COMMUNICATION EFFECTS ON EXERCISE DURATION OF WEAKER GROUP MEMBERS WITH VIRTUALLY - PRESENT PARTNERS By Emery J. Max A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Kinesiology Master of Science 201 4 ii ABSTRACT COMMUNICATION EFFECTS ON EXERCISE DURATION OF WEAKER GROUP MEMBERS WITH VIRTUALLY - P RESENT PARTNERS By Emery J. Max The purpose of this study w as to examine the effects of encouragement on exercise duration within the conceptual framework of the Köhler motivation gain effect , which boosts t ask motivation for weaker group members in conjunctive tasks . Recent rese arch on exercise with virtually - present partners found that encouragement attenuated the Köhler effect (Irwin, Feltz, & Kerr, 2013 ) . The current study compare d he Köhler motivation gain effect. Female and male college students ( n = 240) were assigned to one of five conditions ( individual - control , individual - with - encouragement, partner - no - encouragement, partner - inclusive - encouragement, partner - exclusive - encouragement) and each performed two blocks of isometric abdominal plank exercises. A significant motivation gain was observed in all partnered conditions compared to the control, t (235) = 8.37, p < .001. Encouragement from a virtually present partner, regardless of inclusivity, did not moderate performance outcomes attributed to the Köhler effect. Encouragement in the absence of a partner altogether also boosted exercise motivation over the control group, but to a lesser degree than the Köhler effect, t (235) = 3.23, p = .001. These findings suggest that encouragement from a superior partner does not moderate the Köhler effect in exergames . I n games without a partner, an encouraging voice may be better than playing in silence. iii ACKNOWLEDGMENTS This thesis is not the product of one individual, working alone. I have had input, support, guidance, and helpful criticism from many. A special thank you to my advisor, Dr. Deborah Feltz , for your guidance, support, and patience throughout this project. Thank you to my committee members Dr. Gwen Wittenbaum, Dr. Daniel Gould, and Dr. Norbert Kerr for your guidance on this project. Thank you to Alison Ede, Samuel Forlenza, Dr. Kaitlynn Sedabres , Tayo Moss, Hannah Pilarski, an d Brendan Kennedy, whose assis tance was instrumental to the completion of this project. Finally, a thank you to my brother, for inspiring me to persist , and my family, for their unwavering support. iv TABLE OF CONTENTS LIST OF TABLES CHAPTER 1 I Primary Hypotheses Secondary Research Questions Definitions CHAPTER 2 REVIEW OF RELEVANT LITERATURE The Problem of Physical Inactivity Psychosocial Factors Influencing Exercise Participation Self - efficacy Extensions of self - efficacy Social influence Group Motiv ation Theories Motivation losses Motivation gains The Köhler effect Availability of partner - 2 2 ..24 The Köhler effect in exergames CHAPTER 3 ..36 METHOD Design and Participants Exercise Task Measures Self - efficacy v Post - Procedures CHAPTER 4 RESULTS Descriptive Statistics, Confound Checks and Manipulation Checks Preliminary Analyses Hypothesis Testing Ancillary Analyses Exercise self - efficacy Subjective effort Task evaluation Intention to exercise Perceptions of task ability Social comparison Social indispensab ility Team perceptions Communication effectiveness DISCUSSION Limitat ions Future Directions Conclusions APPENDICES AP PENDIX B SELF - APPENDIX F COMMUNIC APPENDIX J AUXILIARY vi REFERENCES vii LIST OF TABLES Table 1 Correlations Between Study Variables, Performance and RPE by Block 45 Table 2 Weighted values for planned contrasts of mean Block 2 Block 1 difference scores 50 Table 3 Bivariate correlations of primary dependent variables 82 Table 4 Bivariate correlations of responses to partner information questionnai re items 84 Table 5 Bivariate correlations of responses to communication questionnaire items 84 Table 6 Frequencies of noted potential confounds 85 Table 7 Means and standard deviations of all study variables by condition 86 Table 8 Means and standard deviations of all study variables by gender 87 viii LIST OF FIGURES Figure 1 Images of exercises performed 41 Figure 2 Mean duration (s) of performance difference scores (Block 2 Block 1) by condition . 52 Figure 3 88 1 CHAPTER 1 INTRODUCTION The American population is becoming increasingly sedenta ry and decr easingly healthy ( Barnett, Cerin, & Baranowski, 2011 ; Daley, 2009 ) . Less than 5% of U.S. adults adhere to the recommended guidelines for daily activity, and half of all persons who begin an exercise program drop out within 6 months ( Dishman & Buckworth, 1996 ; Trojano et al., 2008 ) . These changes have been accompanied by an increase in average time sp ent in front of a television screen ( Daley, 2009 ) . In response to decline s in health in the U.S. population , an ( exergames ) is growing that involve s kinetic, active involvement to get people on their feet and moving while allowing them to still enjoy an engaging gaming experience ( Barnett et al., 2011 ; Lyons et al., 2011 ) . The majority of the exergames on the market, however, have been reported to elicit light to moderate intensity energy - expenditure, which may not meet the American College of Sports Medicine recom mendations for daily activity ( Barnett et al., 2011 ; Biddiss & Irwin, 2010 ; Daley, 2009 ; Peng, Lin, & Crouse, 2011 ; White, Schofield, & Kilding, 2011 ) . Fortunately, newly released games are becoming increasingly metabolically challenging ( Daley, 2009 ; Lyons et al., 2011 ) . Though some of t he newly relea sed exer games may have the potential to be more physiologically challenging than earlier iterations, few are based on theoretical principles of group dynamics to boost motivation to continue to play the games . effectiveness, minimize parti cipant attrition from exergame programs , and encou rage more wide - spread participation, future exercise video games should consider the integrat ion of group dynamics principles of motivation into the games . 2 Research suggests tha t so cial support through partner and group exercise may facilitate e xercise participation ( Burke, 2006 ; Dishman & Buckworth, 1996 ) , but for those wit hout access to a fitness center or w ho suffer from social physique anxiety, group exercise may no t be an option. Furthermore, interdependent group exercise situations where progress and outcomes are mutually determined (group performance outcomes are determined by the interdependent effort of both partners) have yet to be tested in the real world, d espite research indicating that this may be a more effective method than traditi onal group programs ( Feltz, Irwin, & Kerr, 2012 ; Feltz, Kerr, & Irwin, 2011 ; Irwin, Scorniaenchi, Kerr, Eisenmann, & Feltz, 2012 ) . Thus, i ndividuals for whom exercise gami ng is the most appealing option, games that incorporate social dynamics that focus on social comparison, interdependence, and obligation may provide the most motivation. A substantial body of research has been conducted over the last 15 years that has focused on the motivation gains of performing a physically taxing task collaboratively in a group ( Hertel, Kerr, & Messé , 2000 ) . Motivation gains in which weaker group members exert greater effort at a task than they would were th Köhler named after a German industrial psychologist who first documented the phenomenon ( Köhler , 1926 ; Köhler , 1927 ) . Mo tivation and performance gains for weaker group members o ccur with the greatest magnitude when moderate discrepancies of ability exist between partners, when the tasks are conjunctive, and with contin u ous s ( Hertel et al., 2000 ; Kerr, Messé , Park, & Sambolec, 2005 ) . productivity is equal to the productivity of its least capable member ( Hertel et al., 2000 ) . That is , once the weaker member quits, the group can persist no longer. 3 Several recent studies ( Feltz et al., 2011 ; Gockel et al., 2008; Hertel et al., 2000 ; Kerr et al . , 2008; Irwin et al., 2013 ; Kerr et al., 2000; Kerr & Bruun, 2003; Kerr et al., 2007; Kerr et al., 2012; Kerr et al., 2013; Lount et al., 2000 ) have tested the Köhler effect ( and a meta - analysis ( Weber & Hertel, 2007 ) has documented its robustness (e.g., mean effect size d = .82). Additionally, a new program of research has demonstrated the utility of harnessing the K öhler effect to increase motivation in exergames and how the effect is moderated by task and features of a virtually presented partner ( Feltz et al., 2011 ) . This research has shown that exercising with a moderately superior partner led to a 24% improvement in effort in a series of abdominal plank exercises compared to exercising alone (Feltz et al. , 2011). Further studies showed that the Kö hler compared with slightly or extremely more capable partners ( Feltz et al., 2012 ) ; that the Köhler weight (i.e., heavier than the participant) ( Forlenza, Kerr, Irwin, & Feltz, 2012 ) ; and that there were no additional motiva tion gains from exercising with a moderately more capable part ner who provides encouragement ( Irwin, Feltz, & Kerr 2013 ) . Feltz and her colleagues used an exergame designed for the Play S tation 2 (PS2) gaming module for all of their studies. The software was EyeToy:Kinetic, a camer a - based game that offers a variety of fitness activities. The EyeToy abdominal plank exercises within the strength module were used, which minimized the import ance of motor skill and emphasized the effort of the task performance , since participants were instructed to hold each plank for as long as possible movements serve as the inter face to the games. Feltz and her colleagues adapted the game to 4 include a remote partner (confederate) who was presented virtually (e.g. visible over a 2 - way video hookup) in the partnered conditions. Most of the studies on the Köhler effect have restrict ed communication between teammates in order to control for extraneous effects. However, most exercising teammates do communicate with each other during the activity . In real group settings, words of encouragement are often exchanged between active partne rs (most typically from the more c apable to the less capable , such ) , but the effect o f such verbal encouragement is difficult to ascertain without controlling for content of the messages. Thus, in the study by Irwin et al. ( 2013 ) , researchers investigated the effect of verbal encouragement by the virtually - presented partner on the weaker group member. The researchers theorized that verbal encouragement could have positive or negative effects on motivation depending on how the wor ds are interpreted by the receiver. That is, the receiver of the encouragement (the weaker member) could perceive the encouragement (a) as supporting the receiver to do well, (b) as indicating the importance of the task to the encourager, or (c) as patron izing the weaker member, which would have a negative effect. Irwin et al. employed a 3 (conditions: individual, partner - without - encouragement, and partner - with - encouragement ) x 2 (performance block) factorial design with college - age participants . Particip ants performed a s eries of five abdominal planking exercises which they were instructed to sustain for as long as they could , first alone in Block 1, and then, for the partnered conditions, with a same - sex virtually - presented partner in Block 2. As with o ther studies by Feltz and her colleagues, the participants in the partnered conditions exercised conjunctively with a virtual partner who was presented as moderately superior. In the verbal encouragement condition, a pre - recorded series of phrases of enco uragement was played as coming from the virtually - presented partner every 5 15s (+/ - , , , good , , Results supported the performance enhancing Köhler e ffect, but contrary to the , v erbal encouragement mitigated effort gains (Irwin et al. , 2013) , contradicting both reasonable expectations and research on verbal encouragement in ind ividual exercise tasks ( Campenella et al., 2000; Guyatt et al., 1984; McNair et al., 1996 ; Moffat et al., 1994 ) . The researchers surmised that the decrease in effort in the verbal encouragement condition may have been due to the language used; the virtual when communicating with the pa rticipant, which may have been perceived as patronizing by the participant. This perception could decrease the appeal of the virtual partner, a problem that has been demonstrated to reduce gain s from the Köhler Effect ( Kerr & Seok, 2011 ; Kerr, Seok, Poulsen, Harris, & Messé, 2008 ) . effort by increasing the perceived gap of ability e.g. , ( Feltz et al., 2012 ; Köhler , 1926 ; 1927 ; Messé, Hertel, Kerr , Lount, & Park, 2002 ) . Irwin et al. offered, alte interpreted, not as teammate support, but rather as a method of self - encouragement by the partner . The authors offered that ing the message as self - encouragement might suggest to t he participant s that the supposedly superior partner was in fact struggling with the task, (p. 22). Unfortunately, the 6 be as ambiguous to the participant as in a team performanc e context. Inclusive pronouns have been shown to increase self - efficacy beliefs one of the most influential performance related psycholo gical constructs (Feltz , 1988), collective efficacy, a nd other performance indicators when used in self - talk exercise s of verbal persuasion ( Son, Jackson, Grove, & Feltz, 20 11 ) . interdependence , relational stability and longevity in dyadic relationships than exclusive ( Sillars, Shellen, McIntosh, & Pomegranate, 1997 ; Simmons, Gordon & Chambless, 2005 ) , an d use of plural pronouns may even increase feelings of interpersonal closeness (Fitzsimons & Kay, 2004) . To test whether exclusive language was indeed the factor mitigating the Köhler e ffect in Irwin 201 3 ) study, a partial replication and extens ion study with mo dified language w as conducted for the purpose of this thesis . The current study included another verbal encouragement condition that use d pronoun . The study also assessed perceptions of their partner. In addition, Irwin et al . (2011) also suggested that the feedback indicating the virtual the constant veridical feedback indicating that participants were being outperformed by their partner on Block 2 may have overridden or diluted any self - efficacy - boosting effect that positive encouragement may have had on the participant. In order to tes t the motivati o n and efficacy boosting effects of positive encouragement, without the confound of social comparison with a superior partner, we included a verbal encouragement condition without the virtual partner performance. This individual with encoura gement 7 condition used the There is literature that has shown positive performance effects from verbal persuasion ( Brown, 2003 ; G ould, Hodge, Peterson, & Giannini, 1989 ; Vargas - Tonsing, Myers, & Feltz, 2004 ; Wise & Trunnell, 2001 ) . According to Social Cognitive Theory, self - efficacy beliefs , and therefore motiva tion , are derived from several factors, one of which is verbal persuasion in the form of encouragement from trusted peers, authority figures, or oneself ( Bandura, 1977 ) . Primary Hypotheses This study test ed the following hypotheses: Hypothes is 1: Compared to working alone with no encouraging statements, participants will exercise longer when working together with a moderately superior virtual partner , regardless of encouragement, under conjunctive task demands. Hypothesis 2: Compared to worki ng together with a moderately superior virtual partner who provides no encouragement , participants will exercise longer when working with a partner who Hypothesis 3: Compared to working together with a moderately superior v irtual partner who provides no encouragement , participants will exercise longer when working with a partner who provides encour Hypothesis 4: Compared to working alo ne with no encouragement , participants will exercise longer when working alone with a trainer providing Secondary Research Questions In addition, the following exploratory research questions w ere addressed: Does task self - eff icacy change in tandem with motivation gains? Do ratings of perceived exertion (RPE) change in tandem with motivation gains? 8 Does intention to exercise change in tandem with motivation gains? Definitions Conjunctive Task: the group outcome is determined by the least capable member . Exclusive Language: l anguage that uses . Inclusive Language: Exergames : video games that require physical exertion in order to play Köhler Effect: motivation gains from working with a partner . Self - efficacy: succeed in specific situations . 9 CHAPTER 2 REVIEW OF RELEVANT LITERATURE The purpose of this chapter is to review the literature relevant to physical inactivity and means of its reduction . The chapter begins by addressing the health problems associated with physical inactivity. Next, the chapter delves into factors influencing exercise participation and adhe rence. This is followed by an exploration of group motivation theories including the Köhler effect . Finally, the chapter concludes with a review of the Köhler effect in exergames , including team communication and the use of language cues . The Problem o f Physical Inactivity Declines in physical activity, especially in developed countries such as the United States, have been documented repeatedly (Brownson, Boehmer, & Luke 2005 ; Dumith , Hallal, Reis & Kohl, 2011 ; U.S. Department of Health and Human Services, 1996 ). While the importance of this change may seem obvious, little change has occurred in actual activity levels (Brownson et al. , 2005). In a culture that prizes technological conveniences, physical activity is less necessary and people becom e less active . However, in the case of physical inactivity, conveniences now may often result in severe health repercussions later. Physical inactivity is known to lead to the accumulation of body fat: low caloric e xpenditure without concomitant low cal oric intake results in a caloric surplus that the human body stores in adipose cells. Over time, the excessive storage can accumulate and hinder both bodily movement and function. As essentially non - functional (non - contractile, non - support ive) tissue, ex cessive body fat is potentially burden some on the bodily organs and suppor ting structures ( Montani et al. 2004; Navina , 2011). Many comorbidities of obesity, including stroke, cardiovascular disease, some types of cancer, and diabetes type II, fall high on the list of leading 10 killers in America (Wen & Wu, 2012). While physical inactivity may be a clear logical precursor of obesity and many of these illnesses, some research has determined that physical in activity alone, regardless of any other habits and a variety of physical traits including fatness, may be enough to increase risk of mortality ( Bellocco et al., 2010 ; Esteghamati et al., 2012 ; Haapanen - Niemi et al., 2000 ; Patrick et al., 2005 ; Wei et al., 1999). Of course, the problem extends beyond the scope of the individual. Nearly 5 million deaths in the world each year have been attributed to physical inactivity - roughly the same as that of tobacco ciga rettes ( Wen & Wu, 2012 ). If physical inactivity were decreased by as little as 25%, an estimated 1.3 million premature deaths per year could be averted and the average life expectancy would increase by nearly a year (Lee et al., 2012). It is estimated th at in the United States alone, the economic burden of physical inactivity exceeds $23.7 billion per year and the amount continues to grow ( Kohl et al., 2012; Wang , Pratt, Macera, Zheng & Heath, 2004). While this information may persuade an individual to begin an exercise program , the current guidelines for minimum daily physical activity remain daunting for many. Thankfully, mortality has been shown to decrease by 14% with just 15 minutes of physical activity per day and 4% decreases with each addi tional 15 minutes beyond that (Wen et al., 2011). Of course , even with an attainable minimum, a key obstacle remains: can the public be convinced to actually try to exercise ? While there are myriad factors influencing exercise participation, some of the m ost powerful are psychosocial in nature. Psychosocial Factors Influencing Exercise Participation The most prevalent psychosocial factors that can potentially influence exercise participation include perceived self - efficacy, perceived social support , and behavioral intentions. 11 These factors are based primarily on social - cognitive theories and principles of group dynamics. Each set of factors is elaborated in the paragraphs that follow. Self - efficacy. Bandura (1977) defined self - s beliefs that they have the - efficacy beliefs ultimately determined whether or not one would engage in a behavior and persist in the face of adversity (1978). Self - efficacy was determined to be derived from four sources: personal mastery experiences, vicarious experiences through observing others, verbal persuasion from trusted others, and physiological arousal level. Mastery experiences were proposed to serv - that is, if one had performed the task before, the individual was likely to believe he or she could perform the task again. Vicarious experiences were theorized to operate similarly to mastery experiences albeit t o a lesser degree. By observing another person complete the task, it became easier to envision oneself performing the task. Verbal persuasion, or the encouragement from a trusted source or authority figure, was hypothesized to boost self - efficacy. Again , the effect of this route was expected to be weaker than mastery experiences and was believed to be contingent on the credibility of the persuader. Finally, Bandura predicted a relationship between self - efficacy and physiological arousal. An s interpretation of his or her own physiological changes (i.e. , quickened heartbeat, palmar sweating, fatigue, etc.) as signs of fear or inadequacy could decrease self - efficacy ; whereas , perceptions of ability could increase self - efficacy beliefs. The concept of self - efficacy has been applied within the context of sport and exercise model of self - efficacy and an anxiety - based model of avoidance behavior on a high - avoidance 12 back - diving task and found that self - efficacy was a significant predictor of approach behavior to the task. Self - efficacy also has been shown to predict effort on muscular endurance tasks (Weinberg et al., 1979; Weinberg, Gould, Yukelson , & Jackson, 1981; Weinberg. Yukelson, & Jackson, 1980). In a rigged competitive leg - (1979) study were randomly assigned to either a high or low self - efficacy condition, where participant self - efficacy was manipulated by identifying the confederate competitor as either a varsity track athlete who performed well on a related task (low self - efficacy condition) or an individual with a knee injury who had performed poorly on a related task (high self - efficacy). Despite both losing to the confederate, individuals in the high self - efficacy condition persisted significantly longer in trials when compared to individuals in the low self - efficacy condition who osed relationship between self - efficacy and performance. In terms of exercise behavior, McAuley, Lox, and Duncan (1993) found self - efficacy to strongly predict exercise adherence in older adults. The study was performed on middle - aged, sedentary adults who, 9 months prior, had participated in a 5 month structured exercise program. Self - efficacy measurements pertaining to physical activity (3 specific exercises) and adherence were collected at four points: before and after graded exercise tests first at the end of the 5 - month program and then at the 9 - month follow - up. Though self - efficacy beliefs at the end of the 5 - month intervention predicted exercise adherence in the following 9 months, those beliefs tended to decline over time, as evidenced by signif icantly lower self - efficacy measures at the follow - up p ost - test. Interestingly, participation in the exercise test at the 9 - month follow - up boosted self - efficacy to levels comparable to those reported at the end of the 5 - month intervention. These 13 finding s suggest that the relationship between self - efficacy and exercise behavior is reciprocal and dynamic, in that while self - efficacy predicts exercise behavior over time, acute behavior (e.g., mastery experiences providing updated performance information) ca n also modify self - efficacy. Thus, self - efficacy has been shown to be an effective motivating factor in sport and exercise contexts as it relates to the initiation of a task, the persistence of effort at muscular - endurance tasks, and the adherence of exer cise behavior over time. Extensions of self - efficacy . Integrations of self - efficacy with other related theories as a determinant of performance have been noted (Bandura, 1982). Cognitive Evaluation Theory erception of competence (a similar concept to self - efficacy) along with attributions of performances to internal causes and a sense of relatedness to others will facilitate intrinsic motivation, which may in turn enhance the enjoyment derived from the acti vity ( Deci & Ryan, 1985 ; Ryan & Deci, 2000 ). Thus, this thesis also examined enjoyment of the exercise activity. Initiating the activity, however, required a difference explanation. The Theory of Reasoned Action, developed by Fishbein and Azjen ( 1975 ) , stated that behavior was determined by behavioral intentions, which were in turn determined by attitudes about the behavior and subjective norms about the behavior ( Ajzen & Fishbein, 1977; 1980 ; Fishbein & Ajzen , 1975). Intention has been identified, along with past behavior, as a powerful predictor of future behavior (Oulette & Wood, 1998 ). However, behavioral intention means little without belief in behavioral ability . Self - efficacy was added to the Theory of Reasoned Action as an agency - enabling me chanism to create the Theory of Planned Behavior (Ajzen 1991 ) , where engagement in behavior was predicted to be highly contingent on perceived behavioral control , which may take 14 the form of self - efficacy beliefs. Research on i ntentions specific to exercise behavior has provided additional support for the Theory of Planned Behavior (Hagger, Chatzirsarantis & Biddle, 2002 ) . A study by Downs, Graham, Yang, Bargainnier & Vasil (2006) investigated exercise intention and behavior in a longitudinal study on adolescents. Data w ere collected on past exercise behavior, attitudes, subjective norms, perceived behavioral control and intention pertaining to exercise behaviors. Intentio n was best predicted by perceived behavioral control and most predictive of past exercise behavior. In a study by Dzewaltowski, Noble and Shaw ( 1990), exercise intention in college age individuals was best predicted by perceived beha vioral control and att itudes and strongly predictive of actual exercise behavior . The degree to which intention predicted behavior was only exceeded by self - efficacy perceptions. Even so, the importance of behavioral intention and its precursors especially subjective social components - ought not to be discounted. Social influence . Within the Theory of Planned Behavior model, subjective norms, or individuals perceived attitude of their cohort regarding a behavior, is believed to be a predictor of behavioral intention. That is, if one perceives that others maintain positive attitudes and expectations of behavioral action, one is more likely to form intentions to engage in that behavior. Little support has been found for this relationship ( Courneya, Nigg & Estabrook, 200 0 ; Godin & Kok, 1996) , however, and consequently researchers have argued that p erceived social support , which can be viewed as a form of verbal persuasion within self - efficacy theory, may play a more significant role (Courneya & McAuley 1995a; 1995b; Courneya et al., 2000). In a study by Rhodes, Jones and Courneya (2002) , 192 undergraduate psychology students completed surveys in large groups pertaining to the theory planned behavior, social support in exercise, and exercise habits. A 2 week follow up indicated that while subjective norms 15 contributed to exercise participation, social support was found to be independently predictive of both intention to exercise and perceived behavioral control. These findings suggest that social support may play a sig nificant role in behavioral outcomes through two processes: reinforcing both intention and perceived self - efficacy. Indeed, humans are highly social creatures; existing and thriving within groups. The effect of social context on motivation changes in gro ups has been duly documented in a number of studies, examined below, on effort, motivation, and productivity in working groups. Group Motivation Theor ies Motivation in groups has been studied from a number of perspectives. Much of the literature has focuse d on motivation losses, or social loafing (Everett, Smith, & Williams, 1992; Karau & Williams, 1993; Williams & Karau, 1991; Williams, Nida, Baca, & Latané, 1989). However, recent research has focused on motivation gains in which performance increases with in a group setting compared to individual performance (e.g., Weber & Hertel, 2007). Researchers have identified possible causes for both concepts. Motivation losses. In 1913, Max Ringelmann observed an inverse relationship between group size and group pro ductivity in a rope - pulling task: individuals exerted less force on the rope when working as a team than when pulling alone. He proposed two explanations: coordination losses, where the group was simply less efficient at the task than individuals, and mot ivation losses - where the individuals relied on their coworkers t o exert the necessary effort to achieve the task and consequently exerted less effort themselves (cited in Kravitz & Martin, 1986). 1972) taxonomy of group tasks, coordination losses in a unitary (unable to be divided into subtasks) , additive (task outcome is the sum of individual inputs) , and 16 maximizing (outcome is dependent on quantity rather than quality) task s, such as that used in the Ringelmann experiment where potential productivity is equal to the summation of individual efforts - are an unlikely explanation for performance decrements. In a 1974 replic ation of able to isolate coordination with the use of - groups , blindfolded and performed the task with confederate teammates who refrained from actually pulling the rope . Individuals in the group condition expended le ss effort than when working alone. T he false knowledge of a supportive team was enough to reduce the output of an individual rope - puller (Ingham , Levinger, Graves & Peckham, 1974). Consequently, motivation losses appear to be a more plausible explanation for performance decrements than coordination losses. In a similar study, albeit with clapping and shouting instead of rope - pulling, Latané , Williams and Harkins (1979) found further evidence of social loafing. In their first experiment, the researcher s instructed participants to shout as and clap as loudly as they could, both alone and as a member of a group. The groups produced markedly lower decibel levels than would be expected: the noise intensity increased from the individual condition , but not i n proportion to the addition of group members. To determine whether the performance losses were due to a decrease in efficiency or a decrease in effort, the subjects were blindfolded and instructed to wear sound - dampening earmuffs. As with the first expe riment, subjects failed to produce decibel level increases that would be expected with the addition of new noise sources. The rese archers attributed these performance decrements to social loafing and proposed several possible explanations but ultimately explained social loafing with the Social Impact Theory . This theory essentially states that the impact of a social force is positively related to the ratio of sources (influential agents) to targets (receiving agents). That is, social impact increases when 17 the target number is fixed but source number is increased; whereas social impact decreases when the source number is fixed and the target size increases, such as with group membership ( Latané 1981). Thus, i n a performance situation where individual outputs are not identifiable, the resultant decrease in effort and output is logical ( Latané et al., 1979). In addition to lack of identifiability as a possible cause for decreases in performance in a group enviro nment, Baron and Kerr (2003) also suggested that group members may recognize that in some instances they may be able to free - contribute to what they perceive to be more than thei r fair share of the collective effort. A m eta - analysi s of 78 studies d emonstrated social loafing to be a robust effect across several task domains and effort modalities (Karau & Williams , 1993). Several key elements were shown to consistently affect motivation in group contexts, namely: evaluation, task meaningfulness, group member familiarity, expectations of co - worker ability, and dispensability. The researchers determined that social loafing is reduced when individual performances within the group are readily identifiable to all involved, there is a moderate to high degree of personal involvement, when there are group - level comparison standard s available, when individuals know or value their co - workers, when individuals do not expect their group to perform exceedingly well, and when individual contributions are seen as crucial to group outcome (Karau & Williams 1993) . By increasing identifiability of group member performance to hide in the crowd (Davis, 1 969) is reduced, thereby reducing the likelihood of sub - par performances. Identifiability is also important to promote recognition for honest efforts to avoid feeling lost in the crowd ( Latané et al., 1979), so group member contributions are seen as valua ble and not doomed to be unrecognized and unrewarded. When contributions are not identifiable, i.e. , indispensability) countervails performance 18 losses (Harkins & Petty, 1982). Unfortunately, in many colle ctive tasks, individual performance data are not always readily identifiable or even easy to determine. Situations with highly cohesive groups - loaf because they know they ca n profit from the work of others without working t hemselves (Baron & Kerr 2003). Thus, while the nature of some collective tasks may undermine m otivation, when there are social pressures (e.g . , when individuals may be held accountable for their output and group comparison information is available , when relationships produce a sense of obligation to the team, and when there are feelings of indispensability ) and internal drives to perform (e.g., when involvement is high) , moti vation decrements may be avoided. Motivation gains . While motivation losses in groups have been explored extensively , not all groups are consigned to diminishing performance returns. In the past 50 years , the field of group motivational processes has seen an accumulation of empirical research on the possibility of performance gains in group contexts . While competition is an important factor in some performance situations, cooperative groups such as those in industry (i.e. , sales teams, factory lines, military units) and ath letics (i.e. , sports teams) are more prevalent . Of the social effects within these contexts , which produce motivation gains rather than decrements, the two most frequently cited are the social compensation effect and the Köhler effect. Social compensation is the tendency for people to exceed individual performance expectations on a collective task to compensate for a weaker member ( Williams & Karau , 1991). In a three - study series, Williams and Karau investigated the extent to which expectations of co - worker performance and personal involvement moderated motivation changes in group tasks. In t he first experiment, individual tendency to trust others predicted their reliance on co - workers as 19 evidenced by loafing behavior on collective tasks (tasks requiring unique contributions from each member) but not on coactive ones (tasks completed in the p resence of another) . In the second two experiments, high co - worker ability or intended effort mitigated compensatory behavior from participants in collective tasks but not coactive ones. Perhaps most importantly, loafing prevailed over compensation rega rdless of co - worker output if the task was not valued by the participant. Thus, individuals who saw co - workers as unreliable, unwilling, or unable to fulfill their duty in a collective task were likely to work harder to ensure the end goal was met but o nly when they were genuinely involved in the task or the outcome held personal relevance ( Williams & Karau, 1991). The authors inferred from these findings that social compensation occurs as a result of concerns regarding reputation is on the line - as it is in a collective task with high meaningfulness where the outcome is directly reflective of the performers (e.g., highly competitive sports teams) more capable and willing individuals will contribute in excess of their part to ensure that the group goal is met. The Köhler effect. In contrast t o social compensation, where strong er individuals pick up the slack of weaker group members, there are instances where weaker group members will exceed performance expectations under collecti ve task demands. In the late 1920s, German industrial psychologist Otto Köhler first observed this phenomenon while studying male members of a Berlin rowing club. Köhler (1926) devised an experim ent in which participants curled a weight for maximal repetitions individually , paced by a metronome, and then performed the same task again albeit yoked to a partner with twice the weight. While persistence was the objective in both individual and group tasks, the rules for each slightly differed. The dyadic task called for cessation of the activity Surprisingly, calculations of the 20 group potential, based on observed individual performances from preliminary trials, underestimated actual group performance . Actual group performance exceeded not only the performance of the weakest member but also the av erage of the two members - an effect which was replicated in a follow - up study with winch - winding ( Köhler , 1927) , but was moderated by the discrepancy in ability between partners (St roebe, Diehl, & Abakoumkin, 1996; Witte , 1989). Köhler e highest group productivity levels when the ratio of partner ability fell within the range of 3:5 to 4:5, with optimal performances occurring at a ratio of 3: 4. Upward social comparison. Subsequent experiments obtained similar results, confirming that m oderate ability discrepancy is a prerequisite for the effect because it encourages up ward social comparison one of the two key psychological mechanisms underlying the Köhler motivation gain ( Plante & Madden, 2010; cf. Seta, 1982; Stroebe, Diehl, & Abako umkin, 1996. ). Motivation gains derived from upward social comparison may be due to an elevation of the personal pe rformance goal as a result of a new performance standard or, perhap s, increased goal saliency through competition, with the objective being to outperform partner (Kerr et al. 2005). Regardless of whether group members view their partner as chase members and their stronger counterparts gen erally serve to boost motivation in weaker group members. Availability of partner - related information. Partner ability discrepancy is not the only factor determining whether or not upward social comparison will occur, however. The availability of partner - related information namely, ability and contribution may moderate the Köhler effect. Partner abili ty needs to be known prior to working to elicit a discrepancy effect ( Messé et al., 2002), and c onsistent performance updates facilitate social comparison by 21 continually reminding weaker members of their relative status in the task. An increase in the num ber of opportunities to compare (with consistent feedback) may cause greater motivation gains than in the conditions when either no feedback is provided or feedback is delayed until after trials ( Hertel et al., 2008), but lack of feedback does not preclude the Köhler effect entirely it merely attenuates it (Kerr et al., 2005) . Part of this appears to be due to recency or frequency of partner performance updates . I n Hertel an d study (2008), a 20 minute task was employed which found that if fee dback was not promised until after the trial, motivation gains seen with continuous feedback failed to surface. Motivation gains remain intact , for the most part, when partner performance is fresh on the mind of the participants - even if delayed until af ter trials, as demonstrated in the results from a study by Kerr and colleagues (2005), which took only 2 - 3 min. per trial. Identifiability . Though performance feedback undoubtedly serves as a vector for upward social comparison, it also serves to increase group member identifiability. That is, participants their partner knows of their contributions to the team outcome . The importance of identifiability of group member input was outlined earlier within the context of social loafing, and it applies to the Köhler effect as we ll. Unless highly involved with the task or performing an act of altruism, an individual has little reason to fully exert himself if he knows his own contribution (or lack thereof) will go unnoticed. While its ability to moderate upward social comparison is evident, availability of Social indispensability. I n addition to upward social comparison, the Köhler effec t is strongest when a task is constructed in such a fashion that it incurs feelings of indispensability within an individual in regard to her contribution to the group task outcome ( Kerr & Bruun, 22 1983). According to Instrumentality x V al ue models of motivation (Karau, Williams, Bourgeois, Carlston & Eagly, 1993; Shepperd , 19 93 ) , individual motivation in a collective task is contingent upon the degree to which that individual values a positive outc ome on the task and the importance of her contribution to the task to achieve that outcome. Thus, collective tasks where positive outcomes necessitate an earnest individual contribution will likely increase individual motivation . The conditions under whi ch Köhler first observed his effect fulfilled this criterion: the nature of Köhler determined by the weakest individual performance, conveyed a high sense of instrumentality in the weake r group member. One way to determine if the task design is indeed as crucial as 1972) . Task structure. In an experiment to test this prediction, Hertel, Kerr, Scheffler et al. (2000) compared the individual persistence alone and in groups under a variety of task demands, ones in which there was high instrumentality (conjunctive), and low instrumentality (additive). Participants performed an endurance task where they were instructed to hold a weight in an extended arm over a tripwire alongside a partner doing the same thing. In the conjunctive condition, the team score was determined by the performance of the weaker member (i.e. when the weaker member tripped th e wire , the other had to stop ). In the additive condition, the stronger member was allowed to persist for as long as possible with the team score being determine d by the sum of member performances. Individuals in the conjunctive condition showed a robust motivation gain (45.7s), while the individuals in the additive condition showed significant , albeit less robust, change . Because both tasks allowed for upward social comparison 23 (weaker members in both groups could see their more capable partner), the dif ference in performance was attributed entirely to indispensability of the weaker group member. Support for the instrumentality hypothesis was found again in computer - supported groups shortly thereafter (Herte l , Deter & Konradt 2003). Participants were instructed to assemble computer hardware packages for customer requests in such a way as to maximize sales first alone, and then, if in one of the two group conditions, as part of an internet - connected group for the second round. Participants in all thr ee conditions were promised a reward for correct assemblages, but the two group conditions had different stipulations . One group condition was additive, where if one member assembled a correct package for a customer , a reward was earned for the team, and the other group condition was conjunctive, where both team members had to assemble the package correctly in order for the reward to be disbursed , thus limiting the team productivity to the ou t put of the weaker member. After each trial, participants in all groups completed a survey of their feelings with respect to instrumentality, effort, and enjoyment. Consistent with findings in other domains, actual group productivity and perceptions of individual effort were greatest in the conjunctive task condition, lending further credibility to the instrumentality hypothesis and highlighting the importance of task structure ( Gockel, Kerr, Seok & Harris, 2008 ; Kerr et al., 2007 ; ) . Group performance did not exceed the actual potential, but only the predicted potenti al based on individual performances. It is important to note, then, that the conjunctive task paradigm does not cause individuals to exceed their predicted capabilities, but instead to realize their actual potential. There are several plausible explanat ions for why additive task structure, which seems to offer the most promising potential for increasing total group output, fails to realize motivation gains like those seen under conjunctive task constraints . One such explanation is that an additive 24 task - superior effort or that their own effort contributes little to the team output, - Kerr & Bruun , 2003 ; Williams & Karau, 1991 ). In the case of work under additive task constraints, where the reward is a financial incentive contingent on a relatively simple task and the more capable member is motivated to conti nue for his or her own sake, there is also a possibility of social compensation unless, however, the lower - output member is in fact capable. Should the more capable team member realize her efforts are being capitalized upon by an opportunistic but capab 2003). Evaluation concerns. Both upward social comparison and fe elings of indispensability (operating through consequent feelings of obligation) are strongest when group members are physically present , a phenomenon which has been attributed to evaluation concerns (i.e. , concerns stemming from the potential of negative evaluation of others) (Lount et al., 2008 ) . Lount and colleagues investigated the degree to which mutual observation of coworkers performing collective tasks (either physically present or virtually present ) moderated the Köhler effect. In a task very similar to the one used in Hertel, Kerr and Scheffler ( 2000 ) experiments examining the Köhler effect, participants suspended a weight over a tripwire either side by side with a partner or concurrently with a p artner in another room, visible through a video feed. Consistent with predictions, the greatest motivations gains were seen when coworkers were physically present. Virtually present coworkers induced motivation gains greater than the control condition (a n individual persistence task), but those gains paled in comparison to the 25 physically present coworker. These findings were consistent both with Collective Effort Model (Karau & Williams, 1993), which postulates that evaluation potential will predict moti vation gains, as well as Social Imp act Theory ( Latané et al., 1979 ), which suggests that social forces will be experienced at their greatest power when proximity between force and target is reduced, and virtual presence increases proximity and dilutes thos e forces. Resultant decreases in performance and motivation are likely due, at least in part, to a decline in self - presentation concerns over pending evaluation from others (Carron, Burke & Prapavessis 2004). Gender composition. One potential moderator of evaluation concerns and social he majority of research on the Köhler effect has been restricted to same - gender teams, but the interaction of normative gender roles and task demands could p - gender team task. In an effort to investigate the degree to which group gender composition could moderate the Köhler effect, Lount, Messé and Kerr (2000) performed a study on 95 college students who com pleted four (two per arm) endurance exercises where they suspended their arm with a wrist - weight over a tripwire. All participants performed the first two exercises alone, and then were split in to one of three conditions: individual control, where they p erformed the second block alone, conjunctive same - gender , where they performed the task conjunctively with a same - sex confederate, or conjunctive opposite - gender , where they performed the task conjunctively with an opposite - gender confederate. Males and f emales used different weights so that task difficulty was relatively similar for both genders . This controlled for participant instrumentality, thereby restricting performance differences to social comparison processes associated with any potential gender differences. Results aligned with hypotheses, where both males and females received motivation gains from working with a partner in a conjunctive task, males working with female 26 teammates exhibited even greater motivation gains than males with male teamm ates, and female performances with male teammates demonstrated a high degree of variation. Th e researchers surmised that these differences would be due to normative beliefs of gender - expectations and the . Thus, males would be most motivated when working with an opposite - gender partner in order to fulfill his normative role, lest he be outperformed by a female partner and la who received additional motivation from working with a male partner were hypothesized to do so because they may have wished to defy their gender stereotype, while females who did not were hypothesized to do so becaus e they wished to adhere to normative expectations. Unfortunately, the key measure the researchers had intended to use to distinguish between individuals who subscribed to normative gender expe ctations and those who did not - th e Bem Sex - Role Inventory, 19 74 - measured self - evaluation of traits rather than broad beliefs on the gender roles. As su ch, little other than performance differences were left to be analyzed, and a variety of ex planations were given for them, including variation in the social comparison proces s and self - presentation concerns (Lount et al., 2000). Group identity. Because the strength of the Köhler effect hinges on feelings of indispensability, consideration of the conditions under which in dispensability arises (i.e. , under conjunctive task demands) and under which it matters (i.e. , when there is a personal investment the latter is the degree to which an individual identifies with his or her group (Kerr et al., 2008) . Group cohesiveness, or the strength of the bond teammates have and their familiarity with one another, was demonstrated to facilitate motivation in collective tasks (Karau & Williams, 1997), but much of the research on the Köhler effect involved ad hoc groups where the relationship and 27 identity was limited to the shared goal and interdependence arising from the task structur e. An e arly study on ostracism (Geller, Goodstein, Silver & Sternberg, 1974) suggested that individuals were unlikely to work hard if their payoffs in any way benefitted an individual who excluded them. To test how social exclusion could moderate the Köhler effect, Kerr and colleagues (2008) divided parti cipants into three conditions each assigned to perform a persistence arm - lifting task one control, who worked alone, one conjunctive, and one coactive. Participants in group conditions either proceeded from the first to second block of exercise with no intervention , or they participated in an electronic task that facilitated feelings of either exclusion or inclusion. Results from the study supported their hypotheses: ostracism attenuated the Köhler effect in conjunctive conditions, where participant exe rtion could benefit their ostracizing partner s , but made little difference in the coactive condition where individual performances were independent of one another. The researchers concluded that social exclusion undermines group identity and, accordingly , hinders the indispensability mechanism by decreasing feelings of obligation, but has no e ffect on social comparison processes (Kerr et al., 2008). The Köhler effect in exergames. Köhler l study was on athletes performing an exercise task. Taking note of this and the subsequent research on the Köhler effect with virtually present teammates, resea rchers Feltz et al. (2011) identified the potential for the Köhler effect as one way to increase the effectiveness of exergame s . They designed a study where 181 undergraduates were divided into four work conditions (individual control, coactive, additive, and conjunctive) and each performed two rounds of an exergame first alone, and then, if in one of the partnered conditions, with a virtually present partner designed for the PlayStation 2 gaming console. The exercise blocks consisted of five variations of an abdominal planking task and the user interface 28 showed a live webcam feed of the participant side - by - side with a software - g enerated trainer , who modeled the exercises . In the partnered conditions, the second block of exercise s appeared the ther exercise r (the teammate, a pre - recorded confederat e) was displayed alongside it. Contrary to previous findings on the Köhler effect, motivation gains among coactive, additive, and conjunctive conditions were powerful but indistinguishable from one an other. The researchers attributed the atypical pattern of motivation gains to either the inherently competitive nature of an exergame or the demotivating potential of the extrinsic reward (i.e. , money) offered to successful teams (Feltz et al., 2011) . Extrinsic incentives . To test whether extrinsic incentives played a role, the same research team (Kerr, Feltz & Irwin, 2012) performed a similarly designed study with two blocks of abdominal planking exercises, albeit with only two task designs (individual control and conjunctive) and the added variable of the presence or absence of an extrinsic reward. Data from the two extrinsic reward conditions were borrowed from the initial 2011 study and compared with new data collected with the same protocol (for both the individual and conjunctive conditions) where an extrinsic reward was not offered. When working alone, individuals offered a rew ard responded similarly to those who were not, but conjunctive teams who were offered no financial reward persisted 43% longer their rewarded peers . The researchers noted several plausible explanations: extrinsic incentives could have undermined the soci al comparison process if participants saw partner performances not as reflections of ability but as reflections of partner desire for the reward, or extrinsic incentives could have decreased the sense of obligation to a partner if participants saw their pe rformance as essential not solely for social reward, as with the non - incentive conjunctive condition, but as essential for a financial reward, which may or 29 may not have been valued by the participant (Kerr et al., 2012). While these findings are important to further our understanding of the Köhler effect and incentives for performing exergames, it seems unlikely that rewards such as those seen in most video games (tokens meaningless in the real world) would have a negative impact on the Köhler effect as si gnificant as the one demonstrated by Kerr and his coworkers. Still, the study serves to underline the importance of upward social comparison untainted by motive - questioning of a superior partner and a sense of indispensability that leads to obligational m otivation and personal investment in the team outcome. Perpetual inferiority. Of course, when designing an exergame, one must consider the long term viability motivation gains, while robust and consistent, are meaningless in an exergame context if a p layer discontinues use after just a few sessions. The effectiveness of the Köhler effect rests on ability discrepancy (to create upward social comparison) and indispensability, but perpetual inferiority presents a potential problem for ongoing exercise pa rticipation. Previous studies on the Köhler effect in exergames had examined the effect on a one - time persistence abdominal planking task, but for long - term health changes, repeated aerobic exercise bouts offer greater potential. As alluded to in the Fel tz et al. (2011) study, the researchers ( Irwin et al., 2012) divided participants into three conditions - individual control or a partnered condition (coactive or conjunctive) where they cycled with a superior virtually present partner for 6 separate days. Not only was there strong support for the Köhler effect in the conjunctive condition in the first trial, but performance in the conjunctive condition actually increased over the six trials - a remarkable finding made more remarkable when compared to moderate performance declines in the coactive and individual conditions ( Irwin et al., 2012 ). The irrelevance of perpetual partner superiority over time with an aerobic task was also 30 supported by a study that investigated perpetual vs. intermi ttent partner superiority across two isometric task domains (Kerr et al., 2013). Partner characteristics. Lastly, when considering implementation of the Köhler effect into exergame design, one must examine the characteristics of the partner (in additio n to gender cha ra cteristics) , which may affect either of the two functional mechanisms. Two factors which may serve to convey ability and may be relevant to social comparison are age and weight. Forlenza et al. (2012), utilizing a task design similar to the original Feltz et al. (2011) study, had participants perform a series of abdominal planking task s over two exercise blocks either alone or with a partner who was either similar or dissimilar in age (college age or 48 years old) and similar or dissimila r in weight (average weight or obese). Unexpectedly, the Köhler effect was unmoderated by age , and males worked even harder when paired with an obese partner, suggesting that, in the realm of exergames, if partner dissimilarity do es not impinge upon the social comparison processes or remove feelings of indispensability, the Köhler effect will persist . Team communication. Appearance, however, is only one facet of partner - player dynamic in the Köhler - Exergame research series . Thoug h verbal communicat ion between partners was limited in most of the Köhler research, there are a number of studies that have shown verbal encouragement to increase performance in exercise tasks when used independently (Campenella, Mattacola & Kimur a, 2000; Guyatt et al., 1984; McNair, Depledge & Stanley, 1996; Moffat, Chitwood & Biggerstaff, 1994) . Additionally, as stated in Chapter 1, most exercising teammates communicate with each other during the activity. In real group settings, words of encouragement are often exchanged between active partners (most typically from the , ntrolling 31 for content of the messages. Thus, in the study by Irwin et al. (2013) , researchers investigated the effect of verbal encouragement by the virtually - presented partner on the weaker group member. The researchers theorized that verbal encouragement could have positive or negative effects on motivation depending on how the wor ds are interpreted by the receiver. That is, the receiver of the encouragement (the weaker member) could perceive the encouragement (a) as supporting the receiver to do well, (b) as indicating the importance of the task to the encourager, or (c) as patron izing the weaker member, which would have a negative effect. Irwin et al. employed a 3 (conditions: individual, partner - without - encouragement, and partner - with - encouragement) x 2 (performance block) factorial design college - age participants. Participants performed a series of five abdominal planking exercises which they were instructed to sustain for as long as they could, first alone in Block 1, and then, for the partnered conditions, with a same - sex virtually - presented partner in Block 2. As with other studies by Feltz and her colleagues, the participants in the partnered conditions exercised conjunctively with a virtual partner who was presented as moderately superior. In the verbal encouragement condition, a pre - recorded series of phrases of encourage ment was played as coming from the virtually - presented partner every 15s (+/ - 3 s) , , , , strong here , Results supported the performance enhancing Köhler effect, but contrary to expectations, verbal encouragement mitigated effort gains (Irwin et al., 2013). The researchers surmised that the decrease in effort in the verbal encouragement condition may have been due to the language which may have been perceived as patronizing by the participant. This perception could decrease the appeal of the virtual partner, a problem which has been demonstrated to reduce 32 gains from the Köhler Effect ( Kerr & Seok, 2011 ; Kerr et al., 2008 ) have a negative impact on effort by increasing the perceived gap of ability (Feltz et al., 2012; Köhler , 1926; Köhler , 1927; Messé et al., 2002). could be interpret ed, not as teammate support, but rather as a method of self - encouragement. - encouragement might suggest to the participants that the supposedly superior partner was in fact struggling with the task , thereby Due to the coactive and relational dynamic of the Köhler effect, communication research in interpersonal relationships offers some insight to illuminate the unusual results of the Irwin et al. (2013) study. Communication among group members should enhance collective effort when it enhances perceptions of task importance or soci Karau & Williams , 1993 , p. 702). An increase in team identity could foster feelings of social responsibility and subsequent obligational motivation gains to enhance t he team outcome. As noted before, most groups in the Köhler [who] share a common fate [and] exhibit structured patterns of interaction and modes of communication (Carron, Hausenblas & Eys , 1998, pp. 13 14). In the context of the present study, which hypothesizes that when a superior partner communicates inclusive encouragement Köhler effect will be strengthened by 33 establishing and reinforcing a group identity and consequen t feelings of obligation enacted through the indispensability mechanism. Language cues. Strength of group identity and cohesiveness is known to be positively related to individual effort in group tasks ( Karau & Williams, 1997; Kerr et al., 2008; Worchel , 1998 ) . Though most real groups identify as such for a variety of reasons such as a common fate, shared goals or other similarities , it is possible to induce feelings of closeness and group identity through nothing more than subtle manipulations of languag e ( Brewer & Gardner, 1996 ; Fitzsimons & Kay, 2004). Language has special and unique power in the social world in that the labels it provides determine and perpetuate the perception s of an individual , a group, or the self , simultaneously describing qualities and ascribing values. One way to elicit th e positive associations of a group identity in the absence of a true group is to merely prime individuals with collectiv e language ( Brewer & Gardner, 1996; Perdue, Dovidio, Gurtman & Tyler, 1990 ). To d emonstrate this phenomenon, researchers Brewer and Gardner (1996) performed a three study series on the effects of pronoun priming on perceived attitudinal similarity when reading ambiguous statements and, more notably, the production of social self - descri ptions and declarations of group membership and a collective identity (1996). In experiments one and two, d their degree of agreement with an ambiguous statement. Results supported the hypothesis that individuals would find similarity in indifference when primed with inclusive pronouns. The third experiment used a similar priming task but, rather than rating ambiguous statements, participants were asked to form spontaneous self - descriptions. Individuals primed with inclusive pronouns more frequently described themselves within a social context than individuals primed with exclusive pronouns or 34 adjectives. T he pattern of their research suggested that, w hen primed with inclusive pronouns, individuals reac t with a tendency to relate to others, categorize themselves within a social context, and modify their self - perceptions to identify with a collective (Brewer & Gardner, 1996). This tendency can even be used to increase perceptions of closeness in actual interpersonal interactions ( Fitzsimons & Kay , 2004 ) . In the first three studies of a four - study series, Fitzsimons and Kay investigated the effects of pronoun priming on perceptions o f individual closeness, first in the context of a written, fictitious story (Experiment 1), next, in the context of an existing, ongoing relationship (Experiment 2), and finally in the context of a real - time interpersona l interaction (Experiment 3). In the first experiment , individuals read a story where characters referenced one another with either exclusive pronouns, highlighting their individual identities, or a single inclusive pronoun to instead identify them as a uni t. Participants rated the characters , who referred to one another as a unit (via inclusive pronouns) as having a closer relationship than the characters who referred to one another with dis crete, exclusive pronouns. Consequently, in the second experiment , participants were asked to recall a current, ongoing relationship of their own and record details of the relationship either with distinct, exclusive pronouns (i.e. , , Participants in the collective pronoun condition rated their 551). Experiment 3 tested this effect in a live interaction with a confederate, where participants completed a fill - in - the - blank brainstorming task side by side with a confederate. Embedded in the task, sentences were structured to either refer to the participant and the confederate as a unit (inclusive pronoun) or as distinct entities (e xclusive pronouns). After the brainstorming task, a 35 brief scripted interaction took place, after which the participant and confederate were separated and asked to reflect on their interaction. Participants in the inclusive language not only perceived the ir interaction as closer than those in the exclusive language condition, but they also predicted that, were they to interact in the future, their relationship with the confederate would be closer. Experim ent 4 replicated E xperiment 1 but included a post - e xperimental questionnaire to closeness. Consistent with the first three studies, results supported the hypothesized role of inclusive language in the formatio n and perpetuation of interpersonal relationship perceptions. The identification of inclusive language as potential causal factor in group identification and relationship development is promising in light of the aforementioned indispensability mechanism of the Köhler effect, the relationship dynamic required to optimize that mechanism, and the findings of Irwin et al. ( 2013 ) study. While inclusive encouragement may bolster the Köhler effect, it should be noted that there is significant evi dence that verbal encouragement alone (typically heard from a trainer, teammate, or coach) may also p romote effort gains (Campenella et al., 2000; Guyatt et al., 1984; McNair et al., 1996 ; Moffat et al., 1994 ). Despite consistently helping performance, it is unlikely that encouragement alone a unilateral approach to motivation - could provide the motivation gains seen with the Köhler effect, which offers increases in effort through at least two potently mot ivating social processes (i.e. , upward social compari son and social indispensability. 36 CHAPTER THREE METHOD Design and Participants The study employed a 5 ( condition: individual - control , in dividual - with - encouragement , partner - without - encouragement , partner - exclusive - encouragement , partner - inclusive - encouragement ) x 2 (Performance B lock: Block 1 & Block 2) factorial design with repeated measures on the second factor. As with the Irwin et al . (201 3 ) study, most of the data for two of the conditions (i. e., 35 participants in the individual - control condition and 40 in the partner - without - encouragement condition) were co llected as part of the Kerr et al. ( 2012 ) and Irwin et al. (201 3 ) studies. In the current study , a new wave of data were collected for the individual - with - encouragement ( n = 50) , partner - inclusi ve - encouragement ( n = 49) , partner exclusive - encouragement ( n = 46) conditions. Additional data in the two other conditions (10 per condition) were also collected to contrast the latter with those collected for the Kerr et al. (2012) and Irwin et al. (2013) studies to probe for possible history or cohort effects (total: individual - control n = 45; partner - without - encouragement n = 50 , see Figure 3 Participant flow in Appendix K for an illustration of participant sources) . Irwin et al. did not find any systematic differences between the two waves of data collection, because the lab settings, participant populations, and procedures for both data collection periods were identical. We also did not find any differences (see Chapter 4 Preliminary analyses) . A one - way ANOVA examining potential differences between the control and partner - without - encouragement conditions from the data obtained during this wave of collected and the data used in the Irwin et al. (2013) study found no significant differences in ex ercise duration between waves according to the Tukey HSD procedure ( p s > .97 ). 37 Students were recruited from introductory psychology (online) and kinesiology courses (online and face - to - face) at a large Midwestern university and were given course credit for participation. Students were recruited based upon their interest in exercise and were told they would be playing an exercise video game and performing abdominal planking exercises for as long as they felt comfortable. The fin al total sample consisted of 240 participants (121 female, 1 19 ma le) of college age ( M = 20.32 , SD 1.83 ). Overall the average participant was a sophomore/junior (mean of 2.86, SD 1.26 where 1=1 st year, 2=2 nd year, etc.). Exercise Task The task for this study was the same exer game designed for the PlayStation 2 (PS2) gaming module as used in the Feltz et al. (2011) study. The software used was EyeToy: Kinetic, that operates in conjunction with an additional accessory called the EyeToy , designed specifically for the PS2 system. The E yeToy is essentially a small camera that connects to the PS2 system via a USB cable and allows images of the user to be displayed on the TV monitor and interact with virtual environments supported by the software. The abdominal plank exercises within the s trength - t raining module of the EyeToy: Kinetic software were used for this experiment. These are a type of bodyweight exercise where participants are required to suspend their own body weight using their abdominal muscles. These exercises are also isometri c in nature and require very little coordination, and thus are highly effort based. Each exercise targets the abdominal muscles, but there are slight differences between each. On the first exercise, participants were face down on a cushioned mat, with leg s extended straight, and they lifted their body upward by resting their elbows and toes on the mat and using their abdominals to lift their body. In this way, the body w as in a straight line, the spine was 38 directly in line with their head and legs and nothing w as touching the ground except for the elbows, forearms, and toes. In a similar fashion, the second exercise achieved the same elevated position, but the participant was on the left side with only the left forearm and left foot on the ground, empha sizing the use of the outer abdominal muscles. The third exercise was the same as the first exercise except that the participant had the left leg raised in the air and thus w as be balancing on only the right foot, which emphasize d the lower abdominal regio n. The fourth exercise was the same as the second, except the participant performed this on the right side. The fifth exercise w as the same as the third, except the participant performed this with the right foot in the air (see Figure 1). Participants perf ormed each exercise once within each of two blocks. Measures Effort . Effort was measured via task performance as the total number of seconds that the exercise will be held. Block scores w ere calculated by taking the summed total of the five exercises within each trial. Self - efficacy (SE ). Task SE w as measured with a scale developed specifically for this program of research (e.g., Feltz et a l., 2011). The measure contain s five items, each corr esponding to one of the five exercises within each trial. All items w ere preceded by the stem five exe rcises. Respondents w rote in the number of seconds in a blank box following each item. The questionnaire w as administered at three time points: once before Trial 1 (after the participant had watched a brief instructional video demonstrating the exe rcises), a second time before performing the five exercises at Trial 2, and a third time after Trial 2. A total SE score for each trial w as calculated by taking the sum of the five items within each trial. 39 Ratings of perceived exertion (RPE ). RPE w as used to provide a subjective rating of g the 6 - 20 version of the Borg ( 1998 ) RPE scale. The scale ranges from 6 - w e re asked to rate their exertion at t he end of each exercise, with particular reference to their perceived exertion at the moment right before the end of the exercise. Task enjoyment. Task enjoyment w as measured using a short 8 - item version of the Physi cal Activity Enjoyment Scale ( Kendzierski, 1991 ) . Each item w as rated on a 7 - point I loved it I hated it Intention to exercise . Intention was assessed with a single item, - Not at all true for me Completely true for me Post - experimental questionnaire . In addition to understanding of the instru ctions and procedures, there were questions probing their perceived task ability, a rating of task difficulty, and a rating of effort expended on th e task, each made on 8 - point scales. Particip ants w ere also - point scale (where 1 = I am much more capable and 9 = my partner is much more capable ). Additionally, s caled questions measuring the particip virtual partner as well as their own attitude toward the partner (or trainer where appropriate) w ere employed to determine if or how language used by the virtual partner changes the relational dynamic. Participant s were asked to answer the following questions with scaled response 40 options: scale where 1= no comparison and 7 = maximum comparison). How do you think your partn er would rate your performance? ( with a response scale where 1= I performed very poorly and 7= I performed very well), How do you think your partner would rate your ability? scale w here 1= my partner would rate me as very incapable to 7= my partner would rate me as very capable), my performance is not important at all to 7= my performance is very imp ortant), Was your partner (trainer) encouraging to you? ( with a response scale where 1= my partner (trainer) was very encouraging to me to 7= my partner (trainer) was very discouraging to me), How much did your partn er care ( with a response scale where 1= my partner did not care at all about my performance to 7= my partner cared very much about my performance), How did you perform as a team (h ( with a response scale where 1= my partner and I worked v ery well together to 7= my partner and I did not work well together at all), How much did you like your partner? ( with a response scale where 1= I like my partner very much to 7= I strongly dislike my partner). Specifically for the encouragement conditions, participants w ere scale where 1= the statements hindered my performance to 7= the statements were very helpful to my performance - confidence for the task? (with a response scale where 1= the statements strongly lowered my self - confidence for the task to 7= the statements strongly boosted my self - confidence for the task ) encouraging options: A. You B. Her /him self C. Both ) e statements affect your focus during distracting and 7= focusing ). 41 Figure 1. Images of exercises performed 42 Procedures Before conducting this study, permission w as obtained from the institutional review board. Before each session, an experimenter ensured that none of the participants ha d any disabling injuries to their arms, shoulders, back, or legs. Once an informed consent form ha d been signed, participants w ere a sked to remove any wrist jewelry/watches. Participants initially watched a brief instructional video from the PS2 - Eye Toy Kinetic software in which a virtual trainer demonstrated the five exercises. A baseline measure of self - efficacy w as then recorded. All participants then perform ed the first block of exercises, holding each of the five exercises for as long as they could and with 30s rest periods between each exercise. Immediately after each exercise, the participant reported his/her perceiv ed exertion on the 15 - point RPE scale. All participants w ere be given veridical feedback on their performance (i.e., the average of the number of seconds they held each exercise). The work condition manipulation w as introduced at this point. Participant i n the individual - control condition simply rested for 10 minutes. Participants in the individual with encouragement condition were told that the lab ha d recorded a trainer's communication phrases to test if they should be built into the exercises , and that while they are performing the next block of trials they will . Then, they rested for the remaining time before the next trial block start ed . Participants in the partnered conditions were t old that another participant wa s being run simultaneously at another lab on campus, and that the two participants w ould be able to see one another over an internet video connection during future t rials. The participants then met briefly with the other, same - sex participant in a controll ed Skype - like interaction (we will refer to that was an experimental confederate whose side of the interaction was pre - recorded. After th e interaction, participants 43 w ere also be given b ogus feedb ack on how well the partner performed on the firs t trial. That feedback score w as , since this appears to be the optimal discrepancy for inducing motivation gains ( Messé et al., 2002) . Participant s w ere then told that they and their pa rtners w ould be a two - person exercise team. In the two partner - with - encouragement condition s , participants were told that their partner would be able to communicate with them verbally during the next series of exercises but would not be able to respond to the participant . It w as then further explained that the team score w ould be the persistence score of the first teammate to quit an exercise (i.e., as soon as either partner quit, the exercise was over). Followi ng these instructions, all participants were again administered the self - efficacy measure. Block 2 then beg an . The participant was only able to see actually prerecorded) before and during the ex ercise; the participant knew th at the partner c ould ticipant suggest ed that s/he was always the first to quit e ac h exercise. The video link w as allegedly frozen as soon as either teammate quit an exercise and until just before the start of the next exercise. The participant, therefore, knew only that his/her partner had been able to persist longer, but not just how much longer. In the encouragement condition s , a pre - recorded serie s of phrases of encouragement w as played through a set of computer speakers controlled by th e experimenter. The phrases w ere be audible approximately every 15s ( + 3s) and followed a fixed progression: in the exclusive language condition, , , , good , rong here , , In the individual with encouragement phrases were provided through headphones in the same order as delivered at the same pace. 44 After Block 2 was over, the participant completed a series of questionnaires (self - efficacy, intention to exercise, enj oyment of physical activity, manipulation checks , and perception of partner ). The participants were t hen debriefed, thanked, and excused. 45 CHAPTER 4 RESULTS The purpose of this study was to examine the effects of encouragement on the Köhler effect in exergames . This chapter is organized into four main sections. The first section provides results on descriptive statistics and manipulation checks. The second section provides results of the preliminary analyses examining history and cohort effects. The third sec tion provides results for the main hypotheses. A final, fourth section presents results on ancillary analyses used to help interpret the main hypotheses. Descriptive Statistics, Confound Checks and Manipulation Checks Correlations . Bivariate correlations were calculated between all major study variables. Table 1, below, shows correlations of all study variables with Block 1 and Block 2 performance and RPE. Table 1 Correlations Between Study Variables, Performance, and RPE by Block Variable Performance 1 Performance 2 RPE 1 RPE 2 SE 1 .371** .305** .170** .150* SE 2 .439** .477** - .046 .042 SE 3 .345** .370** - .045 - .054 Intention to exercise in future .162* .100 .010 .041 Exercise enjoyment .135* .221** .068 .112 Teamwork .036 .192* - .128 - .010 Liking - .072 .081 - .248** - .185 Prediction of rating of S performance .152* .327** - .003 .026 Prediction of S ability .151* .317** .016 .036 Social indispensability .070 .271** - .021 .019 Upward social comparison .115 .332** .025 .097 Partner c aring .091 .267** .001 .062 Communication e ncouraging .053 .094 - .031 - .026 Communication h elpful .044 .141* .018 .014 Communication c onfidence b oosting .084 .183** .007 - .001 Communication f ocusing .052 .127* .020 .104 46 *p <.05. **p <.01 Notable patterns in these correlations include positive and signi ficant relationships among all three self - efficacy measures with both performance blocks, S elf - efficacy at T ime 1 and RPE in both performance blocks, P erformance at B lock 1 and I ntention to exercise in the future , and E xercise enjoyment and P erformance in both blocks. In the partnered conditions, there was a positive a nd significant relationship between Block 2 performances and P erceptions of teamwork, S ocial indispensability, U pward social comparison, and P artner caring. There was a negative and significant relationship between RPE at Block 1 and P artner liking, suggesting that so hard and this person outworked me? In the encouragement conditions, perceptions of encouragement helpfulness, confidence boosting, and focusing were positively and significantly correlated with Block 2 performance, suggesting that participant perceptions of encouragement effectiveness were accurate. Tables containing other correlations may be found in Appendix J. See Table 3 for corr elations among E xercise endurance, S elf - efficacy, RPE, E xercise enjoyment and E xercise intention. Table 4 contains correlations among respo nses to the partner information questionnaire items. Table 5 contains correlations among responses to the c ommunica tion questionnaire items. Missing data. For the main dependent variable, there was one count of missing data. No participants dropped out of the study before completing the session, though four were excluded from analysis because previous participation i n a similar study was discovered after testing was complete. 47 Confound checks. Experimenters were asked to record signs of suspicion, discomfort, previous a participant knew each other prior to the study. Upon completion of data collection, participant responses to the open - response item in the post - ere anything n = 29). A 5 (Condition) x 2 (Sex) ANOVA on performance scores was performed, excluding all participants who showed signs of suspicion, discomfort, boredom, and any observed factors that may have influenced the integrity of the experiment ( n = 61 , see Table 6 in Appendix J ). Results show a condition main effect , F (4, 179) = 20.9, p < .001, = .33 , which did not differ when the same analysis was performed with these participants included (see Hypothesis Testing section). A post - hoc Tukey test revealed results statistically indistinguishable from the analyses performed with all participants. Conseq uently, the participants were included in all subsequent analyses. It is interesting to note, however, that of the 29 participants showing suspicion in the partner conditions, more than half were in the exclusive - encouragement condition , though a Chi - squar e test revealed no statistically significant difference in suspicion among the partner conditions . Both male and female experimenters tested participants, but a 2 (Experimenter gender ) x 5 (Condition) analysis found no performance differences by experimen ter gender , F (2, 240) = .071, p = .790 . Consequently, experimenter gender was excluded from all subsequent analyses. Manipulation checks. An examination of the measure seeking whether or not participants in this data collection wave understood the ru les and design of the game (Appendix participants in the control condition (100%) reported correctly that they were performing the 48 exercises alone. Of the p articipants in the individual - with - encouragement condition who responded (98.0%), most (84.0%) reported correctly that they were performing the exercises alone. Of the participants exercising with a partner who responded to the item, ( partner - without - enco uragement , 100%; partner - exclusive - encouragement, 100%; partner - inclusive - encouragement, 97.9%), most participants exercising with a partner reported correctly that they were working with another person over an internet connection or were part of a 2 - perso n team, whether their partner was silent (90.0%), their partner offered exclusive encouragement (93.5%), or inclusive encouragement (91.8%). An examination of the measure examining whether or not participants from this collection wave understood how their score was determined (Appendix H, that their score was determined by their own score or the score of the weakest member). Participants in the partner - withou t - encouragement condition reported no incorrect responses, while most participants in the partner - exclusive - encouragement condition reported correctly (92.5%) and most participants in the partner - inclusive - encouragement condition reported correctly (88.4%) . Encouragement direction. O ne possible explanation that Irwin, et al. (2013) offered for self - encouragement, which could have undermined the perceived super iority of the partner (i.e., the partner was struggling and was granting audible self - affirmations). The perceived direction options 1 = you, 2 = her/himself, and 3 = both. Participants in the partner - exclusive - encouragement condition reported the encouragement as participant - directed 63.0% of the time, as self - directed 13.0% of the time, and as directed toward both team members 24% of the time 49 suggesting that particip ants generally felt that the encouragement was, at least in part, directed toward them and was rarely seen as solely self - that exclusive encouragement may have been mistaken as self - encouragement is not valid . Participants in the partner - inclusive - encouragement condition who responded to the item (95.9%) reported the encouragement as participant - directed 42.5% of the time, as self - directed 8.5% of the time, and as directed toward both team members 48.9% of the time , suggesting that inclusive encouragement was not perceived as directed at both exercisers more often than exclusive encouragement, but was also rarely perceived as solely self - directed. Preliminary analyses Stage 1 analyses looked for possible history or cohort effects attributable to the time interval between the new data and the data used by Irwin et al. (2013). In a 4 (Condition: Individual - control - old , individual - control - new , partner - without - encouragement - old , partner - without - encour agement - new ) x 2 ( Gender ) ANOVA, a boost in exercise duration was observed for the partnered conditions in both sets of data (Condition main effect , F (3, 95) = 19.57, p < .001, = .403). A post - hoc Tukey test found no significant differences in exerc ise duration between the collection waves in the control condition ( p = .975) and the partner - without - encouragement condition ( p = .977). All means, standard deviations, by condition, gender, and trial block for all variables employed in t he study are rep orted in Table 7 and Table 8 in Appendix J . Hypothesis Testing The main hypotheses made predictions regarding exercise duration and type of encouragement . Hypothesis 1 stated that compared to working alone with no encouraging statements, participants w ill exercise longer when working together with a moderately superior virtual partner under conjunctive task demands. Hypothesis 2 stated that compared to working 50 together with a moderately superior virtual partner who provides no encouragement , participan ts will exercise longer when working with a partner who provides encouraging statements that use stated that compared to working together with a moderately superior virtual partner who provides no encouragement , pa rticipants will exercise longer when working with a partner who provides encouraging state ments that use the exclusive . Hypothesis 4 stated that compared to working alone with no encouragement , participants will exercise longer when working alone with a trainer providing statements. Because the five exercises were small variations of one another, the total persistence ( in s econds ) across all five exercises was computed. In order to control for individual differences in strength and fitness, the primary dependent variable used was the difference score between both blocks (Block 2 Block 1), to show any changes in persistence. This approach has generally produced the same pat tern of results as using the Block 1 scores as a covariate in the analysis of Block 2 scores in previous research (Forlenza et al., 2012; Kerr et al., 2013). A one - way ANOVA examining these difference scores resulted in a significant condition main effect , F (4, 240) =19.99, p < .001. Planned contrasts were employed to test the four hypotheses as outlined in Table 8 . Table 2 Weighted values for planned contrasts of mean Block 2 Block 1 difference scores. I C I WE PWE PEE PIE Hypothesis 1 - 3 0 1 1 1 Hypothesis 2 0 0 - 1 0 1 Hypothesis 3 0 0 - 1 1 0 51 Table 2 ( Hypothesis 4 - 1 1 0 0 0 Note. IC = Individual - control , I W E = Individual - with - encouragement , PWE = Partner - without - encouragement, PEE = Partner - exclusive - encouragement, PIE = Partner - inclusive - encouragement. Results of the planned contrasts demonstrate d support for only Hypothesis 1 , t (235) = 8.37, p < .0 01 and Hypothesis 4, t (235) = 3.23, p = .001, but not Hypothesis 2, t (235) = - 1.04, p = .298 or Hypothesis 3, t (235) = 1.40, p = .16 4 . Thus, participants paired with a partner all showed dramatic improvements in exercise duration from Block 1 to Block 2, whether their partner was silent ( M = 43.60, SD = 53.25), provided exclusive encouragement ( M = 27.89, SD = 58.85), or provided incl usive encouragement ( M = 32.06, SD = 61.87) (see Figure 2), demonstrating a strong Köhler effect and providing evidence in support of Hypothesis 1 . Individuals exercising alone received a performance boost when offered encouragement ( M = - 7.58, SD = 45.03 ) when compared to individuals exercising in silence ( M = - 44.11, SD = 55.03), providing support for Hypothesis 4. However, support was found neither for the original hypothes is of the Irwin et al. (2013) study that exclusive encouragement would boost the Köhler effect (Hypothesis 3) n or for the hypothesized superiority of inclusive encouragement over silence (Hypothesis 2). 52 Figure 2 Mean duration (s) of performance difference scores (Block 2 Block 1) by condition. Note. I C = Individual - control , I W E = Individual - with - encouragement , PWE = Partner - without - encouragement, PEE = Partner - exclusive - encouragement, PIE = Partner - inclusive - encouragement. Ancillary Analyses Exercise self - efficacy. Post - block efficacy judgments (the number of seconds partic ipants estimated they could persist at each of the five exercises) were examined in a 2 (Block) x 5 (Condition) x 2 ( Gender ) ANCOVA, which included pre - block 1 self - efficacy scores or to performing any exercise) to control for chronic differences among participants of their self - efficacy belief pertaining to the task. As in previous studies, there was a Block main effect , F (1, 240) = 8.58, p 53 = .004, = .036, where participants were less optimistic about their potential for performance after Block 2 (adjusted M = 145.6s, SD = 80.11) than after Block 1 (adjusted M = 186.9s, SD = 92.56). Males were generally more optimistic about their performance than females , F (1, 240) = 7.96, p < .01, = .034. Unlike Irwin, et al. , F (4, 240) = 2.56, p = .034, = .043. Participants in both of the individual conditions reported lower self - efficacy estimates than participants in the part ner - without - encouragement and partner - exclusive - encouragement conditions, an effect which disappeared after the second block for the individual - with - encouragement condition but not for the individual control ( p s < .05). Subjective effort. To determine t was examined: the ratings of perceived exertion (reported after each exercise and averaged across exercises within blocks). Consistent with the previous studies, a 5 (Work condition) x 2 ( Gender ) x 2 (Block) analysis of RPE data found only a Block main effect: participants reported greater exertion at Block 2 ( M = 14.82, SD = 1.83) than Block 1 ( M = 14.10, SD = 1.77; F (1, 240) = 114.18, p < .001, = .33), as one might expect given the demands o f the task. However, the discrepancy did not differ significantly among the conditions, suggesting that perception of effort did not accurately reflect effort expended (i.e., participants exercising longer did not report that they felt more fatigued. Task evaluation. Overall task enjoyment was measured by means of the 8 - item PACES scale. The mean ( M = 4.09 on the 7 - point scale, SD = .822) was not significantly different from the scale midpoint , t (239) = 1.72, p = .087, suggesting that the participants wer e, at worst, neutral regarding the task. A 5 (Condition) x 2 ( Gender ) ANOVA of the measure found a condition main effect , F (4, 240) = 2.87, p = .024, = .048. A post - hoc Duncan test showed that the mean scores of the individual - without - encouragement (M = 3.87; SD = .956) was significantly lower ( p 54 < .05) than both the partner - inclusive - encouragement ( M = 4.29, SD = .81) and the partner - exclusive - encouragement ( M = 4.24, SD = .12). - experimental rating of the difficulty of the task was significantly higher than the scale midpoint ( M = 4.89, SD = 1.63, p < .001), ( t (236) = 3.64, p < .001) suggesting that participants found the task moderately challenging. A 5 (Condition) x 2 ( Gender ) ANOVA revealed no significant differences among gr oups on this measure , F (4, 237) = 1.63, p = .167. Intention to exercise. Intention to exercise was assessed with a single item in the post - experimental questionnaire. The mean (5.63 on the 7 point scale, SD = 1.72) was significantly greater than the sc ale midpoint , t (239) = 14.67, p < .001 , suggesting that in general, participants intended to exercise the following day. Unfortunately, the absence of a pre - intervention measure precludes any conclusion that the intention can be attributed to the experime nt. A 5 (Condition) x 2 ( Gender ) ANOVA found no significant differences among the groups on this measure , F (4, 240) = 1.05, p = .380. Perceptions of task ability. As a potential explanation for the deleterious effects of encouragement on the Köhler effe ct, Irwin et al. (2013) offered that encouragement undermined 2 ( Gender ) analysis of a measure examining perceived partner ability revealed that participa nts were unaffected by the communication manipulation , F (2, 145) = 2.33, p = .101. Social comparison. One of the two key components of the Köhler effect is upward social comparison. Because the previous study did not examine it directly, analysis was li mited to the two partnered communication conditions. A 2 (Partner encouragement conditions) x 2 ( Gender ) analysis revealed a significant condition effect , F (1, 95) = 4.48, p = .037, = .047, where individuals who received inclusive encouragement from a partner ( M =3.95, SD = .25) 55 than participants who received exclusive encouragement ( M = 4.72, SD = .26). Only the scores of the participants who received exclusive encouragement significantly differed from the mean ( M = .72 higher than scale midpoint ) , t (45) = 14.67, p = .003, suggesting that participants in the inclusive encouragement condition compared themselves to their partners very little if at all. Social indispensability. The second of the two key components of the Köhler effect is social indispensability. Because the previous study did not examine it directly, analysis was limited to the two partnered communication conditions. A 2 (Partner encou ragement conditions) x 2 ( Gender ) analysis revealed a significant condition effect , F (1, 95) = 4.89, p = .029, = .051, where participants who received exclusive encouragement reported greater feelings of indispensability ( M = 5.56, SD = 1.42) than pa rticipants who received inclusive encouragement ( M = 4.75, SD = 2.00). Both groups were significantly higher than the scale midpoint , t (94) = 6.26, p < .001, suggesting that feelings of social indispensability were strong in both conditions. Team perceptions. To assess perception of team coordination, a single item was from 1 = worked very well together to 7 = did not work well together at all. Becaus e the previous study did not examine it directly, analysis was limited to the two partnered communication conditions. A 2 (Partner encouragement conditions) x 2( Gender ) ANOVA found no significant differences between the groups , F (1, 95) = .651, p = .422. Another 2 (Partner encouragement conditions) x 2 ( Gender ) analysis for perception of found no significant differences between the groups , F (1, 95) = .003, p = .954 . There was, however, a significant interaction between the two , F (1, 95) = 4.63, p = .034, where both sexes 56 responded neutrally to inclusive encouragement, but males responded to exclusive encouragement with ratings of perceived caring significantly high er than the scale midpoint ( M = 4.85, SD = 1.22 ) , t (19) = 3.10, p = .006. This trend was not paralleled in responses to the measure examining how much participants liked their partner, where a 2 (Partner encouragement conditions) x 2 ( Gender ) analysis rev ealed no differences between groups , F (1, 95) = .119, p = .730. Communication effectiveness. communication in the three communication conditions was calculated by examining the mean of three questionnaire items. Though four items assessed communication effectiveness, the reverse scoring of one item appeared to confuse some participants, as it failed to correlate to the other measures at the p < .01 level , so it was excluded from the analyses. A 3 (Condition: individual - encouragement, partner - exclusive - encouragement partner - inclusive - encouragem ent) x 2 ( Gender ) ANOVA found no significant difference among groups , F (3,145) = .514, p = .599, but the mean of all groups was significantly higher than the scale midpoint ( M = 4.46, SD = 1.46 ) , t (144) = 3.77, p < .001, suggesting that participants in all communication conditions felt that the words were be neficial to their performance. 57 CHAPTER 5 DISCUSSION The primary purpose of this study was to test the efficacy of a superior internet partner (providing either inclusive encouragement, exclusive encoura gement, or no encouragement at all) as a strategy for increasing the duration of a series of isometric abdominal plank exercises. Consistent with previous studies, participants exercising with a virtually present partner, where task outcome was contingent on the performance of the weaker participant, persisted for significantly longer than participants silently working alone ( Feltz et al., 2011; Forlenza et al., 2012 ; Irwin et al., 2012; Irwin et al., 2013; Kerr et al., 2012; Kerr et al., 2013 ) . Thus, we found evidence in s upport of our first H ypothesis, that compared to working alone with no encouraging statements, participants will exercise longer when working together with a moderately superior virtual partner under conjunctive task demands. This motiv ation gain is not unlike gains seen previously, providing additional support for the Köhler effect in exergames as a potentially viable method for achieving fitness goals. und among the partner conditions. Accordingly, we found no evidence in support of our second Hypothesis, that Köhler effect compared to working together with a moderately superior virtual partner who provides no communication. We also found no evidence in support of our third Hypothesis, that the use of would boost the Köhler effect . Given that the same materials for the partner - exclusive - encouragement condition were use d in both studies, there are several potential explanations for the discrepancy between the Irwin et al. (2013) study and the present investigation. The mean performance difference of the partner - without - encouragement 58 one of the highest values for the Köhler effect in the studies employing this same task paradigm and was 10.02 s higher than in the present study. In addition, the mean for the partner with exclusive encou ragement was 8.40 s lower in in the failure to detect significant effects attributable to communication in the present study. Because mean performance diffe rence scores in the partner without encouragement condition for the present study are more reflective of the motivation gains typically seen in this paradigm, it seems plausible that encouragement, regardless of inclusivity, does not attenuate the Köhler e ffect. Alternatively, it is possible that there was an error in the collection methods of th e present study that resulted in a failure to obtain significance on a measure with suc h subtle variation between conditions (i.e., pronoun use and encouragement) . It may also be possible that , due to the subtlety of the communication manipulation in the partnered conditions, the study did not reach sufficient power to obtain statistical significance between the partnered conditions with communication. Finally, gi ven that performance in the communication conditions ( PEE, M = 27.89 and PIE, M = 32.06) were not as short as the PEE communication condition in the Irwin et al. (2013) study ( M = 19.49), it is possible that neither communication manipulation attenuated th e Köhler effect. Nevertheless, a performance scores. This suggests that encouragement (whether in the presence or absence of a partner), and especially the presence of a silent partner, can allow an individual to work harder than they think they are working. Perhaps more importantly, participants working with a partner receiving encouragement reported greater levels of enjoyment than when working silently alone. Given th at exercise enjoyment plays an important role in exercise adherence (Wininger & 59 Parqman, 2003), inclusion of encouragement or some type of verbal interaction may be important for implementation in an exergame for long - term success as a realistic primary ex ercise modality. Participants in all conditions reported strong intentions to exercise the following day, but due to the absence of a measure of this item before the intervention, these high levels cannot necessarily be attributed to gameplay. While it is possible that individuals already interested in fitness were more likely to enroll in the study than the average individual, the demographic would not suggest this. More plausibly, participation in strenuous exercise in the presence of an experimenter and sometimes a partner inspired or renewed fitness goals. Several participants indicated that they did not realize the inadequacy of their fitness level until after being eally need to Future studies may benefit from a baseline measure of this item to compare exercise intentions before and after exercise and identify changes attributable to gameplay. ( 2013 ) findings, but consistent with the hypotheses they presented, self - efficacy did indeed vary by condition in this study, with all partner conditions reporting higher self - efficacy measures than both individual conditions after B lock 1, an effect that disappeared for participants wor king alone with encouragement by the end of B lock 2. This finding suggests that the disembodied verbal encouragement did indeed have a mild efficacy - in the for mation of efficacy - beliefs. Despite this, according to both the self - efficacy measurements and actual performance values, which did not significantly differ among the partnered conditions, verbal persuasion did not seem to have a synergistic effect when c ombined Köhler effect, in which 60 motivation has increased maximally and any motivation boosting qualities of verbal encouragement are limited by actual physical performance abilities. It could also be indicative of a manipulation weakness , as may be alluded to by the number of suspicious participants in the encouragement conditions. Although encouragement should have increased the self - efficacy and performance scores of participants in the partner - encouragement conditions, it is plausible that the lack of believability of the voices in those conditions could have undermined any performance - boosting potential. Though the removal of suspicious subjects from a preliminary analysis resul ted in no significant differences, many participants ( n = 29 ) show ed or report signs of suspicion, especially in the encouraging - partner conditions. Participant laughter as a response theme in experimenter reports, suggesting that the participants either found the actual voices humorous or simply too contrived (or with a timeline too prescribed) to be believable. It is also possible that the laugh may have been a nervous laugh that re flected discomfort with recei ving encouragement in the presence of the experimenter , which may have highlight ed the However, u pon retrospective analysis of the encouragement used, it is quite plausible that the participants found it difficult to believe the realism of the participant after hearing the encouragement, as the voices sounded unstrained (contrary to what would be expected when an individual is speaking with flexed abdominals) , failed to reflect any fatigue as the session give encouragement when the participant was struggling, as would be seen in a truly live interaction. This realism, however, would not have been an issue in the individual - with - encouragement condition, where the encouragement was presented as pre - recorded and, as such, 61 could not have been perceived as deceptive. The attenuation of motivation loss from encouragement tha t is typically seen when working alone is especially interesting because typically research on encouragement in exercise has not used pre - recorded phrases like those used in the present study , but rather the encouragement of a live experimenter ( Campenella , Mattacola & Kimura, 2000; Guyatt et al., 1984; Moffat, Chitwood & Biggerstaff, 1994; McNair, Depledge & Stanley, 1996) . This suggests that perceived meaningfulness or honest intention to encourage may not be required for performance gain through encoura gement. Lastly, it is possible that the mere novelty of a pre - recorded voice may have entertained or distracted participants, as suggested by Soltani and Salesi (2013), who saw a performance boost from participants hearing pre - recorded encouragement and m usic in a gaming situation with virtual avatars of co - exercisers. Some support was found for the explanations offered by Irwin et al. (2013) for the lower performance scores in the encouragement condition in their study. Irwin et al. offered that th e - tations of the direction of communication suggest the participants in the partner - exclusive - encouragement condition - directed or team - directed. No differences in ratings of how much participan ts liked their partner were found, suggesting that the encouragement was not interpreted negatively as either patronizing or condescending: rather, most participants provided favorable ratings of the encouragement they heard, regardless of its source. 62 Des pite insignificant performance differences between the partnered conditions with encouragement, a pattern emerge d with questionnaire responses to two items assessing key social factors contributing to the Köhler effect: upward social comparison and social indispensability. The primary rationale for Hypothesis 2, that pronoun would enhance the Köhler effect compared t o a no n - communicati ve partner , was that inclusive language would foster a stronger group identity and boost perc eptions of partner - inclusive - encouragement condition reported lower levels of social indispensability than those who received exclusive - encouragement, and reported no social comparison. It seems that social comparison and social indispensability may both rely, to some degree, on a self - focus by the weaker partner that emphasizes him or her as the weaker/indispensable partner that is absent or reduced by ove rt inclusion through the use of inclusive language. This relationship may conform to Self - Categorization Theory, which posits that as a collective identity becomes more asingly interchangeable (Turner, Oakes, Has lam, & McGarty, 1994 ). Specifically, the theory states that rsonal identity, this is ordinar y and normal s elf - experience in which the self is defined in terms of others who exist outside the individual person doing and experiencing and therefore of the self - focus or discrete individual identity, the opportunity to social compare is absent and any feelings of obligation to a group (i.e., indispensability) may be reduced because of a shift in the perception of self from individual (i.e., with personal own ership of performance) to group (i.e., a divided ownership of performance) by reducing perceived social impact ( Latané et al., 63 1979). Unfortunately, since most data for the partner - without - encouragement condition was collected in a previous wave where que stionnaire items examining these factors were not employed, we have no baseline of these measures for comparison. Though differences between the partner communication conditions were not observed in measures of teamwork, partner liking, or partner caring, the Gender x Condition interaction on responses indicating perceptions of partner caring suggests that exclusive encouragement indicates to male participants that the partner cares about them. One potential explanation for this finding is that males tend focused, and consequently may dislike being submerged in the group ( Watson et al., 1987 ; Wright et al., 1989). While this phenomenon did not manifest in performance differences, it may be useful to keep in mind in the development of a virtual partner tailored to a specific individual . Limitations While performance differences observed in the Irwin et al. (2013) study were not reproduced here, a trend in line with those differences was seen, as the verbal encouragement conditions with virtually present partners had lower means than the no communication partner condition . It is possible that, after a re - examination of the manipulations, the lack of statistical differences in the current study ma y have been an artifact of poor manipulations (i.e., unauthentic and poorly constructed recordings). E ncouragement may boost the Köhler effect if presented in a more persuasive manner (e.g., a labored, passionate voice), if presented at more opportune tim es (e.g., when the participant is struggling rather than at prescribed intervals), or if presented as the pre - recorded voice of an automaton or software generated partner (i.e., to avoid aversive reactions to detected deception). 64 In addition to the myria d potential problems with the materials used in this study, the researcher failed to control for participant language skills. It was evident that English was not the primary language of some participants, though no formal inquiry was made in to their fami liarity with the English language and consequently this factor could not be examined as a confounding variable post data collection . Though responses to the communication questionnaire items indicate that, in general, communication was retrospectively per ceived as helpful, verbal encouragement encountered during exercise may not have had the same visceral motivation boosting effect on a non - native English speaker as it would on a native English speaker. Future studies ought to check the native language of the participant as well as their familiarity (e.g., number of years in country) with the language used in the manipulation. Future Directions Future directions in a line of research pursuing the effects of communication in exergames ought to first addre ss some of the issues presented in this study. To ameliorate any potential issues with the use of confederates, several solutions may be worth pursuing. The first could be to use trained actors as confederates whose encouragement would be perceived as mo re authentic . The use of a trained actor confederate for pre - recorded interactions and intra - exercise communication may increase believability. Another alternative is to avoid pre - recorded interactions altogether, and rather introduce confederates, live, over an internet connection but present in another room. Because fatigue may become an issue with a live, virtually - present partner, the exercise video could be pre - recorded while intra - game encouragement is delivered live via audio. 65 Examining the part in the design of future research. A communication intervention will undoubtedly diminish in effectiveness if the communication is not understood or detected. It may also be prudent to investigate and optimize the pre - exercise web camera interaction. Though the interaction was consistent across partner conditions, neither of the confederates mirrored the emotions of the participants: they appeared to be neither friendly, nervous, n or as interested as their exercise partners. The male confederate is visibly older than the typical undergraduate college demographic, and the female confederate can even be seen looking up and away from the camera (seemingly rolling her eyes) at the time a llocated for the from these interactions could have affected all subsequent verbal interactions, making them seem less genuine than they otherwise would ha ve , though the researcher did not assess these factors directly . Accordingly, nonverbal immediacy (i.e., nonverbal behaviors that communicate liking and a generally positive evaluation) during the initial interaction may be worthwhile examining. In orde r to eliminate all potential issues with the use of confederates while still ensuring the superiority of the partner, researchers interested in studying this topic may consider the development of an entirely software - generated partner or trainer. A soft - w are generated partner or trainer would rely upon either a prescribed timing for feedback, as used in this study, the input of an experimenter to deliver the feedback at times of need, or perhaps the direct input of the - generated partner would be able to recognize times of need and could be tailored to specific exercisers. A software - and could b e programmed to interact with exercisers consistently in motivation - boosting ways. 66 However, the best methods to boost motivation through communication in this paradigm are still unknown. Accordingly, an interesting avenue for future research may be an inv estigation of the nature of communication in real workout teams. The manipulations used in this study used only a few encouraging phrases, but it is possible that real exercisers use language quite different than that utilized here. Performing an observa tional study investigating what is said in real workout groups, how it is said, when it is said, and the vocal (e.g., inflection, tonality) and facial (e.g., smile, grimace, neutrality), characteristics of the communicator s may illuminate the path for futu re projects. For example, is encouragement really used of performance information (e.g. your form is breaking, in pace tasks)? This type of investigation could include the observation of real, pre - existing working groups exercising in - exercise partners who, a fter meeting in a lab setting, could be instructed to exercise with each could be observed, recorded, and analyzed to empirically identify communication patterns be tween exercise partners. This study examined the effect of encouragement on exergamers performing an isometric abdominal planking task. An examination of the effect of communication on endurance in rhythmic, aerobic exercise (e.g., cycling, rowing, and running) is warranted. Rhythmic, full - body exercise modalities offer greater potential cardiovascular health benefits than localized, isometric exercises. 67 Communication (especially encouragement) from trainers and partners in exergames intended for reha bilitation and physical therapy settings may prove to be useful , especially with older adults . Exergames have only recently begun to be explored as a delivery vehicle for rehabilitation - specific at - home training (Smith & Schoene, 2012). Because many decl ines in bodily function in aging adults occur in tandem with declines in self - confidence, limited social interaction, and a tendency to remain inside the home, the social aspects of a partner (i.e., communication) may be a crucial component of rehabilitati on effectiveness. Finally, communication in this study was restricted to one direction. Communication typically involves an exchange of information rather than merely delivery of it. Future studies may examine how encouragement between partners rather t han encouragement from a designated source to a predetermined target, may affect the motivational climate . Conclusions The current study validates previous research ( Feltz et al., 2011; Forlenza et al., 2012 ; Irwin et al., 2013; Irwin et al., 2012; Kerr et al., 2012; Kerr et al., 2013) that exercising with a moderately superior virtually - present partner can boost motivation and increase exercise duration, regardless of partner communication. Unlike Irwin et al. (2013), verbal encouragement, regardle ss of inclusivity, did not significantly moderate this effect. Encouragement in the absence of a partner elicited effort gains above exergaming alone without encouragement, but those gains paled in comparison to gains seen in partnered gameplay. Future r esearch should examine the effect of communication from a software generated partner, different types of pre - exercise interactions, and the effects of different types of interpersonal . 68 exergames. 69 APPENDICES 70 APPENDIX A THE BORG SCALE OF PERCEIVED EXERTION The Borg Scale 6 No exertion at all 7 Extremely light 8 9 Very light 10 11 Light 12 13 Somewhat hard 14 15 Hard (heavy) 16 17 Very hard 18 19 Extremely hard 20 Maximal exertion 71 APPENDIX B SELF - EFFICACY BELIEFS Please type in a numerical response in each of the boxes in response to the question. What is the number of seconds which you are completely confident that you can hold: The FIRST exercise (front plank)? _________ The SECOND exercise (right side plank)? _________ The THIRD exercise (right one - legged plank)? __ _______ The FOURTH exercise (left side plank)? _________ The FIFTH exercise (left one - legged plank)? _________ 72 APPENDIX C PHYSICAL ACTIVITY ENJOYMENT QUESTIONNAIRE Please rate how you feel at the moment about the physical activity you have been doing according to the following scales (find the scales above and to the left of each row of checkboxes = ex. 1 = _______, 7 = _______) 1. 1= I loved it, 7 = I hated it 1 2 3 4 5 6 7 2. 1= I felt bored, 7= I felt interested 1 2 3 4 5 6 7 3. 1= I disliked it, 7= I liked it 1 2 3 4 5 6 7 4. 1= I found it pleasurable, 7= I found it unpleasurable 1 2 3 4 5 6 7 5. 1= I was very absorbed in this activity, 7= I was not at all absorbed in this activity 1 2 3 4 5 6 7 73 6. 1= It was no fun at all, 7= It was a lot of fun 1 2 3 4 5 6 7 7. 1= It was very pleasant, 7= It was very unpleasant 1 2 3 4 5 6 7 74 APPENDIX D INTENTION TO EXERCISE Please respond to the following statement: 1. - 3 = Not at all true for me, 3= Completely true for me - 3 - 2 - 1 0 1 2 3 75 APPENDIX E PARTNER INFORMATION QUESTIONNAIRE If you worked with a partner in the second block of exercises, please answer the questions below. Otherwise, skip ahead. 1. Did you compare your average planking time to that of your partner? (1= No comparison, 7= Maximum comparison) 1 2 3 4 5 6 7 2. As you were preparing to start the exercise, how important did you think your performance would be to the group score? (1= My performance is not important at all, 7= My performance is very important) 1 2 3 4 5 6 7 3. partner, 4 = My partner and I are equally capable, 9= My partner is much more capable than me) 1 2 3 4 5 6 7 8 9 4. How do you think your partner would rate your ability? (1 = My partner would rate me as very incapable, 7= My partner would rate me as very capable) 76 1 2 3 4 5 6 7 5. How do you think your partner would rate your performance? (1 = I performed very poorly, 7= I performed very well) 1 2 3 4 5 6 7 6. How much did your partner care about your performance? (1= My partner did not care at all about my performance, 7 = My partner cared very much about my performance) 1 2 3 4 5 6 7 7. How did you perform as a team (how well did you work together)? (1 = Worked very well together, 7 = Did not work together at all) 1 2 3 4 5 6 7 8. How much did you like partner? (1 = I like my partner very much, 7 = I strongly dislike my partner) 1 2 3 4 5 6 7 77 APPENDIX F COMMUNICATION QUESTIONNAIRE respond to the questions below. Otherwise, skip ahead. 1. Were the statements helpful to your performance? (1 = the statements greatly hindered my performance, 7 = the statements were very helpful to my performance) 1 2 3 4 5 6 7 2. Did the statements boost your self - confidence for the task? (1 = the statements strongly lowered my self - confidence for the task, 7 = the statements strongly boosted my self - confidence for the task) 1 2 3 4 5 6 7 3. Was your partner (or trainer) encouraging to you? (my partner [trainer] was very encouraging to me, 7 = my partner [trainer] was very discouraging to me) 1 2 3 4 5 6 7 4. Whom was your partner talking to? (1 = you, 2 = her/himself, 3 = both) 1 2 3 78 5. How did the statements affect your focus during the task? (1 = distracting, 7 = focusing) 1 2 3 4 5 6 7 79 APPENDIX G ACTIVE PERFORMANCE QUESTIONNAIRE 1. How much interest would you have in participating in another exercise study like this one? 1 = None at all, 8 = Very much 1 2 3 4 5 6 7 8 2. How difficult did you find the exercises that you did today? 1 = Not at all difficult, 8 = Extremely difficult 1 2 3 4 5 6 7 8 3. How much effort did you exert when performing these exercises? 1 = My absolute minimum, 8 = My absolute maximum 1 2 3 4 5 6 7 8 1. How capable to perform these exercises do you feel? 1 = Extremely incapable, 8 = Extremely capable 1 2 3 4 5 6 7 8 80 APPENDIX H MANIPULATION CHECKS In which of the following conditions did you perform the last series of exercises? (check one) 1. (Except for the experimenter) I performed these exercises alone. 2. I performed these exercises with another person through an internet connection. 3. I performed these exercises with two other persons through an internet connection. 4. I performed these exercises as part of a two - person team. 5. I performed these exercises as part of a three - person team. How was your score determined during the last series of exercises? 1. My score is the sum of the number of seconds I held each exercise. 2. My score is the average number of seconds I held each exercise 3. My score is the average of my is the number of seconds each exercise was held by the first team member to quit the number of seconds each exerci se was held by the first team member to quit 5. My score is the average of how long I held each exercise and how long my partner held each exercise 6. My score is the sum of how long I held each exercise and how long my partner held each exercise Was there anything confusing or odd about the experiment? (open answer) What, in your own words, do you think the purpose of this experiment was? (open answer) 81 APPENDIX I DEMOGRAPHICS 1. Height (inches) ______ 2. Weight (lbs) ______ 3. Sex (M) (F) 4. Age ______ 5. Class a. 1 st year b. 2 nd year c. 3 rd year d. 4 th year e. 5 th year f. >5 th year 6. E - mail address ________ 82 APPENDIX J AUXILIARY TABLES Table 3 Bivariate correlations of primary dependent variables. Measure 1 2 3 4 5 6 7 8 9 10 1. Sum of block 1 performances - 2. Sum of block 2 performances .774 ** - 3. Difference score (Block 2 Block 1) - .407 ** .245 ** - 4. Sum of pre - block 1 self - efficacy scores .371 ** .305 ** - .140 * - 5. Sum of post - block 1 self - efficacy scores .439 ** .477 ** - .004 .563 ** - 6. Sum of post - block 2 self - efficacy scores .345 ** .370 ** .007 .506 ** .750 ** - 7. Mean block 1 RPE score .077 .037 - .097 .170 ** - .046 - .045 - 8. Mean block 2 RPE score .122 .147 * - .015 .150 * .042 - .054 .837 ** - 9. Mean PACES score .135 * .221 ** .111 .090 .178 ** .092 .068 .112 - 83 10. Mean exercise intention score .162 * .100 - .101 .099 .089 .024 .010 .041 .193 ** - Note. ** p < .001, * p < .05 84 Table 4 Bivariate correlations of responses to partner information questionnaire items. Item 1 2 3 4 5 6 7 1. Teamwork - 2. Liking .435 ** - .332 ** .210 * - .199 * .187 .917 ** - 5. Social indispensability .137 .276 ** .664 ** .621 ** - 6. Upward social comparison .124 .091 .650 ** .599 ** .778 ** - 7. Caring .296** .373** .221* .165 .382** .176 - Note . ** p < .001, * p < .05 Table 5 Bivariate correlations of responses to communication questionnaire items. Item 1 2 3 4 1. Encouraging - 2. Helpful .132 - 3. Confidence Boosting .272 ** .825 ** - 4. Focusing .142 .694 ** .696 ** - Note . ** p < .001, * p < .05 85 Table 6 Frequencies of noted potential confounds. Issue N Attire 17 Suspicion 31 Discomfort 0 Boredom 2 Equipment 6 Fitness 2 Activity 8 Know participant 2 Observer present 0 86 Table 7 Means and standard deviations of all study variables by condition. Condition C IE SP PEE PIE Variable M (SD) M (SD) M (SD) M (SD) M (SD) Block 1 performance 276.44 (86.14) 249.56 (92.3) 227.50 (95.90) 278.41 (106.81) 265.41 (106.99) Block 2 p erformance 232.33 (75.06) 241.98 (74.83) 275.32 (97.73) 306.30 (99.15) 297.47 (93.37) Block 2 Block 1 - 44.11 (55.03) - 7.58 (45.03) 43.60 (53.25) 27.89 (58.85) 32.06 (61.87) RPE 1 14.09 (1.95) 14.22 (1.86) 13.77 (1.94) 14.18 (1.21) 14.28 (1.80) RPE 2 14.88 (2.01) 14.56 (1.96) 14.59 (1.92) 15.11 (1.47) 14.99 (1.73) SE 1 253.13 (442.02) 294.02 (153.81) 183.66 (112.04) 302.96 (173.22) 286.51 (145.31) SE 2 158.73 (114.39) 178.74 (86.60) 192.22 (108.47) 218.83 (148.82) 185.53 (98.43) SE 3 119.22 (84.24) 159.72 (100.52) 143.90 (78.21) 157.89 (113.17) 146.12 (78.13) Intention to exercise 6.02 (1.45) 5.60 (1.81) 5.40 (1.81) 5.39 (1.87) 5.76 (1.60) Exercise enjoyment 3.88 (.95) 4.09 (.86) 3.95 (.80) 4.24 (.60) 4.29 (.81) Teamwork - - - 3.96 (1.41) 4.14 (1.41) Liking - - - 5.06 (1.18) 4.96 (1.24) Partner's rating of performance - - - 2.17 (1.18) 2.63 (1.41) - - - 2.74 (1.41) 3.26 (1.71) Social indispensability - - - 5.57 (1.42) 4.76 (2.01) Upward social comparison - - - 4.72 (1.57) 3.94 (1.88) Caring - - - 4.15 (1.65) 4.22 (1.53) Encouraging - 4.70 (1.52) - 5.27 (1.37) 5.17 (1.31) Helpful - 4.32 (1.75) - 4.61 (1.36) 4.27 (1.62) Confidence Boosting - 4.40 (1.66) - 4.50 (1.17) 4.41 (1.64) Focusing - 4.58 (2.01) - 4.85 (1.48) 4.24 (1.70) Note. C = Individual Control without encouragement, IE = Individual Control with encouragement, SP = Partner without encouragement, PEE = Partner with exclusive encouragement, PIE = Partner with inclusive encouragement. 87 Table 8 Means and standard deviations of all study variables by gender . Gender Means Female Male Variable M (SD) M (SD) Block 1 performance 243.70 (81.44) 273.60 (112.11) Block 2 p erformance 263.41 (87.92) 278.02 (96.96) Block 2 Block 1 Difference score 19.71 (58.70) 2.68 (66.40) RPE 1 13.93 (1.67) 14.28 (1.87) RPE 2 14.53 (1.75) 15.11 (1.86) SE 1 223.79 (132.53) 302.61 (299.41) SE 2 160.26 (93.63) 213.01 (124.75) SE 3 126.93 (80.12) 164.17 (99.33) Intention to exercise 5.60 (1.71) 5.66 (1.74) Exercise enjoyment 4.10 (.93) 4.12 (.71) Teamwork 4.26 (1.28) 3.94 (1.50) Liking 4.97 (1.24) 5.06 (1.05) Partner's rating of performance 1.75 (1.75) 1.93 (1.99) 1.47 (1.82) 1.40 (1.92) Social indispensability 2.73 (2.90) 2.28 (2.88) Upward social comparison 2.24 (2.46) 1.84 (2.51) Caring 2.07 (2.28) 1.82 (2.36) Encouraging 5.10 (1.45) 4.84 (1.41) Helpful 3.12 (2.41) 2.54 (2.47) Confidence Boosting 3.12 (2.39) 2.58 (2.47) Focusing 3.20 (2.59) 2.64 (2.47) 88 APPENDIX K Figure 3. Participant flow. 89 REFERENCES 90 REFERENCES Ajzen, I., & Fishbein, M. (1977). Attitude - behavior relations: A theoretical analysis and review of empirical research. Psychological Bulletin , 84 (5), 888 918. doi:10.1037//0033 - 2909.84.5.888 Ajzen, I. & Fishbein, M. (1980). Understanding attitudes and predicting social behavior. Englewood Cliffs, NJ: Prentice - Hall. Ajzen, I. (1991). The Theory of Planned Behavior. Organizational Behavior and Human Decision Process , 50 , 179 211. Bandura, A. (1977). Self - eff icacy: Toward a Unifying Theory of Behavioral Change. Psychological Review, 84 (2), 191 - 215. Bandura, A. (1982). Self - efficacy mechanism in human agency. American psychologist , 37 (2), 122 147. Retrieved from http://psycnet.apa.org/journals/amp/37/2/122/ Barnett, A., Cerin, E., & Baranowski, T. (2011). Active video games for youth: a systematic review. . Journal of Physical Activity and Health. , 8 (5), 724 - 737. Baron, R. S., & Kerr, N. L. (1983). Group Process, Group Decision, Group Action (p. 56). Bello cco, R., Jia, C., Ye, W., & Lagerros, Y. T. (2010). Effects of physical activity, body mass index, waist - to - hip ratio and waist circumference on total mortality risk in the Swedish National March Cohort. European Journal of Epidemiology , 25 (11), 777 88. do i:10.1007/s10654 - 010 - 9497 - 6 Biddiss, E., & Irwin, J. (2010). Active video games to promote physical activity in children and youth: a systematic review. . Archive of Pediatric Adolescence., 164 (7), 664 - 672. Borg, G. (1998). Borg's perceived exertion and pain scales. Human Kinetics Publishers . representations. Journal of Personality and Social Psychology , 71 (1), 83 93. doi:10.1037//0022 - 3514.71.1.83 Brown, T. C. (2003). The effect of verbal self - guidance training on collective efficacy and team performance. Personnel Psychology, 56 , 935 - 964. Brownson, R. C., Boehmer, T. K., & Luke, D. A. (2005). Declining rates of physical activity in the United States: What are the contributors? Annual Review of Public Health, 26 , 421 - 443. doi: 10.1001/jama.291.10.1238 91 Burke, S., Carron, A., Eys, M., Ntoumanis, N., Estabrooks, P. . (2006). Group versus Individual Approach? A Meta - Analysis of the Effectiveness of Interventions to promote physical activity. Sport & Exercise Psychology Review, 2 (113), 1 - 39. Campenella, B. , Mattacola, C. G., & Kimura, I. F. (2000). Effect of visual feedback and verbal encouragement on concentric quadriceps and hamstrings peak torque of males and females. Isokinetics and Exercise Science , 8 , 1 6. Carron, A. V., Burke, S. M., & Prapavessis, H . (2004). Self - Presentation and Group Influence. Journal of Applied Sport Psychology , 16 (1), 41 58. doi:10.1080/10413200490260044 Carron, A. V., Hausenblas, H. A., & Eys, M. A. (1998). Group dynamics in sport. Fitness Information Technology. Morgantown, W V . Courneya, K. S., & McAuley, E. (1995). Reliability and discriminant validity of subjective norm, social support, and cohesion in an exercise setting. Journal of Sport & Exercise Psychology , 17 (3), 325 337. Retrieved from http://www.getcited.org/pub/103 340222 Courneya, K. S., Nigg, C. R., & Estabrooks, P. A. (2000). Relationships among the theory of planned behaviour, stages of change, and exercise behaviour in older persons over a three year period. (pp. 189 - 205) Harwood Academic Publishers, Amsterdam. Retrieved from http://ezproxy.msu.edu/ login?url=http://search.proquest.com/docview/619670400?accounti d=12598 Courneya, K. S., Plotnikoff, R. C., Hotz, S. B., & Birkett, N. J. (2000). Social Support and the Theory of Planned Behavior in the Exercise Domain. American Journal of Health Behavior , 24 (4), 300 308. doi:10.5993/AJHB.24.4.6 Daley, A. J. (2009). Can exergaming contribute to improving physical activity levels and health outcomes in children? . Pediatrics, 124 (2), 763 - 771. Davis, J. H. (1969). Group Performance. Reading, Mass.: Addison - Wesley. Deci, E. L., & Ryan, R. M. (1985). Intrinsic motivation and self - determinaton in human behaviour . New York: Plenum. Dishman, R. K., & Buckworth, J. (1996). Increasing physical activity: a quantitative synthesis. Medicine & Science in Sports & Exercise. . Medicine & Science in Sports and Exercise, 28 (6), 706 - 719. Downs, D., Graham, G., Yang, S., Bargainnier, S., & Vasil, J. (2006). You th exercise intention and past exercise behavior: Examining the moderating influences of sex and meeting exercise recommendations. Research Quarterly for Exercise and Sport , 77 (1), 91 99. 92 Dumith, S. C., Hallal, P. C., Reis, R. S., & Kohl, H. W. (2011). W orldwide prevalence of physical inactivity and its association with human development index in 76 countries. Preventive Medicine: An International Journal Devoted to Practice and Theory, 53 (1 - 2), 24 - 28. Dzewaltowski, D. (1990). Physical activity participa tion: Social cognitive theory versus the theories of reasoned action and planned behavior. , 12 , 388 405. Retrieved from http://psycnet.apa.org/psycinfo/1991 - 18363 - 001 Esteghamati, A., Morteza, A., Khalilzadeh, O., Anvari, M., Noshad, S., Zandieh, A., & Nakhjavani, M. (2012). Physical inactivity is correlated with levels of quantitative C - reactive protein in serum, independent of obesity: results of the national surveillance of risk factors of non - communicable diseases in Iran. Journal of Health, Population, and Nutrition , 30 (1), 66 72. Feltz, D.L. (1988). Self - confidence and sports performance. Exercise and sport sciences reviews, 16, 423 - 57. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/3292264 Feltz, D. L., Irwin, B. C., & Kerr, N. L. (2012). Two - player partnered exergame for obesity prevention: Using discrepancy in players' abilities as a strategy to motivate physical activity. Journal of Diabetes Science and Technolology., 6 (4), 820 - 827. Fel tz, D. L., Kerr, N. L., & Irwin, B. C. (2011). Buddy up: the Köhler effect applied to health games. Journal of Sport & Exercise Psychology, 33 (4), 506 - 526. - efficacy and an anxiety - based model of avoidance behavior. Journal of Personality and Social Psychology , 42 (4), 764 781. doi:10.1037//0022 - 3514.42.4.764 Feltz, D. L., & Mugno, D. A. (1983). A replication of the path analysis of the causal elements in of self - efficacy and the influence of autonomic perception. Journal of Sport Psychology , 5 (3), 263 277. Feltz, D. L., Chow, G. M., & Hepler, T. J. (2008). Path analysis of self - efficacy and diving performance revisited. Journal of Sport & Exercise Psychol ogy , 30 (3), 401 11. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/18648112 Fishbein, M. & Ajzen, I. (1975). Belief, attitude, intention, and behavior: An introduction to theory and research . Reading, MA: Addison - Wesley. [1] Fitzsimons, G. M., & Kay, A. C. (2004). Language and interpersonal cognition: causal effects of variations in pronoun usage on perceptions of closeness. Personality & social psychology bulle tin , 30 (5), 547 57. doi:10.1177/0146167203262852 Forlenza, S. T., Kerr, N. L., Irwin, B. C., & Feltz, D. L. ( 2012 ). Is my exercise partner similar enough? Partner characteristics as a moderater of the Köhler effect in exergames. Games for Health Journal . 93 Gage, L. (1990). An examination of the utility of the health belief model for predicting adult participation in aerobic exercise. (Ph.D. 9115983), University of Florida, United States -- Florida. ProQuest Dissertations & Theses (PQDT); ProQuest Disserta tions & Theses A&I database. Effects of the Violation of Implicit Rules of Social Interaction. Sociometry , 37 (4), 541 556. Gockel, C., Kerr, N. L., Seok, D. - H., & Harris, D. W. (2008). Indispensability and group identification as sources of task motivation. Journal of Experimental Social Psychology , 44 (5), 1316 1321. doi:10.1016/j.jesp.2008.03.011 Godin, G., & Kok, G. ( 1996 ). The Theory of Planned Behavior: A Revi ew of Its Applications to Health - related Behaviors. American Journal of Health Promotion , 11(2), 87 - 98. Gould, D., Hodge, K., Peterson, K., & Giannini, J. (1989). An exploratory examination of strategies used by elite coaches to enhance self - efficacy in a thletes. Journal of Sport & Exercise Psychology, 11 (2), 128 - 140. D. W. (1984). Effect of encouragement on walking test performance. Thorax , 39 (11), 818 22 . Retrieved from http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=459930&tool=pmcentrez&rend ertype=abstract Haapanen - Niemi, N., Miilunpalo, S., Pasanen, M., Vuori, I., Oja, P., & Malmberg, J. (2000). Body mass index , physical inactivity and low l evel of physical fitness as determinants of all - cause and cardiovascular disease mortality - 16 y follow - up of middle - aged and elderly men and women. International Journal of Obesity , 24 , 1465 1474. Hagger, M., Chatzisarantis, N., & Stuart, J. (2002). A m eta - analytic review of the theories of reasoned action and planned behavior in physical activity: Predictive validity and the contribution of additional variables. Journal of Sport & Exercise Psychology , 24 , 3 32. Retrieved from http://psycnet.apa.org/psyc info/2002 - 12499 - 001 Harkins, S. G., & Petty, R. E. (1982). Effects of task difficulty and task uniqueness on social loafing. Journal of Personality and Social Psychology , 43 (6), 1214 1229. doi:10.1037//0022 - 3514.43.6.1214 Herman, C. M. (2010). The Burden of Obesity and Physical Inactivity Across the Lifespan, with a Focus on Health - Related Quality of Life. (Ph.D. NR78350), Queen's University (Canada), Canada. ProQuest Dissertations & Theses (PQDT); ProQuest Dissertations & Theses A&I database. 94 Hertel, G., Deter, C., & Konradt, U. (2003). Motivation Gains in Computer - Supported Groups1. Journal of Applied Social Psychology, 33 (10), 2080 - 2105. doi: 10.1111/j.1559 - 1816.2003.tb01876.x Hertel, G., Kerr, N. L., & Messé , L. A. (2000). Motivation gains in performance groups: paradigmatic and theoretical developments on the Köhler Effect. Journal of Personality and Social Psychology, 79 (4), 580 - 601. Hertel, G., Niemeyer, G., & Clauss, A. (2008). Social Indispensability or Social Comparison: The Why and When of Motivation Gains of Inferior Group Members. Journal of Applied Social Psychology , 38 (5), 1329 1363. doi:10.1111/j.1559 - 1816.2008.00350.x Ingham, A. G., Levinger, G., Graves, J., & Peckham, V. (1974) . The Ringelmann E Studies of Group Size and Group Performance. Journal of Experimental Social Psychology , 10 , 371 384. Irwin, B.C., Feltz, D.L., & Kerr, N.L. (2013). Silence is Golden: Effect of Encouragement in Motivating the Weak Link in an Online Exercise Video Game. Journal of Medical Internet Research, 15(6), e104. doi:10.2196/jmir.2551 Irwin, B. C., Scorniaenchi, J., Kerr, N. L., Eisenmann, J. C., & Feltz, D. L. (2012). Aerobic Exercise Is Promoted when Individual Performance Affects the Group: A Tes t of the Köhler Motivation Gain Effect. Annals of Behavioral Medicine, 44 (2), 151 - 159. Jarvenpaa, S. L., & Leidner, D. E. (1999). Communication and trust in global virtual teams. Organization Science, 10 (6), 791 - 815. Karau, S., Williams, K. D., Bourgeo Meta - Analytic Review and Theoretical Integration. Interpersonal Relations and Group Processes , 65 (4), 681 706. Karau, S. J., & Williams, K. D. (1997). The effects of group cohesiveness on social loafing and social compensation. Group Dynamics: Theory, Research, and Practice, 1 (2), 156 - 168. Kendzierski, D., & DeCarlo, K. J. . (1991). Physical Activity Enjoyment Scale: Two validat ion studies. Journal of Sport & Exercise Psychology . Kerr, N. L., Feltz, D. L., & Irwin, B. C. (2012). To pay or not to pay? Do extrinsic incentives alter the Köhler group motivation gain? Group Processes & Intergroup Relations . doi: 10.1177/136843021245 3632 Kerr, N. L., Messé , L. A., Dong - Heon, S., Sambolec, E. J., & et al. (2007). Psychological Mechanisms Underlying the Köhler Motivation Gain. Personality and Social Psychology Bulletin, 33 (6), 828 - 841. Kerr, N. L., Messé , L. A., Park, E. S., & Sambol ec, E. J. (2005). Identifiability, performance feedback and the Köhler Effect. Group Processes and Intergroup Relations, 8 (4), 375 - 390. 95 Kerr, N. L., & Bruun, S. E. (1983). Dispensability of member effort and group motivation losses: Free - rider effects. Jo urnal of Personality and Social Psychology , 44 (1), 78 94. doi:10.1037//0022 - 3514.44.1.78 Kerr, N. L., & Seok, D. and group motivation gain. Journal of Managerial Psychology, 26 (3 ), 205 - 218. Kerr, N. L., Seok, D. - H., Poulsen, J. R., Harris, D. W., & Messé, L. A. (2008). Social ostracism and group motivation gain. European Journal of Social Psychology, 38 (4), 736 - 746. Kerr, N. L., Forlenza, S. T., Irwin, B. C., & Perpetual vs. intermittent inferiority and the Köhler group motivation gain in exercise groups. Group Dynamics: Theory, Research, and Practice , 17 (2), 67 80. doi:10.1037/a0031588 Kohl, H. W., 3rd, Craig, C. L. , Lambert, E. V., Inoue, S., Alkandari, J. R., Leetongin, G., & Kahlmeier, S. (2012). Physical Activity 5: The pandemic of physical inactivity: global action for public health. The Lancet, 380 (9838), 294 - 305. Köhler , O. (1926). Kraftleistungen bei Einzel - und Gruppenabeit [Physical performance in individual and group situations]. Industrielle Psychotechnik, 3 , 274 - 282. Köhler , O. (1927). Über den Gruppenwirkungsgrad der menschlichen Körperarbeit und die Bedingung optimaler Kollektivkraftreaktion [Human physical performance in groups and conditions for optimal collective performance]. Industrielle Psychotechnik, 4 , 209 - 226. Kravitz, D. a., & Martin, B. (1986). Ringelmann rediscovered: The original article. Journal of Personality and Social Psychology , 5 0 (5), 936 941. doi:10.1037//0022 - 3514.50.5.936 Latané and Consequences of Social Loafing. Journal of Personality and Social Psychology , 37 (6), 822 832. Latané , B., & Zip f, G. K. (1981). The Psychology of Social Impact, (April), 343 356. Lee, I. M., Shiroma, E. J., Lobelo, F., Puska, P., Blair, S. N., & Katzmarzyk, P. T. (2012). Effect of physical inactivity on major non - communicable diseases worldwide: an analysis of burden of disease and life expectancy. The Lancet, 380 (9838), 219 - 229. Lount, R. B., Jr., Messé , L. A., & Kerr, N. L. (2000). Trying Har der for Different Reasons: Conjunctivity and Sex Composition as Bases for Motivation Gains in Performing Groups. Zeitschrift fur Sozial Psychologie, 31 (4), 221 - 230. Lount, R. B., Park, E. S., Kerr, N. L., Messé , L. a., & Seok, D. - H. (2008). Evaluation Co ncerns and the Köhler Effect: The Impact of Physical Presence on Motivation Gains. Small Group Research , 39 (6), 795 812. doi:10.1177/1046496408323215 96 Lyons, E. J., Tate, D. F., Ward, D. S., Bowling, J. M., Ribisl, K. M., & Kalyararaman, S. (2011). Energy expenditure and enjoyment during video game play: differences by game type. Medicine & Science in Sports and Exercise, 43 (10), 1987 - 1993. McAuley, E., Wraith, S., & Duncan, T. E. (1991). Self - efficacy, perceptions of success, and intrinsic motivation for exercise. Journal of Applied Social Psychology, 21 (2), 139 - 155. McNair, P., Depledge, J., & Stanley, S. (1996). Verbal encouragement: effects on maximum effort voluntary muscle: action. , (December 1995), 243 245. Retrieved from http://bjsportmed.com/content/30/3/243.abstract Messé, L. A., Hertel, G., Kerr, N. L., Lount, R. B., & Park, E. S. (2002). Knowledge of partner's ability as a moderator of group motivation gains: An exploration of the Köhler discrepancy effect. Journal o f Personality and Social Psychology, 82 (6), 935 - 946. doi: 10.1037/0022 - 3514.57.2.239 Messé, L. A., Hertel, G., Kerr, N.L., Lount, R.B.J., Park, E.S. (2002). Knowledge of partner's ability as a moderator of group motivation gains: An exploration of the Köh ler discrepancy effect. Journal of Personality and Social Psychology, 82 (6), 935 - 946. Moffatt, R. J., Chitwood, L. F., & Biggerstaff, K. D. (1994). The influence of verbal encouragement during assessment of maximal oxygen uptake. The Journal of Sports Medicine and Physical Fitness , 34 (1), 45 9. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/7934010 Montani, J. - P., Carroll, J. F., Dwyer, T. M., Antic, V., Yang, Z., & Dulloo, a G. (2004). Ectopic fat storage in heart, blood vessels and kidneys in the p athogenesis of cardiovascular diseases. International Journal of Obesity , 28 , S58 65. doi:10.1038/sj.ijo.0802858 (2011). Lipotoxicity causes multisystem o rgan failure and exacerbates acute pancreatitis in obesity. Science Translational Medicine , 3 (107), 1 26. doi:10.1126/scitranslmed.3002573 Ouellette, J. a., & Wood, W. (1998). Habit and intention in everyday life: The multiple processes by which past beha vior predicts future behavior. Psychological Bulletin , 124 (1), 54 74. doi:10.1037//0033 - 2909.124.1.54 Patrick, W., Elaine, H., Holly, R., & James, O. (2005). Obesity , Inactivity , and the Prevalence of Diabetes and ... Diabetes Care , 28 (7), 1599 1603. P lante, T., & Madden, M. (2010). Effects of perceived fitness level of exercise partner on intensity of exertion. Journal of Social Sciences , 6 (1), 50 54. Retrieved from http://my.psychologytoday.com/files/attachments/34033/jssarticle.pdf 97 Peng, W., Lin, J. H., & Crouse, J. (2011). Is playing exergames really exercising? A meta - analysis of energy expenditure in active video games. Cyberpsychology, Behavior and Social Networking, 14 (11), 681 - 688. Perdue, C. W., Dovidio, J. F., Gurtman, M. B., & Tyler, R. B. (1990). Us and them: Social categorization and the process of intergroup bias. Journal of Personality and Social Psychology , 59 (3), 475 486. doi:10.1037//0022 - 3514.59.3.475 Rhodes, R. E., Jones, L. W., & Courneya, K. S. (2002). Extending the theory of planned behavior in the exercise domain: A comparison of social support and subjective norm. Research Quarterly for Exercise and Sport, 73 (2), 193 - 199. Ryan, R. M., & Deci, E. L. (2000). Self - determination theory and the facilitatoin of intrins ic motivation, social development, and well - being. American Psychologist, 55 (1), 68 - 78. Journal of Personality and Social Psychology , 42 (2), 281 291. Psychological Bulletin , 113 (1), 67 81. Sillars, A., Shellen, W., McIntosh, A., & Pomegranate, M. (1997). Relational characteristics of language: elaboration and diffe rentiation in marital conversations. Western Journal of Communication, 61 (4), 403 - 422. Simmons, R. a, Gordon, P. C., & Chambless, D. L. (2005). Pronouns in marital interaction. Psychological science , 16 (12), 932 6. doi: 10.1111/j.1467 - 9280.2005.01639.x Smith, S., & Schoene, D. (2012). The use of Exercise - based videogames for training and rehabilitation of physical function in older adults. Aging Health, 8 (3), 243 - 252. Soltani, P., & Salesi, M. (2013). Effects of Exergame and Music on Acute Exercise Re sponses to Graded Treadmill Running. Games for Health Journal , 2 (2), 75 80. doi:10.1089/g4h.2012.0077 Son, V., Jackson, B., Grove, J. R., & Feltz, D. L. (2011). "I am" versus "we are": Effects of distinctive variants of self - talk on efficacy beliefs and m otor performance. Journal of Sports Sciences, 29 (13), 1417 - 1424. Steiner, I. D. (1972). Group process and productivity, New York: Academic Press. Stroebe, W., Diehl, M., & Abakoumkin, G. (1996). Social compensation and the Köhler effect: Toward a theore tical explanation of motivation gains in group productivity. Understanding group behavior, Vol. 2: Small group processes and interpersonal relations. p. 37 - 65. Turner, J. C., Oakes, P. J., Haslam, S. A., & McGarty, C. (1994). Self and Collective: Cognition and Social Context. Personality and Social Psychology Bulletin , 20 (5), 454 465. 98 Trojano, R. P., Berrigan, D., Dodd, K. W., Masse, L. C., Tilert, T., & McDowell, M. (2008). Physical activity in the Unite d States measured by accelerometer. Medicine & Science in Sports and Exercise, 40 (1), 181 - 188. U.S. Department of Health and Human Services. (1996). Physical Activity and Health: A Report of the Surgeon General. Vargas - Tonsing, T. M., Myers, N. D., & Felt z, D. L. (2004). Coaches' and Athletes' Perceptions of Efficacy - Enhancing Techniques. Sport Psychologist, 18 (4), 397 - 414. Wang, G., Pratt, M., Macera, C. a, Zheng, Z. - J., & Heath, G. (2004). Physical activity, cardiovascular disease, and medical expendit ures in U.S. adults. Annals of Behavioral Medicine , 28 (2), 88 94. doi:10.1207/s15324796abm2802_3 Watson, P. J., Taylor, D., & Morris, R. J. (1987). Narcissism, sex roles, and self - functioning. Sex Roles , 16 (7 - 8), 335 350. Weber, B., & Hertel, G. (2007). Motivation gains of inferior group members: A meta - analytical review. Journal of Personality and Social Psychology, 93 (6). Wei, M., Kampert, J. B., Barlow, C. E., Nichaman, M. Z., Gibbons, L. W., Paffenbarger, R. S., & Blair, S. N. (1 999). Relationship between low cardiorespiratory fitness and mortality in normal - weight, overweight, and obese men. Journal of the American Medical Association , 282 (16), 1547 53. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/10546694 Weinberg, R. S., Gould, D., & Jackson, A. (1979). Expectations and performance: An empirical - efficacy theory. Journal of Sport Psychology , 1 (4), 320 331. Weinberg, R. S., Yukelson, D., & Jackson, A. (1980). Effect of public and priv ate efficacy expectations on competitive performance. Journal of Sport Psychology , 2 , 340 349. Retrieved from http://psycnet.apa.org/psycinfo/1981 - 07981 - 001 Weinberg, R. S., Gould, D., Yukelson, D., & Jackson, A. (1981). The effect of preexisting and mani pulated self - efficacy on a competitive muscular endurance task. Journal of Sport Psychology , 3 (4), 345 354. Wen, C. P., & Wu, X. (2012). Stressing harms of physical inactivity to promote exercise. Lancet , 380 (9838), 192 3. doi:10.1016/S0140 - 6736(12)60954 - 4 Wen, C. P., Wai, J. P. M., Tsai, M. K., Yang, Y. C., Cheng, T. Y. D., Lee, M. - (2011). Minimum amount of physical activity for reduced mortality and extended life expectancy: a prospective cohort study. Lancet , 378 (9798), 1244 53. doi:10.101 6/S0140 - 6736(11)60749 - 6 White, K., Schofield, G., & Kilding, A. E. (2011). Energy expended by boys playing active video games. Journal of Science and Medicine in Sport, 14 (2), 130 - 134. 99 Williams, K. D., & Karau, S. J. (1991). Social loafing and social comp ensation: The effects of expectations of co - worker performance. Journal of Personality and Social Psychology, 61 (4), 570 - 581. Wininger, S. R., & Pargman, D. (2003). Assessment of factors associated with exercise enjoyment. Journal of music therapy , 40 (1) , 57 73. Wise, J. B., & Trunnell, E. P. (2001). The Influence of Sources of self - efficacy upon efficacy strength. Journal of Sport & Exercise Psychology, 23 (4), 268 - 280. Worchel, S., Rothgerber, H., Day, E. a, Hart, D., & Butemeyer, J. (1998). Social identity and individual productivity within groups. The British journal of social psychology / the British Psychological Society , 37 ( Pt 4) , 389 413. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/9933910 ary, J., & Balkin, J. (1989). Shame, guilt, narcissism, and depression: Correlates and sex differences. Psychoanalytic Psychology , 6 (2), 217 230. doi:10.1037//0736 - 9735.6.2.217 Zajonc, R. B. (1965). Social facilitation. Science, 149 (3681), 269 - 274.