 Cantor, David; Fisher, Bonnie; Chibnall, Susan; Townsend, Reanne; Lee, Hyunshik; Bruce, Carol; Thomas, G. (2015). Report on the AAU Campus Climate Survey on Sexual Assault and Sexual Misconduct. Westat, 3129(301), i-A6-8. Carr, A., Dooley, B., Fitzpatrick, M., Flanagan, E., Flanagan-Howard, R., Tierney, K., … Egan, J. (2010). Adult adjustment of survivors of institutional child abuse in Ireland. Child Abuse and Neglect, 34(7), 477–489. https://doi.org/10.1016/j.chiabu.2009.11.003 Carr, E. M., & Sandmeyer, J. (2018). Exploring the Vicissitudes of Safety and Danger in Psychoanalysis: Developing Trust Through Mutual Engagement. Psychoanalytic Inquiry, 38(8), 557–568. https://doi.org/10.1080/07351690.2018.1521219 Carvalho, S. W., & de Oliveira Mota, M. (2010). The role of trust in creating value and student loyalty in relational exchanges between higher education institutions and their students. Journal of Marketing for Higher Education, 20(1), 145–165. https://doi.org/10.1080/08841241003788201 Castle, E., Eisenberger, N. I., Seeman, T. E., Moons, W. G., Boggero, I. A., Grinblatt, M. S., & Taylor, S. E. (2012). Neural and behavioral bases of age differences in perceptions of trust. Proceedings of the National Academy of Sciences of the United States of America, 109(51), 20848–20852. https://doi.org/10.1073/pnas.1218518109 187 Cave, R. (2022, May 30). Victim describes effect of abuse by police officer as lawsuit against Saint John begins . CBC News. Retrieved from https://www.cbc.ca/news/canada/new- brunswick/kenneth-estabrooks-case-may30-1.6468380 Chan, H. Y. (2021). Reciprocal Trust as an Ethical Response to the COVID-19 Pandemic. Asian Bioethics Review 2021 13:3, 13(3), 335–354. https://doi.org/10.1007/S41649-021-00174-2 Chan, M. L. E. (2009). “Why Did You Hurt Me?” Victim’s Interpersonal Betrayal Attribution and Trust Implications. Review of General Psychology, 13(3), 262–274. https://doi.org/10.1037/a0017138 Charmaz, K. (2006). Constructing Grounded Theory: A Practical Guide through Qualitative Analysis . London: Sage. Chen, S. (Joseph), Waseem, D., Xia, Z. (Raymond), Tran, K. T., Li, Y., & Yao, J. (2021). To disclose or to falsify: The effects of cognitive trust and affective trust on customer cooperation in contact tracing. International Journal of Hospitality Management, 94, 102867. https://doi.org/10.1016/j.ijhm.2021.102867 Citrino, D. (2018). Moving from Institutional Betrayal to Institutional Courage. Bar Journal: Women In Law. Retrieved from https://www.oig.ca.gov/media/reports/ARCHIVE/BOI/Spe- Clark, H., & Fileborn, B. (2011). Responding to women’s experiences of sexual assault in institutional and care settings. ACSSA Wrap, (10), 1–19. Retrieved from www.aifs.gov.au/acssa Clarke, V., & Braun, V. (2017). Thematic analysis. Journal of Positive Psychology, 12(3), 297– 298. https://doi.org/10.1080/17439760.2016.1262613 Colquitt, J. A., Scott, B. A., & LePine, J. A. (2007). Trust, Trustworthiness, and Trust Propensity: A Meta-Analytic Test of Their Unique Relationships With Risk Taking and Job Performance. Journal of Applied Psychology, 92(4), 909–927. https://doi.org/10.1037/0021- 9010.92.4.909 Crenshaw, K. (1991). Mapping the Margins: Intersectionality, Identity Politics, and Violence against Women of Color. Stanford Law Review, 43(6), 1241. https://doi.org/10.2307/1229039 Cresswell, J. (2003). Qualitative Inquiry and Research Design: Choosing Among Five Approaches. Health Promotion Practice, Vol. 16. https://doi.org/10.1177/1524839915580941 Crossmaker, M. (1991). Behind locked doors-Institutional sexual abuse. Sexuality and Disability, 9(3), 201–219. https://doi.org/10.1007/BF01102393 Daigle, L. E., Fisher, B. S., & Cullen, F. T. (2008). The violent and sexual victimization of 188 college women: Is repeat victimization a problem? Journal of Interpersonal Violence, 23(9), 1296–1313. https://doi.org/10.1177/0886260508314293 Damodaran, A. K., Jones, P., & Shulruf, B. (2021). Trust and risk pitfalls in medical education: A qualitative study of clinical teachers. Medical Teacher, 43(11), 1309–1316. https://doi.org/10.1080/0142159X.2021.1944613 De Aguiar, A., & Camargo, B. V. (2014). Romantic relationships, adolescence and HIV: Love as an element of vulnerability. Paideia, 24(58), 165–175. https://doi.org/10.1590/1982- 43272458201404 Decker, M., & Littleton, H. L. (2018). Sexual Revictimization Among College Women: A Review Through an Ecological Lens. Victims and Offenders, 13(4), 558–588. https://doi.org/10.1080/15564886.2017.1390514 Delker, B. C., & Freyd, J. J. (2014). From Betrayal to the Bottle: Investigating Possible Pathways From Trauma to Problematic Substance Use. Journal of Traumatic Stress, 27(5), 576–584. https://doi.org/10.1002/jts.21959 Delker, B. C., Smith, C. P., Rosenthal, M. N., Bernstein, R. E., & Freyd, J. J. (2018). When Home Is Where the Harm Is: Family Betrayal and Posttraumatic Outcomes in Young Adulthood. Journal of Aggression, Maltreatment and Trauma, 27(7), 720–743. https://doi.org/10.1080/10926771.2017.1382639 Denhollander, R. (2019). What is a girl worth? : my story of breaking the silence and exposing the truth about Larry Nassar and USA gymnastics. Tyndale Momontum. Department of Defense. (2019). Department of Defense Annual Report on Sexual Assault in the Military: Fiscal Year 2018. Retrieved from https://int.nyt.com/data/documenthelper/800- dod-annual-report-on-sexual-as/d659d6d0126ad2b19c18/optimized/full.pdf#page=1 Deprince, A. P. (2001). Trauma and posttraumatic responses: An examination of fear and betrayal. Dissertation Abstracts International: Section B: The Sciences and Engineering, Vol. 62, p. 2953. Retrieved from https://psycnet.apa.org/record/2001-95024-199 DePrince, P., & Freyd, J. J. (2002). The intersection of gender and betrayal in trauma. Gender and PTSD., pp. 98–113. Dixon-Woods, M., Agarwal, S., Young, B., & Sutton, A. J. (2005). Synthesising Qualitative and Quantitative Evidence: A Review of Possible Methods . Journal of Health Services Research & Policy, 10(1), 45–53. https://doi.org/10.1258/1355819052801804 Dupont, Q., & Karpoff, J. M. (2020). The Trust Triangle: Laws, Reputation, and Culture in Empirical Finance Research. Journal of Business Ethics, 163(2), 217–238. https://doi.org/10.1007/s10551-019-04229-1 189 Dworkin, E. R., Menon, S. V., Bystrynski, J., & Allen, N. E. (2017, August 1). Sexual assault victimization and psychopathology: A review and meta-analysis. Clinical Psychology Review, Vol. 56, pp. 65–81. https://doi.org/10.1016/j.cpr.2017.06.002 Edwards, V. J., Freyd, J. J., Dube, S. R., Anda, R. F., & Felitti, V. J. (2012). Health outcomes by closeness of sexual abuse perpetrator: A test of betrayal trauma theory. Journal of Aggression, Maltreatment and Trauma, 21(2), 133–148. https://doi.org/10.1080/10926771.2012.648100 Eiler, B. A., Al-Kire, R., Doyle, P. C., & Wayment, H. A. (2019). Power and Trust Dynamics of Sexual Violence: A Textual Analysis of Nassar Victim Impact Statements and #MeToo Disclosures on Twitter. Journal of Clinical Sport Psychology, 13(2), 290–310. https://doi.org/10.1123/JCSP.2018-0056 Elsbach, K. D. (2004). Managing images of trustworthiness in organizations. In R.M. Kramer & K. S. Cook (Eds.), Trust and Distrust in Organizations : Dilemmas and Approaches (pp. 275–292). Retrieved from https://psycnet.apa.org/record/2004-16590-011 Ferrin, D. L., Bligh, M. C., & Kohles, J. C. (2008). It takes two to tango: An interdependence analysis of the spiraling of perceived trustworthiness and cooperation in interpersonal and intergroup relationships. Organizational Behavior and Human Decision Processes, 107(2), 161–178. https://doi.org/10.1016/j.obhdp.2008.02.012 Fitness, J. (2012). Betrayal, Rejection, Revenge, and Forgiveness: An Interpersonal Script Approach. In M. Leary (Ed.), Interpersonal Rejection (pp. 73–103). https://doi.org/10.1093/acprof:oso/9780195130157.003.0004 Flashman, J., & Gambetta, D. (2014). Thick as thieves: Homophily and trust among deviants. Rationality and Society, 26(1), 3–45. https://doi.org/10.1177/1043463113512996 Foucault, M. (1977). Discipline and punish : the birth of the prison. London: Allen Lane. Foucault, M. (1982). The Subject and Power. Critical Inquiry, 8(4), 777–795. https://doi.org/10.1086/448181 Frederiksen, M. (2014). Trust in the face of uncertainty: a qualitative study of intersubjective trust and risk. International Review of Sociology, 24(1), 130–144. https://doi.org/10.1080/03906701.2014.894335 French, L., Hanna, P., & Huckle, C. (2022). “If I die, they do not care”: U.K. National Health Service staff experiences of betrayal-based moral injury during COVID-19. Psychological Trauma : Theory, Research, Practice and Policy, 14(3), 516–521. https://doi.org/10.1037/TRA0001134 Freyd, J. J. (1996). Betrayal trauma: the logic of forgetting childhood abuse. Harvard University Press. 190 Freyd, J. J., Klest, B., & Allard, C. B. (2005). Betrayal trauma: Relationship to physical health, psychological distress, and a written disclosure intervention. Journal of Trauma and Dissociation, 6(3), 83–104. https://doi.org/10.1300/J229v06n03_04 Fukuyama, F. (1995). Trust: The Social Virtues and The Creation of Prosperity. New York, NY: Free Press. Fulmer, A. (2018). Multilevel Trust: Antecedents and outcomes of trust at different levels. In R. H. Searle, A. M. I. Nienaber, & S. B. Sitkin (Eds.), ROUTLEDGE COMPANION TO TRUST (pp. 143–160). https://doi.org/10.4324/9781315745572 Fulmer, Ashley, & Dirks, K. (2018). Multilevel trust: A theoretical and practical imperative. Journal of Trust Research, 8(2), 137–141. https://doi.org/10.1080/21515581.2018.1531657 Gartner, R. B. (1999). Betrayed as boys : psychodynamic treatment of sexually abused men. New York, NY: Guilford Press. Gazmararian, J. A., Petersen, R., Spitz, A. M., Goodwin, M. M., Saltzman, L. E., & Marks, J. S. (2000). Violence and Reproductive Health: Current Knowledge and Future Research Directions. Maternal and Child Health Journal, 4(2). Gerson, M. (2022, May 23). Opinion | The report on Southern Baptist abuses is a portrait of brutal misogyny . The Washington Post. Retrieved from https://www.washingtonpost.com/opinions/2022/05/23/southern-baptist-sexual-abuse- culture-of-misogyny/ Gil, E. (1982). Institutional Abuse of Children in Out-of-Home Care. Child & Youth Services, 4(1–2), 7–13. https://doi.org/10.1300/j024v04n01_03 Glaser, B. G. (1963). Retreating Research Materials: The Use of Secondary Analysis by the Independent Researcher. The American Behavioral Scientist, 11–14. Glaser, B. G. (1965). The Constant Comparative Method of Qualitative Analysis. Social Problems, 12(4), 436–445. Glaser, B. G., & Strauss, A. L. (1967). The Discovery of Grounded Theory: Strategies for Qualitative Research. New Brunswick, NJ: Aldine Transaction. Gobin, R. L., & Freyd, J. J. (2009). Betrayal and Revictimization: Preliminary Findings. Psychological Trauma: Theory, Research, Practice, and Policy, 1(3), 242–257. https://doi.org/10.1037/a0017469 Gobin, R. L., & Freyd, J. J. (2014). The impact of betrayal trauma on the tendency to trust. Psychological Trauma: Theory, Research, Practice, and Policy, 6(5), 505–511. https://doi.org/10.1037/a0032452 191 Gobin, R. L., (2012). Trauma, Trust, and Betrayal Awareness. University of Oregon. Goffman, E. (1961). Asylums: Essays on the Social Situation of Mental Patients and Other Inmates. Retrieved from https://www.amazon.com/Asylums-Essays-Situation-Patients- Inmates/dp/0385000162 Goldsmith, R. E., Martin, C. G., & Smith, C. P. (2014). Systemic Trauma. Journal of Trauma and Dissociation, Vol. 15, pp. 117–132. https://doi.org/10.1080/15299732.2014.871666 Gollwitzer, M., Rothmund, T., & De Cremer, D. (Ed. . (2009). When the need to trust results in unethical behavior: The sensitivity to mean intentions (SeMI) model. In Psychological perspectives on unethical behavior and decision making (pp. 135–152). Grant, M. J., & Booth, A. (2009, June 1). A typology of reviews: An analysis of 14 review types and associated methodologies. Health Information and Libraries Journal, Vol. 26, pp. 91– 108. https://doi.org/10.1111/j.1471-1842.2009.00848.x Gu, L., Liu, J., & Peng, Y. (2022). Locality Stereotype, CEO Trustworthiness and Stock Price Crash Risk: Evidence from China. Journal of Business Ethics, 175(4), 773–797. https://doi.org/10.1007/s10551-020-04631-0 Guiora, A. (2020). Armies of Enablers: Survivor Stories of Complicity and Betrayal in Sexual Assaults. Chicago: American Bar Association. Hamm, J. A., Cox, J. G., Zwickle, A., Zhuang, J., Cruz, S. M., Upham, B. L., … Dearing, J. W. (2019). Trust in whom? Dioxin, organizations, risk perception, and fish consumption in Michigan’s Saginaw Bay watershed. Journal of Risk Research, 22(12), 1624–1637. https://doi.org/10.1080/13669877.2018.1501599 Hamm, J. A., & Mollering, G. (2022). How Concerned Are You? Integrating Focal Vulnerability as an Antecedent of Trust. [Unpublished Manuscript]. Hamm, J. A., Searle, R., Carr, J. D., & Rivers, L. (2021). Public Vulnerability to the Police: A Quantitative Inquiry: Https://Doi.Org/10.1177/00938548211008489, 48(12), 1749–1769. https://doi.org/10.1177/00938548211008489 Hamm, J. A., Smidt, C., & Mayer, R. C. (2019). Understanding the psychological nature and mechanisms of political trust. PLOS ONE, 14(5), e0215835. https://doi.org/10.1371/JOURNAL.PONE.0215835 Hamm, J. A., Trinkner, R., & Carr, J. D. (2017). Fair Process, Trust, and Cooperation: Moving Toward an Integrated Framework of Police Legitimacy. Criminal Justice and Behavior, 44(9), 1183–1212. https://doi.org/10.1177/0093854817710058 Hamm, J. A., & Wolfe, S. E. (2021). Using public priorities to disentangle the dimensions of 192 procedural justice and trustworthiness in police-citizen interactions. British Journal of Criminology, 61(2), 558–579. https://doi.org/10.1093/bjc/azaa079 Han, S. D., & Mosqueda, L. (2020, July 1). Elder Abuse in the COVID-19 Era. Journal of the American Geriatrics Society, Vol. 68, pp. 1386–1387. https://doi.org/10.1111/jgs.16496 Harris, J. D., Keevil, A. A. C., & Wicks, A. C. (2013). Public trust in the institution of business. In R Bachmann & A. Zaheer (Eds.), Handbook of Advances in Trust Research. Univ Virginia, Darden Sch Business, Charlottesville, VA 22903 USA. Hauser, C., & Astor, M. (2018, January 25). The Larry Nassar Case: What Happened and How the Fallout Is Spreading - The New York Times. New York Times. Retrieved from https://www.nytimes.com/2018/01/25/sports/larry-nassar-gymnastics-abuse.html Herman, J. L. (1992). Trauma and recovery. BasicBooks. Herman, J. L. (2015). Trauma and Recovery: The Aftermath of Violence—From Domestic Abuse to Political Terror. https://doi.org/10.1176/appi.psychotherapy.2015.69.4.455 Holliday, R., & Monteith, L. L. (2019). Seeking help for the health sequelae of military sexual trauma: a theory-driven model of the role of institutional betrayal. Journal of Trauma and Dissociation, 20(3), 340–356. https://doi.org/10.1080/15299732.2019.1571888 Houde, L. D., Sherman, D. M., White, T. B., & Sheppard, B. H. (2004). The four faces of trust: An empirical study of the nature of trust in relational forms. In N. Haslam (Ed.), Relational Models Theory: A Contemporary Overview (pp. 287–308). https://doi.org/10.4324/9781410611413 Huddy, S. (2015). Vulnerability in the classroom : Instructor’s ability to build trust impacts the student’s learning experience. International Journal of Education Research, 10(2), 96–104. Jinkerson, J. D. (2016). Defining and assessing moral injury: A syndrome perspective. Traumatology, 22(2), 122–130. https://doi.org/http://dx.doi.org/10.1037/trm0000069 Judges, R. A., Gallant, S. N., Yang, L., & Lee, K. (2017). The role of cognition, personality, and trust in fraud victimization in older adults. Frontiers in Psychology, 8(APR), 588. https://doi.org/10.3389/fpsyg.2017.00588 Kamkar, K., Russo, C., Chopko, B. A., Tuttle, B. M. Q., Blumberg, D. M., & Papazoglou, K. (2019). Moral injury in law enforcement. In POWER: Police Officer Wellness, Ethics, and Resilience (pp. 117–128). https://doi.org/10.1016/B978-0-12-817872-0.00008-2 Kerler, W. A., & Killough, L. N. (2009). The effects of satisfaction with a client’s management during a prior audit engagement, trust, and moral reasoning on auditors’ perceived risk of management fraud. Journal of Business Ethics, 85(2), 109–136. https://doi.org/10.1007/s10551-008-9752-x 193 Kirsner, D. (2011). Trust and the global financial crisis. In S. Long & B. Sievers (Eds.), Towards a socioanalysis of money, finance and capitalism : beneath the surface of the financial industry (pp. 278–291). https://doi.org/10.4337/9780857931382.00019 Koehler, J. J., & Gershoff, A. D. (2003, March 1). Betrayal aversion: When agents of protection become agents of harm. Organizational Behavior and Human Decision Processes, Vol. 90, pp. 244–261. https://doi.org/10.1016/S0749-5978(02)00518-6 Kostis, A., & Nasholm, M. H. (2020). Towards a research agenda on how, when and why trust and distrust matter to coopetition. Journal of Trust Research 10(1), 66–90. https://doi.org/10.1080/21515581.2019.1692664 Kraft, S. A., Cho, M. K., Gillespie, K., Halley, M., Varsava, N., Ormond, K. E., … Soo-Jin Lee, S. (2018). Beyond Consent: Building Trusting Relationships With Diverse Populations in Precision Medicine Research. American Journal of Bioethics, 18(4), 3–20. https://doi.org/10.1080/15265161.2018.1431322 Kramer, Roderick M. (1999). TRUST AND DISTRUST IN ORGANIZATIONS: Emerging Perspectives, Enduring Questions. In Annu. Rev. Psychol (Vol. 50). Kramer, Roderick M. (2011). Moving between laboratory and field: A multi-method approach for studying trust judgments. In F. Lyon, G. Möllering, & M. N. K. Saunders (Eds.), Handbook of Research Methods on Trust (pp. 19–28). https://doi.org/10.4337/9780857932013.00010 Kuran, C. H. A., Morsut, C., Kruke, B. I., Krüger, M., Segnestam, L., Orru, K., … Torpan, S. (2020). Vulnerability and vulnerable groups from an intersectionality perspective. International Journal of Disaster Risk Reduction, 50, 1–8. https://doi.org/10.1016/j.ijdrr.2020.101826 Kutsyuruba, B., & Walker, K. D. (2016). The destructive effects of distrust: Leaders as brokers of trust in organizations. Advances in Educational Administration, 26, 133–154. https://doi.org/10.1108/S1479-366020160000026008 Landa, N., Zhou, S., & Tshotsho, B. (2019). Interrogating the Role of Language in Clergy Sexual Abuse of Women and Girls in Zimbabwe. Journal for the Study of Religion, 32(2), 1–20. https://doi.org/10.17159/2413-3027/2019/v32n2a5 Levac, D., Colquhoun, H., & O’Brien, K. K. (2010). Scoping studies: Advancing the methodology. Implementation Science, 5(1), 1–9. https://doi.org/10.1186/1748-5908-5-69 Levenson, E., & Joseph, E. (2018). Victims Confront Larry Nassar in court: “Little girls don’t stay litle forever.” CNN. Retrieved from https://www.cnn.com/2018/01/16/us/larry-nassar- sentencing-gymnastics/index.html 194 Lewicki, R. J., & Bunker, B. B. (1996). Developing and Maintaining Trust in Work Relationships. In Roderick M. Kramer & T. R. Tyler (Eds.), Trust in Organizations: Frontiers of Theory and Research (pp. 114–139). https://doi.org/10.4135/9781452243610.n7 Lind, M. N., Adams-Clark, A. A., & Freyd, J. J. (2020). Isn’t high school bad enough already? Rates of gender harassment and institutional betrayal in high school and their association with trauma-related symptoms. PLoS ONE, 15(8 August). https://doi.org/10.1371/journal.pone.0237713 Luchters, S., Richter, M. L., Bosire, W., Nelson, G., Kingola, N., Zhang, X. D., … Chersich, M. F. (2013). The Contribution of Emotional Partners to Sexual Risk Taking and Violence among Female Sex Workers in Mombasa, Kenya: A Cohort Study. PLoS ONE, 8(8), e68855. https://doi.org/10.1371/journal.pone.0068855 Lumineau, F., & Schilke, O. (2018). Trust development across levels of analysis: An embedded- agency perspective. Journal of Trust Research, 8(2), 238–248. https://doi.org/10.1080/21515581.2018.1531766 Mackelprang, J. L., Klest, B., Najmabadi, S. J., Valley-Gray, S., Gonzalez, E. A., & Cash, R. E. (Gene. (2014). Betrayal Trauma Among Homeless Adults: Associations With Revictimization, Psychological Well-Being, and Health. Journal of Interpersonal Violence, 29(6), 1028–1049. https://doi.org/10.1177/0886260513506060 Maia, C., Guilhem, D., & Freitas, D. (2008). Vulnerabilidade ao HIV/Aids de pessoas heterossexuais casadas ou em união estável. Revista de Saúde Pública, 42(2), 242–248. https://doi.org/10.1590/s0034-89102008005000004 Martindale, M. (2020, December 18). Critics tell MSU trustees to do more to prevent sexual abuse in wake of Nassar scandal. The Detroit News. Retrieved from https://www.detroitnews.com/story/news/local/michigan/2020/12/18/critics-msu-trustees- do-more-prevent-sexual-abuse-wake-nassar-scandal/3958936001/ Mayer, R. C., Davis, J. H., & Schoorman, F. D. (1995). An Integrative Model of Organizational Trust. The Academy Management Review, 20(3), 709–734. https://doi.org/10.1002/j.2050- 0416.1927.tb05040.x Mays, N., Pope, C., & Popay, J. (2005, July). Systematically reviewing qualitative and quantitative evidence to inform management and policy-making in the health field. Journal of Health Services Research and Policy, Vol. 10, pp. 6–20. https://doi.org/10.1258/1355819054308576 McAlinden, A. M. (2006). ’Setting “em up”: Personal, familial and institutional grooming in the sexual abuse of children. Social and Legal Studies, 15(3), 339–362. https://doi.org/10.1177/0964663906066613 195 McCauley, H. L., Campbell, R., Buchanan, N. C. T., & Moylan, C. A. (2019). Advancing Theory, Methods, and Dissemination in Sexual Violence Research to Build a More Equitable Future: An Intersectional, Community-Engaged Approach. Violence Against Women, 25(16), 1906–1931. https://doi.org/10.1177/1077801219875823 McEvily, B., Weber, R. A., Bicchieri, C., & Ho, V. T. (2006). Can groups be trusted? An experimental study of trust in collective entities. In Reinhard Bachmann & A. Zaheer (Eds.), Handbook Of Trust Research (pp. 52–67). https://doi.org/10.4337/9781847202819.00009 McMahon, S. M., Karp, D. R., & Mulhern, H. (2019). Addressing individual and community needs in the aftermath of campus sexual misconduct: restorative justice as a way forward in the re-entry process. Journal of Sexual Aggression, 25(1), 49–59. https://doi.org/10.1080/13552600.2018.1507488 Mendez, X. (2020). Beyond Nassar: A Transformative Justice and Decolonial Feminist Approach to Campus Sexual Assault. Frontiers: A Journal of Women Studies, 41(2), 82– 104. Retrieved from https://muse.jhu.edu/article/765266 Misztal, B. A. (2012). Trust: Acceptance of, precaution against and cause of vulnerability. In Trust: Comparative Perspectives (pp. 209–236). https://doi.org/10.1163/9789004221383_010 Moher, D., Liberati, A., Tetzlaff, J., & Altman, D. G. (2009). Preferred Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement. PLoS Medicine, 6(7), e1000097. https://doi.org/10.1371/journal.pmed.1000097 Molendijk, T., Verkoren, W., Drogendijk, A., Elands, M., Kramer, E. H., Smit, A., & Verweij, D. (2022). Contextual dimensions of moral injury: An interdisciplinary review. Military Psychology. https://doi.org/10.1080/08995605.2022.2035643 Möllering, G. (2001). The nature of trust: From Georg Simmel to a theory of expectation, interpretation and suspension. Sociology, 35(2), 403–420. https://doi.org/10.1017/S0038038501000190 Möllering, G. (2019). Connecting trust and power. Journal of Trust Research, 9(1), 1–5. https://doi.org/10.1080/21515581.2019.1609732 Monteith, L. L., Bahraini, N. H., Matarazzo, B. B., Soberay, K. A., & Smith, C. P. (2016). Perceptions of Institutional Betrayal Predict Suicidal Self-Directed Violence Among Veterans Exposed to Military Sexual Trauma. Journal of Clinical Psychology, 72(7), 743– 755. https://doi.org/10.1002/jclp.22292 Montgomery, K., Jordens, C. F. C., & Little, M. (2008). How vulnerability and trust interact during extreme events: Insights for human service agencies and organizations. Administration and Society, 40(6), 621–644. https://doi.org/10.1177/0095399708321670 196 Morris, J. H., & Moberg, D. J. (1994). Work organizations as contexts for trust and betrayal. In T. R. Sarbin, R. M. Carney, & C. Eoyang (Eds.), Citizen espionage: Studies in trust and betrayal. (pp. 163–187). Morrison, E. W., & Robinson, S. L. (1997). When employees feel betrayed: A model of how psychological contract violation develops. Academy of Management Review, 22(1), 226– 256. https://doi.org/10.5465/AMR.1997.9707180265 Moylan, C. A., Javorka, M., Maas, M. K., Meier, E., & McCauley, H. L. (2021). Campus sexual assault climate: Toward an expanded definition and improved assessment. Psychology of Violence, 11(3), 296–306. https://doi.org/10.1037/vio0000382 Muehlenhard, C. L., Peterson, Z. D., Humphreys, T. P., & Jozkowski, K. N. (2017). Evaluating the One-in-Five Statistic: Women’s Risk of Sexual Assault While in College. Https://Doi.Org/10.1080/00224499.2017.1295014, 54(4–5), 549–576. https://doi.org/10.1080/00224499.2017.1295014 Nienaber, A. M., Romeike, P. D., Searle, R., & Schewe, G. (2015). A qualitative meta-analysis of trust in supervisor-subordinate relationships. Journal of Managerial Psychology, 30(5), 507–534. https://doi.org/10.1108/JMP-06-2013-0187/FULL/PDF Nordberg, A., Crawford, M. R., Praetorius, R. T., & Hatcher, S. S. (2016). Exploring Minority Youths’ Police Encounters: A Qualitative Interpretive Method. Child and Adolescent Social Work Journal, 33, 137–149. Novack, D. (2018). The Analyst’s Experience of Trust and Mistrust. Psychoanalytic Dialogues, 28(6), 733–747. https://doi.org/10.1080/10481885.2018.1538750 Nunno, M. A., & Motz, J. K. (1988). The development of an effective response to the abuse of children in out-of-home care. Child Abuse and Neglect, 12(4), 521–528. https://doi.org/10.1016/0145-2134(88)90069-5 O’Neal, E. N., & Hayes, B. E. (2020). “Most [False Reports] Involve Teens”: Officer Attitudes Toward Teenage Sexual Assault Complainants—A Qualitative Analysis. Violence Against Women, 26(1), 24–45. https://doi.org/10.1177/1077801219828537 Orbuch, T. L. (1997). People’s Accounts Count: The Sociology of Accounts. Annual Review of Sociology, 23(1), 455–478. https://doi.org/10.1146/annurev.soc.23.1.455 Orth, U. (2002). Secondary victimization of crime victims by criminal proceedings. Social Justice Research, 15(4), 313–325. https://doi.org/10.1023/A:1021210323461 Ostrom, E. (2009). Understanding institutional diversity. In Understanding Institutional Diversity. https://doi.org/10.1093/oso/9780190672492.003.0008 197 Owen, J., Quirk, K., & Manthos, M. (2012). I Get No Respect: The Relationship Between Betrayal Trauma and Romantic Relationship Functioning. Journal of Trauma and Dissociation, 13(2), 175–189. https://doi.org/10.1080/15299732.2012.642760 Palmer, C. T. (2005). Mummers and moshers: Two rituals of trust in changing social environments. Ethnology, 44(2), 147–166. https://doi.org/10.2307/3773994 Parrot, A., & Bechhofer, L. (1991). Acquaintance Rape: The Hidden Crime. Retrieved from https://www.ncjrs.gov/App/abstractdb/AbstractDBDetails.aspx?id=132831 Parsons, J., & Bergin, T. (2010). The Impact of criminal justice involvement on victims’ mental health. Journal of Traumatic Stress, 23(2), 182–188. https://doi.org/10.1002/jts.20505 Patent, V., & Searle, R. H. (2019). Qualitative meta-analysis of propensity to trust measurement. Journal of Trust Research, 9(2), 136–163. https://doi.org/10.1080/21515581.2019.1675074 Peters, M. D. J., Godfrey, C. M., Khalil, H., McInerney, P., Parker, D., & Soares, C. B. (2015). Guidance for conducting systematic scoping reviews. International Journal of Evidence- Based Healthcare, 13(3), 141–146. https://doi.org/10.1097/XEB.0000000000000050 Pirson, M., & Malhotra, D. (2011). Foundations of Organizational Trust. Organization Science, 22(4), 1087–1104. https://doi.org/10.1287/ORSC.1100.0581 Platt, M. G., & Freyd, J. J. (2015). Betray my trust, shame on me: Shame, dissociation, fear, and betrayal trauma. Psychological Trauma: Theory, Research, Practice, and Policy, 7(4), 398– 404. https://doi.org/10.1037/tra0000022 Poon, J. M. L. (2013). Effects of benevolence, integrity, and ability on trust-in-supervisor. Employee Relations, 35(4), 396–407. https://doi.org/10.1108/ER-03-2012- 0025/FULL/XML Poppo, L. (2013). Origins of inter-organizational trust: a review and query for further research. In R Bachmann & A. Zaheer (Eds.), Handbok of Advances in Trust Research Univ Kansas, Lawrence: Edward Elgar Publishing. Pratt, M. G., Lepisto, D. A., & Dane, E. (2019). The Hidden Side of Trust: Supporting and Sustaining Leaps of Faith among Firefighters*. Administrative Science Quarterly, 64(2), 398–434. https://doi.org/10.1177/0001839218769252 PytlikZillig, L. M., & Kimbrough, C. D. (2016). Consensus on conceptualizations and definitions of trust: Are we there yet? In Interdisciplinary Perspectives on Trust: Towards Theoretical and Methodological Integration (pp. 17–47). https://doi.org/10.1007/978-3-319-22261-5_2 Rahal, S., & Kozlowski, K. (2018). 204 impact statements, 9 days, 2 counties, a life sentence for Larry Nassar. Retrieved December 17, 2019, from Detriot News website: https://www.detroitnews.com/story/news/local/michigan/2018/02/08/204-impact- 198 statements-9-days-2-counties-life-sentence-larry-nassar/1066335001/ Robinson, S. L. (1996). Trust and breach of the psychological contract. Administrative Science Quarterly, 41(4), 574–599. https://doi.org/10.2307/2393868 Robinson, S. L., Dirks, K. T., & Ozcelik, H. (2004). Untangling the knot of trust and betrayal. In R.M. Kramer & K. S. Cook (Eds.), Trust and Distrust in Organizations : Dilemmas and Approaches (pp. 327–341). Retrieved from https://psycnet.apa.org/record/2004-16590-013 Rosenthal, M., Smith, C. P., & Freyd, J. J. (2017). Behind closed doors: university employees as stakeholders in campus sexual violence. Journal of Aggression, Conflict and Peace Research, 9(4), 290–304. https://doi.org/10.1108/JACPR-02-2017-0272 Roskin-Frazee, A. (2022). “Terrifying and Exhausting”: Secondary Victimization in Title IX Proceedings at U.S. Higher Education Institutions: Feminist Criminology. https://doi.org/10.1177/15570851221105853 Rotenberg, K. J. (2010). The conceptualization of interpersonal trust: A basis, domain, and target framework. In Interpersonal Trust During Childhood and Adolescence (pp. 8–27). https://doi.org/10.1017/CBO9780511750946.002 Rousseau, D. M., Sitkin, S. B., Burt, R. S., & Camerer, C. (1998, July 1). Not so different after all: A cross-discipline view of trust. Academy of Management Review, Vol. 23, pp. 393– 404. https://doi.org/10.5465/AMR.1998.926617 Sampaio, C. H., Perin, M. G., Simões, C., & Kleinowski, H. (2012). Students’ trust, value and loyalty: Evidence from higher education in Brazil. Journal of Marketing for Higher Education, 22(1), 83–100. https://doi.org/10.1080/08841241.2012.705796 Santaularia, J., Johnson, M., Hart, L., Haskett, L., Welsh, E., & Faseru, B. (2014). Relationships between sexual violence and chronic disease: A cross-sectional study. BMC Public Health, 14(1), 1–7. https://doi.org/10.1186/1471-2458-14-1286/TABLES/4 Sapiro, B. (2020). Assessing trustworthiness: Marginalized youth and the central relational paradox in treatment. Children and Youth Services Review, 116, 105178. https://doi.org/10.1016/j.childyouth.2020.105178 Schilke, O., Wiedenfels, G., Brettel, M., & Zucker, L. G. (2017). Interorganizational trust production contingent on product and performance uncertainty. SOCIO-ECONOMIC REVIEW, 15(2), 307–330. https://doi.org/10.1093/ser/mww003 Schoorman, F. D., Mayer, R. C., & Davis, J. H. (2007). An integrative model of organizational trust: Past, present, and future. Academy of Management Review, 32(2), 344–354. https://doi.org/10.5465/AMR.2007.24348410 Scott, S. (2010). Revisiting the total institution: Performative regulation in the reinventive 199 institution. Sociology, 44(2), 213–231. https://doi.org/10.1177/0038038509357198 Serva, M. A., Fuller, M. A., & Mayer, R. C. (2005). The reciprocal nature of trust: a longitudinal study of interacting teams. Journal of Organizational Behavior, 26(6), 625–648. https://doi.org/10.1002/JOB.331 Shackelford, T. K., & Buss, D. M. (1996). Betrayal in mateships, friendships, and coalitions. Personality and Social Psychology Bulletin, 22(11), 1151–1164. https://doi.org/10.1177/01461672962211006 Sheley, E. (2012). Reverberations of the victim’s “voice”: Victim impact statements and the cultural project of punishment. Indiana Law Journal, 87(3), 1247–1286. Siddaway, A. P., Wood, A. M., & Hedges, L. V. (2019). How to Do a Systematic Review: A Best Practice Guide for Conducting and Reporting Narrative Reviews, Meta-Analyses, and Meta-Syntheses. Annual Review of Psychology, 70, 747–770. https://doi.org/10.1146/annurev-psych-010418-102803 Sitkin, S. B., & Pablo, A. L. (1992). Reconceptualizing the determinants of risk behavior. ACADEMY OF MANAGEMENT REVIEW, 17(1), 9–38. https://doi.org/10.2307/258646 Smith, C. P. (2016). First, Do No Harm: Institutional Betrayal in Healthcare. University of Oregon. Smith, C. P. (2017). First, do no harm: Institutional betrayal and trust in health care organizations. Journal of Multidisciplinary Healthcare, 10, 133–144. https://doi.org/10.2147/JMDH.S125885 Smith, C. P., Cunningham, S. A., & Freyd, J. J. (2016). Sexual violence, institutional betrayal, and psychological outcomes for LGB college students. Translational Issues in Psychological Science, 2(4), 351–360. https://doi.org/10.1037/tps0000094 Smith, C. P., & Freyd, J. J. (2013). Dangerous Safe Havens: Institutional Betrayal Exacerbates Sexual Trauma. Journal of Traumatic Stress, 26(1), 119–124. https://doi.org/10.1002/jts.21778 Smith, C. P., & Freyd, J. J. (2014). Institutional betrayal. American Psychologist, 69(6), 575– 584. https://doi.org/10.1037/a0037564 Smith, C. P., & Freyd, J. J. (2017). Insult, then Injury: Interpersonal and Institutional Betrayal Linked to Health and Dissociation. Journal of Aggression, Maltreatment and Trauma, 26(10), 1117–1131. https://doi.org/10.1080/10926771.2017.1322654 Smith, C. P., Gómez, J., & Freyd, J. J. (2014). The Psychology of Judicial Betrayal. Roger Williams University Law Review, (March 2013), 451–475. 200 Smith, S., Zhang, X., Basile, K. C., Merrick, M. T., Wang, J., Kresnow, M., & Chen, J. (2018). The National Intimate Partner and Sexual Violence Survey (NISVS): 2015 Data Brief – Updated Release. Atlanta, GA. National Center for Injury Prevention and Control, Centers for Disease Control and Prevention. Svrluga, S. (2022, May 3). Education Department investigates Liberty University after sexual assault complaints . The Washington Post. Retrieved from https://www.washingtonpost.com/education/2022/05/03/liberty-university-investigation- clery-act/ Tan, H., & Lim, A. (2009). Trust in coworkers and trust in organizations. Journal of Psychology: Interdisciplinary and Applied, 143(1), 45–66. https://doi.org/10.3200/JRLP.143.1.45-66 Tan, H., & Tan, C. S. F. (2000). Toward the Differentiation of Trust in Supervisor and Trust in Organization. Genetic, Social, and General Psychology Monographs, 126(2), 241–260. Tang, S. S. S., & Freyd, J. J. (2012). Betrayal trauma and gender differences in posttraumatic stress. Psychological Trauma: Theory, Research, Practice, and Policy, 4(5), 469–478. https://doi.org/10.1037/a0025765 Taylor, K. M., Thielking, M., Mackelprang, J. L., Meyer, D., Quinn, S., & Flatau, P. (2020). When trauma violates trust: PTSD symptoms among chronically homeless adults in Australia. Journal of Social Distress and Homelessness, 1–12. https://doi.org/10.1080/10530789.2020.1819076 Thomas, J., & Harden, A. (2008). Methods for the thematic synthesis of qualitative research in systematic reviews. BMC Medical Research Methodology, 8(1), 1–10. https://doi.org/10.1186/1471-2288-8-45 Tjaden, P. G., & Thoennes, N. (2006). Extent, nature, and consequences of rape victimization : findings from the National Violence Against Women Survey. Retrieved from https://stacks.cdc.gov/view/cdc/21950 Tracy, M. (2018, February 5). For Nassar Accusers From Michigan State Teams, Feelings of a Trust Betrayed - The New York Times. New York Times. Retrieved from https://www.nytimes.com/2018/02/05/sports/michigan-state-larry-nassar.html Tricco, A. C., Lillie, E., Zarin, W., O’Brien, K. K., Colquhoun, H., Levac, D., Moher, D., Peters, M., Horsley, T., Weeks, L., Hempel, S., Akl, E., Chang, C., McGowan, J., Stewart, L., Hartling, L., Aldcroft, A., Wilson, M., Garrity, C…Straus, S. (2018, October 2). PRISMA extension for scoping reviews (PRISMA-ScR): Checklist and explanation. Annals of Internal Medicine, Vol. 169, pp. 467–473. https://doi.org/10.7326/M18-0850 Weber, J. M., Malhotra, D., & Murnighan, J. K. (2004). Normal acts of irrational trust: Motivated attributions and the trust development process. Research in Organizational Behavior, 75–101. 201 Welter, F. (2012). All you need is trust? A critical review of the trust and entrepreneurship literature. International Small Business Journal: Researching Entrepreneurship, 30(3), 193–212. https://doi.org/10.1177/0266242612439588 Williamson, V., Stevelink, S. A. M., & Greenberg, N. (2018). Occupational moral injury and mental health: Systematic review and meta-analysis. British Journal of Psychiatry, 212(6), 339–346. https://doi.org/10.1192/bjp.2018.55 Wilson, S. J., & Lemoine, J. (2021). Methods of Calculating the Marginal Cost of Incarceration: A Scoping Review: Criminal Justice Policy Review, 1–25. https://doi.org/10.1177/08874034211060336 Winowiecki, E., Lafond, K., & Wells, K. (2018, January 24). TIMELINE: A long history of abuse by Dr. Larry Nassar | Michigan Radio. Michigan Radio. Retrieved from https://www.michiganradio.org/post/timeline-long-history-abuse-dr-larry-nassar Wolfensberger, M. (2016). A conceptual analysis of trust in medicine: its definition, decline, and significance (Keele University). Retrieved from https://eprints.keele.ac.uk/2377/ Yin, R. K. (2015). Qualitative research from start to finish (2nd ed.). https://doi.org/10.1080/09602011.2015.1126911 202