. .:._. A. ~..¢§> ... it ‘ . 5.... fittx .) .... K «.Iuhmhh1. 5.5%}? 1.3.“ I |‘vll| - _ .131 ..IL. .- ulbl. \.|-I’ ‘ . ~\l‘flu.‘lp.vl.,llv ‘ . ‘. .. 9-. . II- v.§.n'$n. .vnHP‘IQJo. p u .‘L?! ‘0 . n . i .OV.Y¢l oui'l'» 13...! ..04\IA |I0~I .' «tll‘tcifl \- .. ‘1 . g...‘l\‘ o a\l. ‘ .~‘v.l f.‘nll. ‘Lu yv‘ti‘! - I...» 1vll1 1Ivv‘ . .l...b. {.II. o? t I I; . ‘bl ‘: .va.&.... v. . I..3:.il..va,....l..9nf .‘ . . 3.3!... (1% 2,111,}: . .lr.vv\. «:.. .... . r. c . .tl._1olh/ . ;. ...: .23... um“. inflilnflflz . . , -IIII. ‘ ..... . I: In gtl-bfllzlnf ‘l 131:», «Haul Iavultl...‘ .1». ... . , W? ‘ "Fa PW litv'fll. u Quilt“ :..-1‘ w: Q‘l?cxlx ‘VOLI\.O. .!)|. . ... ...: .. . ~1‘.Yn‘olu‘ .0..lln . .. . . . :..... I u _ .. . Lia... ....N.J-v.\.\ . , ., ...i,{ o! . . .. I . u .v . . . . . b v .v 4‘15)“, . i .. w I u. . I lu'l Jcik 1 I. 7:) t rlo v. {11- ..i ‘v \a ..‘Ifitl I... ... nod-.51 ”~84! 1!) vllv J41): u)! \.. . ‘VCITJ‘ ‘11}..11 I’l. ‘ ....|.la..\..‘.d . . , .3... ‘ . .. ...-ux). 1..., .I ....i L . . . VIV‘ON4“ . ‘1' II II rt 7!! Itl IA . y ; Ubnoflflfhhr h... \ ....l-..n...buu.....c ind... 4i: ... vi mqtuirflflfibliflth~¢ I to O wright}? ..4 . L LIBRARY ‘mmmmMinimum:1”anam]c “aim“ This is to certify that the dissertation entitled The Impact of Rock Music Video Priming on Appraisal of a Stereotyped Male—Female Interaction presented by Christine Hall Hansen has been accepted towards fulfillment of the requirements for __'Pll_-2o_degree in Psychology Mn njor professor om 7/26/57 ‘ «.N‘i‘ A if MSU ‘ RETURNING MATERIALS: Place in book drop to LJBRARJES remove this checkout from .—:—. your record. FINES will be charged if book is returned after the date stamped below. " 4 ‘ "N » *4 ”IN“ P" ' ' L ., .. . -.. 1 (>0. 1-; f :../g to Q 6’ 0 4 APR 3&0 1:3 . 'M \ l ‘3)“ h 1‘. Lfiij} :51 . . JANE) 201:0 2 n , At; . “ ' L L . ”S’fiié‘ifiil m... 141193463? 290? 'W n A mm- s _. “vr Lg! o '\ ‘1 .‘ kP 7575:}? if"; .5 in; g 5“; R7 is? {647 053550510 1 074‘ “‘3 T THE IMPACT OF ROCK MUSIC VIDEO PRIMING ON APPRAISAL OF A STEREOTYPED MALE-FEMALE INTERACTION BY Christine Hall Hansen A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements tor the degree of DOCTOR OF PHILOSOPHY Department of Psychology 1987 Copyright by CHRISTINE HALL HANSEN 1987 ABSTRACT "II-IE IMPACT OF ROCK MUSIC VIDEO PRIMING ON APPRAISAL OF A STEREOTYPED MALE-FEMALE INTERACTION By Christine Hall Hansen An experiment was conducted to explore the impact of brief exposure to rock music videos on subsequent appraisal of a male and a female. Rock videos were selected to prime one of two sex-role stereotypic schemas: a socially harmonious ”boy-meets-girl" or disharmonious ”boy-dumps-girl" schema. Nonstereotypic (neutral) videos were used as a control. After the video prime, subjects watched a videotaped, scripted interaction in which the behaviors of the male and female actors were congruent (consistent) with one of the two stereotypic primes, resulting in a 3 (Prime: Schema-consistent, Schema-inconsistent, Schema-irrelevant) X 2 (Interaction Script: Boy-meets-girl, Boy-dumps—girl) between-subjects factorial design. Past research (e.g. Hastie, 1981) has suggested that the salience of observed behaviors is increased when behaviors are inconsistent with primed schemas, altering the importance of these behaviors in impressions and enhancing recall. Schema-inconsistency was hypothesized to produce (1) more positive appraisal of harmonious behaviors and (2) more negative appraisal of disharmonious behaviors. Subjects also performed a "unitizing' task (e.g. Newtson, 1973), pressing a button to chunk behaviors into ”meaningful” units. Results showed that subjects generated finer units for schema-inconsistent behaviors, and finer units were associated with greater recall. Schema-inconsistency. but not unitization, influenced impressions. Contrary to predictions, schema-consistency enhanced favorability of both interactions. The boy-meets-girl videos enhanced positive impressions of the harmonious interaction, while the boy-dumps-girl videos attenuated negative impressions of the disharmonious interaction. Results were discussed in terms of the need to address the valence as well as the content of primed schemas to understand assimilation and contrast effects. It was suggested that unitizing may serve to increase attention to behavioral details but may not represent normal schematic encoding processes. This dissertation is dedicated with deep respect to my mentors and with love to my daughters. ACKNOWLEDGEMENTS I wish to acknowledge David Lee Roth, Huey Lewis, and my buddy Jon (Jon Bon Jovi) for providing the impetus for this experiment. My sincere thanks to the members of my committee-- Joel Aronoff, Linda Jackson, Norbert Kerr, and Lawrence Messé--for allowing me to apply what I learned from them to the domain of media influence. TABLE OF CONTENTS List of Tables List of Figures Introduction Schemas As an Explanatory Concept The Impact of Television on Aggression: Past Findings The Impact of Television on Aggression: Berkowitz's Priming Theory The Impact of Television on Aggression: Theoretical Additions to the Priming Approach Media Priming Effects on Sex-Role Stereotypes: Past Findings Extending Priming to Sex-Role Stereotypes Schematic Processing: A Framework for Understanding Social Influences Media Priming of Sex-Role Stereotypes: Research Problems with the Previous Priming Research Some Properties and Consequences of Priming The Current Experiment Solving the Problems with the Earlier Research Priming in the Current Experiment Predicted Effects of Priming on Trait Inferences Predicted Effects of Priming on Unitizing of the Interactions Predicted Effects of Priming on Memory for Unitized Interactions Method vii 10~/ 13¢ 16 / 19 x/ 21 25 25 26 27 29 32 Priming Stimuli: Selection of Videos Interaction Sequences Appraisal of Prime-Script Consistency, Inconsistency, and lrrelevancy Dependent Measures for the Actual Study Procedure, Cover Story. and Instmctions Unitizing Pilot Study Experimental Design and Procedure Judges and Coding Transcribing and Coding of Unitizing Responses Results Preliminary Experiment: Script-Prime Consistency Nonunitizing Pilot Study Unitizing Experiment: Trait Judgments Favorability Getting Acquainted and Audition Segment Favorability Minimally Structured Recall Structured Recall Unitizing Relationship of Unitizing to Recall Relationship of Unitizing to Trait Judgments Discussion Impact of Rock Music Video Primes Did Unitization Fail? List of References 35V 37 39 42 45 46 46 47 49 49 52 53 65 72 76 84 88 90 94 94 99 106 Appendix A: Priming Condition Videos Appendix B: Interaction Scripts Appendix C: Prime and Script Stories Appendix D: Minimally Structured Recall Measure Appendix E: Trait Scales Appendix F: Impression Favorability Scales Appendix G: Structured Recall Measures Appendix H: Cover Story and Instructions Appendix I: Consent Form Appendix J: Partial Debriefing Sheet Appendix K: Scoring and Coding Minimally Structured Recall Appendix L: Scoring and Coding Structured Recall Appendix M: Coded Unitizing Segments Appendix N: Written Debriefing Sheet vi 121 122 131 147 149 151 153 157 160 161 162 166 176 180 10. 11. 12. 13. 14. LIST OF TABLES . Consistency of Scripted Conversations with Rock Music Video Primes Summary of Conscious Appraisals of the Relationships of Script-Stories to Prime-Stories Intercorrelations Among Factors for Male and Female Impressions Pattern of Factor Loadings for Individual Female Trait Judgments Pattern of Factor Loadings for Individual Male Trait Judgments Means for Script Effects on the Female‘s Traits Main Effects of Video Prime on Judgments of the Female's Traits Prime X Script Interaction on Ratings of "Mean and Nasty“ for the Female Prime X Script Interaction on Ratings of ”Openness" for the Female Attenuation and Reversal of Script Effects Produced by Consistent and Inconsistent Video Primes on Trait Ratings of the Female's Positivity and Ability Prime X Script Interaction on Ratings of “Nice Guy" for the Male Prime Main Effects on Overall Favorability on Impressions of the Two Actors Prime X Script Interactions on Overall Favorability of lrrpressions of the Female Actor Main Effects of Video Prime on Favorability of Getting Acquainted and Audition Impressions of the Female and Male Actors vii 50 50 55 56 57 59 60 62 62 65 66 67 15. Prime X Script Interactions on Favorability of Getting Acquainted and Audition Impressions of the Female Actor 16. lnterjudge Reliability on Minimally Structured Recall Categories 17. Video Prime Main Effects on Minimally Structured Recall 18. Video Prime Main Effect on Minimally Structured Recall for All Behaviors Attributed to the Female and to the Male 19. lnterjudge Reliability for Accuracy of Structured Recall Measures 20. Video Prime Main Effects on Accuracy of Stmctured Recall Statements 21. Prime X Scr‘pt Interaction on Structured Recall for the Female's Reciprocation or Nonreciprocation on Statement Four 22. Prime X Script Interaction on Structured Recall for the Male's Praise or Derogation on Statement Five 23. lnterjudge Reliability for Strength of Structured Recall Measures 24. Script Main Effects on Accuracy of Structured Recall Statements 25. Prime X Script Interactions on Structured Recall for Three of the Female's Four Reciprocation and Nonreciprocation Statements 26. Main Effects of Self-Described Unitizing Strategy on Number of Units During the Getting Acquainted Segment 27. Contingency Table of Self-Described Unitizing Strategy X Experimental Condition Indicating Noninterdependence of Strategy and Condition 28. Prime Effects on Unitizing During Portions of the Interaction When the Male was”Coming On,” When the Female Was Reciprocating or Not Recbrocating, When the Male Was Praising or Derogating, and During the Residual viii 69 73 75 76 77 78 79 80 81 82 83 84 85 87 29. Multiple Linear Regressions to Predict Minimally Structured Recall from Unitizing Within Portions of the Interaction 30. Partial Correlations of Unitizing During Statement with Accuracy of Structured Recall for Statement 31. Partial Correlations of Unitizing During Statement with Strength of Stmctured Recall for Statement 32. Multiple Linear Regressions to Predict the Female'sTraits from Unitizing Within Portions of the Interaction 89 91 92 93 LIST OF FIGURES 1. The Effects of Prime, Reciprocation, and Counteraction on Overall Impression Favorability (Hansen & Hansen, 1986) 17 2. Process Model Showing Relationships Among Recall, Unitizing, and Inpressions 103 INTRODUCTION There is a vast literature on the psychological effects of movies, television, and other mass communications media, such as magazines and newspapers. Given the great number of published studies, surprisingly little is really known about the cognitive and behavioral impact of these media stimuli. Most research studies have attempted to look at the impact of television on two developmental trends: aggressive behavior and sex-role orientation. Viewing televised portrayals of aggressive characters and acts of violence appears to have a causal role in the development of a more aggressive personal style (Singer 8. Singer, 1980; Singer, Singer, & Rapaczynski, 1984). Similarly, exposure to stereotypic TV depictions of males and females seems to abet the sex-role stereotypic socialization process of children, helping to produce boys and girls whose behaviors, self-concepts, and stereotypes may be discriminated along a dimension of masculinity and femininity (Eisenstock, 1984: Gerbner, 1978; Piliavin 8 Unger, 1985) To date, there has been very little conceptual integration of the findings within the two seemingly diverse areas of research (sex-roles and aggression), except for the commonsense notion that televised portrayals are a source for Ieaming relevant attitudes and behaviors toward both aggression and sex-roles. The goal of this early section is to integrate many of the research findings bearing on the effects of exposure to media on both aggression and sex-role stereotypes into a simple, theoretically-grounded view of the impact of media on social thinking. The research can be organized around the hypothesis that media-portrayed social behaviors of all types, including sex-role stereotypic and aggressive portrayals, can serve as priming stimuli for social schemas that already are represented in memory. When schemas have been primed by encoding media portrayals, they are more likely to be used to encode future social behavior. This hypothesis, then. bears on media-induced perpetuation of social schemas rather than on the acquisition of these schemas. W The term schema is a hypothetical construct which describes permanent knowledge structures pertaining to people, things, abstract concepts, and events (Fiske & Taylor, 1984). Social schemas contain our stored prior knowledge about frequently encountered, familiar, or irrportant social concepts, such as aggression or sex-roles. We know from information processing research that appraisal of social stimuli most often, if not always, occurs in the context of social schemas of which the stimulus is representative. Perhaps the simplest way of describing schematic appraisal is to think of a schema as an association of all the stored information we have about a particular concept; all its important attributes are stored together in memory as a schema. When we encounter a social stimulus (most often a person or a behavior), an appropriate schema is triggered from memory for use in encoding the new information. Because schemas contain a lot of information about attributes of the stimulus, appraisal is greatly facilitated. Perceivers are not forced to releam the attributes of a stimulus each time it is encountered; the attributes are stored in memory in the schema, and all the perceiver needs to do is match the stimulus to its appropriate schema, and, immediately, the stimulus is understood and appraised. Schemas are indeed useful, then, in taming impressions of people from observing their behavior, because matching an observed behavior to a schema provides the appraiser with a great deal of information about the meaning of the behavior and about the sort of person who would perform it, as well as what can be expected from the type of person seen performing the behavior. For instance, suppose we observe a man hitting his wife (Shotland & Straw, 1976). Because a schema for “wife beater" has been triggered in memory, an instant Impression of him can be formed. We could now generate a complete picture of this “wife beater“ from schematic attributes, and we would have a fairfy useful idea of the kinds of traits and behaviors to expect from this person in the future. Schemas tmly turn the perceivers' experience from a chaotic stream of aimless events into a more predictable and orderly wortd. It is proposed here that in the course of cognitively processing televised social behavior, schemas from memory are activated and used to understand and interpret the current social behaviors, to forrn an impression of the person performing the behavior, and to provide an expectancy for future behavior. The current research will attentpt to show that media portrayals (like real-life observed behaviors) can serve to activate social schemas, and once the schema is accessible, interpretation of subsequent social behaviors can also be altered by the kind of schema that has been primed. As part of an ongoing investigation by the researcher, sex-role stereotypic rock music videos (from Music Television-MTV) were used as priming stimuli. One goal of the present research was to test the effects of stereotypic rock music primes on subsequent appraisal of a man and woman during a social interaction. And, as stated earlier, another purpose of the current dissertation was to offer an alternative conceptualization of past research on the effects of televised portrayals of social behaviors--a priming hypothesis. To that end, it was necessary first to look at the work that has already been done as well as previous theoretical and conceptual underpinnings for some of the important findings. In reviewing the literature in the areas of aggression and sex-roles, it will become evident that (1) priming has substantial explanatory power and heuristic value. and (2) current accounts of media-induced effects on aggression and sex-roles can be strengthened with a few theoretical additions. The majority of published studies have reported a link between viewing television violence and subsequent aggressive behavior (Andison, 1977; Comstock, 1980; Geen, 1976, 1983). In his review of the literature from 1956-1976, Andison presented three theoretical positions used to explain research effects (or lack of effects): (1) aggressive television has a cathartic effect. reducing aggressive behavior in viewers; (2) aggressive or violent television has no significant effect; and (3) televised violence acts as a stimulus promoting aggressive behavior. His (1977) metanalysis supported the third view; there appears to be at least a weak relationship between watching violence on TV and behaving aggressively. Much of the correlational and survey research has centered on the effects of violent programming on devel0ping aggression in children. In longitudinal studies, for example, (9.9. Eron, Huesmann, Lefkowitz, 8 Walder, 1972; Lefkowitz, Eron, Walder, 8 Huesmann, 1977; Singer 8 Singer, 1980; Singer, Singer, 8 Rapaczynski, 1984), watching violence on television has been found to predict later aggressive behavior. In a typical study, Eron et al. (1972) found (in a cross-lagged panel analysis) that, for boys, preference for violent TV at age eight or nine was a good predictor of peer ratings of aggressiveness during the last year in high school; however, the amount of aggression the eight or nine year old child engaged in appeared not to be predictive of his later TV viewing preferences. While results from studies using this type of analysis should be interpreted cautiously (see Rogosa, 1979), Eron et al.'s findings have been interpreted as supporting the hypothesis that television violence produces aggressive behavior--rather than the alternative hypothesis that preference for television violence results from an aggressive personality disposition. More recent correlational studies (e.g. Eron 8 Huesmann, 1980; Singer 8 Singer, 1979) have extended this effect to girls. The most influential theoretical conceptualization for findings such as these has been observational learning and modeling (Bandura, 1965, 1973; Geen, 1976). According to this explanation, violent portrayals may (a) model or teach new aggressive behaviors and (b) reduce restraints against imitating aggression; imitated aggressive behaviors are then maintained by positive reinforcement (positive outcomes from aggression). For learning aggressive behaviors, television models may indeed be a potent force, especially when aggressive television models are rewarded by positive outcomes from their aggression (Zillmann, 1984). Observational learning of aggression, which, incidentally, occurs in both children and adults (Comstrock, Chaffee, Katzman, McCombs, 8 Roberts, 1979), is likely to be one major source of schematic information about aggressive concepts. As noted by Singer, Singer, 8 Rapaczynski (1984), exposure to television violence has indirect effects as well as direct behavioral effects. Being exposed to violent portrayals on TV can lead children to develop a belief that the world is han'nful and unsafe. Gerbner 8 Gross (1976) present evidence that the attitudes of adults, too, may be similarly affected. Adult heavy viewers are more likely to report that they are mistmstful of others, in general, and to overestimate their risk of being personally involved in violence. There is also a strong suggestion that in children, watching a lot of television makes them less likely to report aggressive behavior (playground behavior) to teachers or playground supervisors ( Gerbner, Gross, Morgan, 8 Signiorelli, 1980). Similar effects have been found in adults, suggesting that they, too, may become more tolerant of aggression after long periods of exposure (Horton, Lippincott, 8 Drabman, 1977). Thus, it seems that the attitudes of both children and adults can be negatively affected over the long mn by exposure to violem television programs. But, even short-term exposure to violent television or to movies produces differences in aggressive behavior. For example, both Meyer (1972) and Rosene (1971) found that exposing subjects to one violent film resulted in delivery of significamly more electric shocks to an annoying target than did exposure to a neutral film or no film at all. It could not be argued, then, that a neutral film lowered aggression, but, rather, watching even one violent movie increased aggressive behavior. Similar aggression-facilitating effects have been reported frequently in the literature on the effects of exposure to violent pornography (e.g. Donnerstein, 1983; Malamuth 8 Donnerstein, 1982, 1983; Zillmann, 1984). Observational learning theory can account nicely for the creation of beliefs about violence and learning of aggressive behaviors (i.e. aggression schemas), but it does not provide the most parsimonious explanation for short-term effects, such as those described above. Reviewing the numerous studies which have explored the effects of violent television could leave one with the feeling that television is, universally, a negative influence in our society. In reality, television also can have beneficial effects. Quite a few studies have been conducted to measure the impact of "positive" programming on children (9.9. Coates, Pusser, 8 Goodman, 1976; Friedrich 8 Stein, 1975; Stein 8 Friedrich, 1972). While the particular children's program examined (such as ”Sesame Street” or ”Mister Roger's Neighborhood”) produced some differences, exposure to both shows increased the children's subsequent prosocial behavior. Again, observational learning has been used to explain these effects, a perspective that is consistent with a tendency for stronger effects to occur when the television models receive a positive outcome after behaving in a prosocial way and when the children, themselves, are reinforced for performing the positive behaviors. While prosocial television can actually reduce aggressive behaviors, it is important to note that nonselective and heavy viewers are exposed to many more violent than nonviolent or prosocial programs (Gerbner, Gross, Morgan, 8 Signiorelli, 1986); and because of the sheer prevalence of violence on television, television's negative effects probably outnumber its positive or prosocial effects. Other interesting theories have been advanced in the aggression literature to account for the findings produced by exposure to violent television and films, such as Berkowitz's (1965) frustration-aggression model and Huesmann's (1982) information-processing theory. Berkowitz has proposed a drive model which posits that frustration (produced by being insulted, for example) produces a tendency toward overt aggressive behavior. Observing violence, he argued, could act as a trigger for this kind of drive-like aggressive behavior. Huesmann's hypothesis, on the other hand, extends the earlier observational Ieaming explanation into the domain of cognitive information processing. He views observational learning as an example of encoding specificity, meaning that because aggressive behaviors from violent TV are encoded in memory in a specific context (the circumstances contained in the program), recall of the information and imitated aggressive behaviors are most likely to occur when the aggression-provoking situation closely resembles the original encoding context. His theory, then, is not dissimilar from a statedependent Ieaming model (Bower 8 Cohen, 1982), but it resembles Berkowitz's early model in one important respect. Both models agree that television violence can serve as a cue which triggers aggressive behavior. Quite recently, however, Berkowitz (Berkowitz 8 Rogers, 1986) has proposed an explanation for media effects that bears greater resemblance to Huesmann's theory. Berkowitz has moved away from a frustration—aggression explanation for aggressive behavior toward a hypothesis that places the greatest emphasis on the specific cues that trigger aggression. Berkowitz now strongly advocates a priming theory for aggression. His term for knowledge structures is somewhat different (he prefers the term “associative network" to "schema” ), but his proposal shares many similarities with the approach taken in the current work. Berkowitz proposes that aggressive cues (real-life or from the media) serve to prime prior aggressive thoughts and feelings in memory, as well as their associated behaviors, resulting in increased aggression. In most respects, his explanations parallel those advocated here: aggressive media cues trigger prior knowledge structures, or schemas, which can evoke information stored in memory as well as an "action tendency” toward aggression. The well-reasoned arguments put forth by Berkowitz (Berkowitz 8 Rogers, 1986) for a priming explanation, however, are limited to explaining a single act of behavior, in which extemal cues “match” information which is stored in memory, thus triggering all the associated information and action tendencies from memory. According to Berkowitz, media violence can provide such a cue: it might be an aggressive character with which the viewer identifies, or it might be aggressive words or actions, or an aggressive object, such as his well-known “guns on the table“ (Berkowitz 8 LePage, 1967). When the cue corresponds with a memorial representation, it can trigger the associated information stored in memory, producing aggressive behavior. Berkowitz's new approach can also account for reduced concern toward observed aggressive acts (discussed earlier in this section) by assuming that the more frequently aggression is primed, the less physiologically aroused the person becomes when viewing violence. Physiological arousal produced from exposure to aggression could become less and less with frequent priming (i.e. for heavy viewers), and physiological reactivity could become lower when watching aggressive acts. This pattern of reduced arousal is precisely what Thomas, Horton, Lippincott, and Drabman (1977) and Geen (1981) found in their investigations of desensitization to aggressive portrayals. It is important to note here, though, that lowered physiological reactivity does not mean less overt aggression. Heavy viewers, while less physiologically aroused when exposed to aggressive stimuli, are still more aggressive when provoked (Thomas, 1982). Probably, it simply takes a greater level of aggression to physiologically or psychologically arouse heavy viewers, resulting in their apparent higher tolerance for observed violence. It may be that frequently exposed subjects are less aroused by the appraisal of aggression, whether it is observed aggression or their own tendency toward aggression, so that they are more tolerant of aggression in general. It: no.0: o ' ‘ ' or or an ‘ .'-| I‘0 ' ' . Act-Iron - I: 'Inm AOJLO- I Berkowitz has provided a nice conceptual approach to understanding many of the effects of media violence on aggression. Moreover, his explanation can be expanded to explain and predict other cognitive and behavioral phenomena by integrating it further with current social information processing theories of priming. These additional concepts have been described previously (Hansen 8 Hansen, 1986), but, as they relate directly to the conceptual approach taken in the current research, as well as to Berkowitz's media priming explanation, they are presented here also. The current social information-processing approach to media priming has four major concepts. 1. Priming by media stimuli increases the accessibility of a particular representative schema, such as aggression, and frequently primed schemas can become chronically accessible. 2. In the course of appraising media portrayals, a schema has been made accessible through priming, and subsequently observed social behavior will be more likely to be appraised using the accessible schema, altering appraisal of the behaviors and impressions formed of the people performing them. 3. Observed behavior which is consistent with a primed schema will be automatically assimilated to the accessible schema, resulting in appraisal that more closely resembles schematic content than did the appraised stimulus. 4. Observed behavior which is inconsistent with a primed schema (because it can not be processed schematically) will tend to involve greater conscious attentional processing, resulting in added weight to inconsistent information in impressions and a contrast effect on appraisal. Relating these proposed concepts to priming by media violence, then, suggests additions to Berkowitz's model. First, recent activation through priming will result in subsequently observed social behaviors being more likely to be ”captured by“ the primed aggressive schema. For Berkowitz, the "guns on the table” (whether seen in a violent film or in an aggressor's hand) can trigger ”associative connections among units of feelings and (involuntary or controlled) expressive-motor reactions (page. 60)." The current aproach agrees with Berkowitz, but also suggests that in the course of encoding media violence, the aggressive schema becomes accessible, providing an informational expectancy which makes subsequently observed behaviors more likely to be perceived as aggressive stimuli. Thus, after priming, not only would the ”gun on the table“ be appraised as an aggressive stimulus, but other, more ambiguous cues also would be more likely to be appraised as aggressive stimuli. Carver, Ganellan, Framing, and Chambers (1983) demonstrated this priming process. In their second experiment, they exposed subjects to a hostile or neutral (word list) prime and then, using a classic “learning paradigm” (Milgram, 1960), gave them the opportunity to shock a ”poor Ieamer.” After the hostile prime, subjects punished the learner's errors with shocks that were significamly more intense than those given by subjects exposed to the neutral prime. The prime induction was sufficiently disconnected (in time and purpose) from the Ieaming experiment that it is unlflrely that subjects formed a conscious association between the two events. Nevertheless, the hostile prime produced more aggressive behavior in the subsequent ambiguous situation, and the authors contend that the effect was due to the increased accessibility of a hostile 1O category (i.e., an aggressive schema). Thus, (consistent with the current approach) transient accessibility of an aggressive schema induced by a single, brief prime can make a subsequently encountered ambiguous situation appear more representative of aggressive schematic information--it will not be appraised as ambiguous-producing more aggression in an ambiguous situation. In addition, frequent activation of an aggressive schema by frequent viewing of television and film violence can increase aggressive behavior by making an aggressive schema chronically accessible (Bargh 8 Pratto, 1986; Higgins, Bond, Klein, 8 Strausman, 1984). When a schema is chronically accessible due to frequent priming, it is more easily triggered by environmental cues (Bargh 8 Thein, 1985; Higgins, King, 8 Mavin, 1982). In addition, it appears that chronic and primed (transient) accessibility can summate to increase the likelihood that the schema will be activated and used to encode subsequent stimuli (Bargh, Bond, Lombardi, 8 Tota, 1986). Heavy viewers of violence would be expected to be most susceptible to both kinds of accessibility effects. It is hardly surprising, then, that heavy viewers engage in more frequent hostile appraisals and in more aggressive behavior. In addition to increasing the accessibility of aggressive schemas, watching a lot of violence may produce more complex schemas as well. The aggression schemas of heavy viewers, for example, might contain more stored information relating to aggression and more interconnections between the stored pieces of information (Linville, 1982). Thus, the more elaborate schemas of heavy viewers, with their greater interconnections, would also be expected to increase the probability that heavy viewers will engage in more aggressive appraisals and behaviors by making components of the schema easier to access. III'E"EI[| S-BISI | 'EIE'I' Viewing media aggression has a well-researched influence on aggressive beliefs and behavior. A review of the growing literature on the impact of exposure to sex-role stereotyped media portrayals suggests that the processes hypothesized to account for media violence effects 11 can also be directly applied to understanding the impact of media stereotyping. Whereas media violence leads to aggressive beliefs and behavior, sex-role stereotyped portrayals have been found to produce stereotyped beliefs and behaviors. Sex-role researchers seem to agree that stereotyped media portrayals of men and women, which abound in television, films, and magazines, exert a socially negative influence on viewers by promoting traditional (some say sexist) attitudes and perpetuating the unequal treatment of women in society (e.g. Eisenstock, 1984; Goffman, 1979; McArthur 8 Resko, 1975; Mohr 8 Zanna, 1987; O'Leary, 1977; Piliavin 8 Unger, 1985; Rossi 8 Rossi, 1985). Morgan (1982) sums up the findings of numerous content analyses of television programs by suggesting that ”a symbolic annihilation of women has been accomplished through underrepresentation, overvictimization, and trivialism (page 947)." Women are seen less frequently, with ratios ranging from slightly above 1:1 (males to females) in daytime soap operas to about 3:1 in prime time dramas (Gross 8 Jeffries-Fox, 1978; McNeil, 1975; Tuchman, 1978). And when they are shown, women tend to be portrayed in a traditional, stereotyped way, as having lower status and deferring to the males around them (Morgan, 1982; Piliavin 8 Unger, 1985; Stemglanz 8 Serbin, 1974). All too often, women are emotionally and physically victimized as well (Tuchman, 1978). Both women and men are often depicted in the media as traditionally sex-role stereotyped characters. Children's programs also portray men and women as traditionally sex-typed (Feldstein 8 Feldstein, 1982; Stemglanz 8 Serbin, 1974). Surprisingly, even children's ”educational TV" is sexist. A content analysis by Dohrmann (1975) found that in four popular educational programs (”Mister Roger's Neighborhood", ”Sesame Street”, ”Captain Kangaroo”, and 'The Electric Company“) males were typically characterized as active and masterful, leaders and protectors: whereas females were portrayed as nurturant, passive, dependent, and as followers-and generally of lower status. Television and films (especially with the large numbers of homes with access to cable television and VCRs) are doubtless teaching children sex-roles, and what they are Ieaming is of great concern to many researchers (Tuchman, 1979). 12 There is extensive evidence that exposure to stereotyped media portrayals influences sex-role orientation (e.g. Eisenstock, 1984; Gerbner, 1978; Gross 8 Jeffries-Fox, 1978; Sprafkin 8 Liebert, 1978). As in the media and aggression literature, an observational Ieaming model is often used to explain how both stereotypes (Ashmore 8 Del Boca, 1979) and sex-roles (Atkin 8 Miller, 1975; Gross 8 Jeffries-Fox, 1978; Wolf, 1973) are learned from the medi As Ashrnore and Del Boca (1981) have pointed out, howeverf'a great deal of work has been done to document the stereotyped content of the mass media, but little attention has been paid to the question of how this content influences the beliefs of individuals (page 17)." ( Laboratory experiments designed to look at short-term effects of stereotyped programming on the attitudes and behaviors of both children and adults have shown that even brief exposure alters behavior (Cobb, Stevens-Long, 8 Goldstein, 1982; Jennings, Geis 8 Brown, 1980) and sex-typed beliefs)O'Bryant 8 Corder-Bolz, 1978). For example, Geis, Brown, Jennings, and Porter (1984) were able to influence women's achievement expectations by exposing them to either four traditional (women as homemaker or sex object) or four counterstereotypic (reversed-role, female-dominant) television commercials. After the traditional commercials, essays of female subjects projecting “how you see yourself ten years from now” contained more homemaking themes than achievement themes. After the reversed-role commercials, though, the proportion of achievement to homemaking themes was significantly increased. While Geis et al.'s experiment can be explained (as the authors did) through modeling, I think that a priming explanation better captures their findings. In Geis et al.'s study, It was not the behaviors of the actors that were imitated, but rather ”unspoken but implied assumptions about women's place in society (page 514) ." Subjects appeared to have been affected by a rrrore abstract, common theme from the four commercials exemplifying a traditional or nontraditional female role. The responses of subjects to stereotypic commercials were more general expressions of stereotypic behaviors than direct imitations of the behaviors of the actors they had seen. Thus, the commercials seem to have triggered consistent 13 stereotypic associations from memory (such as the general homemaker theme) which the subjects reproduced in their ten-years-later stories. The Geis et al. findings are important because they illustrate that (a) modelling studies can be subsumed by a priming model, and (b) 'stereotyped media portrayals can prime the content of broad concepts (i.e., schemas). El l' E" IS -BIS|| fl (The current research explored the utility of a priming approach to understanding the process by which stereotypic beliefs are perpetuated (if not, indeed, learned) by exposure to sex-typed mass media portrayals.)t focused on a relatively new medium—the rock music video-which is both a prevalent and popular entertainment format among preteens and teenagerqA recent content analysis of currently popular rock videos (Waite 8 Paludi, 1986) substantiated that like other forms of popular media, the portrayals of both sexes in rock videos are stereotyped and often sexist) i‘ Waite and Paludi found that male characters occurred more frecpently and were more dominant than females. nly in sexist videos were females and males equally often represented. And, in general, females were portrayed as deferent, sexually suggestiIIe, and harm-avoidanltY Rock videos are complex stimuli, and because so much information is presented in such a short period of time (usually three to four minutes), it is not surprising that stereotypic images abound (Hansen 8 Hansen, 1985)) The popularity of rock music videos is such a recent phenomenon, though, that almost no research has been done to explore the effects of this type of stimulus event. .\\ \ ‘ Hansen 8 Hansen (1986) recently conducted an experiment involving music videos that \ /’ \ / ' It. /' tested the priming explanation described eariier. Previous studies had already demonstrated that \.. brief exposure to stereotyped media portrayals could alter attitudes, behavior, and ... _~_ m-— a" w..-" _,._.r self-presentations. Hansen and Hansen wished to extend these findings by examining whether short-term exposure to stereotyped portrayals of women and men found in rock music videos ‘s— “M could alter appraisal of others. Using rock music video primes consisting of videos which were / either traditionally sex-role stereotypic or neutral, it was discovered that the type of videos subjects watched influenced the appraisal of a subsequemly observed social interaction . _ i 14 / & between a young man and a young woman) When the actors' behaviors were consistent with the primed stereotypic theme (i.e. schema-consistent), both actors were judged more favorably than they were after exposure to the neutral prime. When their behavior was inconsistent with the stereotypic prime, the actors were less well-liked after the stereotypic than after the neutral videos. When a schema had been primed by the stereotypic videos, subjects preferred actors who engaged in stereotyped behaviors. In addition, prime influenced judgments of the actors' traits. When the interaction was consistent with the primed stereotype, the actors were appraised more stereotypically than after the neutral prime (assimilation). But, when the interaction was inconsistent with the stereotypic prime, a contrast effect occurred, and the actors were judged as less traditionally stereotyped than after a neutral prime. Hansen and Hansen's findings indicate that even brief video primes can affect subsequent appraisal of a social interaction. The priming model posits that the stereotypic videos primed sex-role stereotypic schemas, making them highly accessrble and, thus, more likely to be used to encode and appraise a subsequent boy-girl interaction. The current research had two goals. The first was to replicate and extend the findings reported by Hansen and Hansen (1986). In that earlier study, the impact of priming sex-role stereotypic schemas on appraisal of schema-consistent or -inconsistent social interactions was contrasted with appraisals of the same interactions in the absence of sex-role stereotypic priming. In this study, schema-consistency was manipulated by altering the interaction so that it was either consistent or inconsistent with the stereotypic prime. In the current experiment, two stereotypic primes were used that allowed for the same interaction to be made both consistent and inconsistent. Unlike the earlier study, schema-consistency and interaction could be orthogonally manipulated to examine differences in appraisal on manipulated components within the social interactions that were consistent with one of the primes but inconsistent with the other. The expanded design permitted a more precise test of the third and fourth basic concepts of the priming model described earlier, namely that schema-consistent and schema-inconsistent social information is processed differently by t 15 perceivers. Since, in the current experiment, congmence between priming and social interaction conditions was orthogonally manipulated, the findings can be integrated into the vast literature of schematic social information processing-the second goal of this research. Memory of behavior, and of the person who performed it, has been conceptualized within a social-cognitive framework loosely termed schema theory. Impressions we forrn of people and what we remember about them may depend as much on long-term memory structures (schemas) as on the actual social information available to us (Fiske 8 Taylor, 1984). Schemas are cognitive structures which contain information (prior knowledge) about the attributes of certain concepts, including persons, events, social roles, and even the self, which aid in encoding, interpreting, and remembering social information. When a particular schema is active during impression formation, information that is relevant to the active (or 'primed') schema is advantaged. Schema-relevant information is more likely to be selectively attended to (Johnston 8 Dark, 1986), encoded (Howard 8 Rothbart, 1980; Wyer, Srull, Gordon, 8 Hartwick, 1982), and recalled (Snyder 8 Uranowitz, 1978; Taylor 8 Crocker, 1981; Zadny 8 Gerard, 1974) than schema-irrelevant information. Researchers have looked extensively at two types of schema-relevant information: information that is consistent or congruent with schema content and information that is inconsistent or incongruent with schema content. When incoming information is consistent with a primed schema (such as librarian-like behaviors to a primed librarian schema), this consistent information is processed with little relative effort (Brewer, Dull, 8 Lui, 1981). Perceivers can process large amounts of schema-consistent information in a relatively short time. Schema-inconsistent information, on the other hand, takes longer to process. Thus, when a perceivers time is limited, recall for schema-consistent information will probably be advantaged (Bargh, 1984; Hastie, 1981; Srull, 1981). Given enough time, however, it appears that schema-inconsistent information can have a recall advantage (Erber 8 Fiske, 1984; Hastie, 1981 ; 16 Hastie 8 Kumar, 1979). Some researchers have suggested that inconsistent information is better recalled because perceivers must process the information more deeply in order to resolve the inconsistent impression formed by incongruous social Information (Hastie, 1981). Resolving inconsistencies, then, might result in added attention being paid to inconsistent information at the time of encoding, producing a recall advantage. Inconsistent information, especially if there is only one or a few inconsistent pieces of information, can be given added weight in impressions (Erber 8 Fiske, 1984; Hastie 8 Kumar, 1979), but whether inconsistent information can carry more weight because it receives more conscious attention or conscious processing is still an open question. III'E" iS-BISI l “B I In the previous rock video priming experiment (Hansen 8 Hansen, 1986), subjects were exposed to a prime consisting of either neutral or sex-role stereotypic rock music videos and were then asked to watch and evaluate a male-female interaction. It was argued that sex-role stereotypic schemas logically implied a sex-role stereotypic script (Bower, Black, 8 Turner, 1979; Schank 8 Abelson, 1977) prescribing a sex-role stereotypic interaction between a man and woman. This script--the 'boy-meets-girl" script--was congruent with sex-role stereotypic schemas: boy makes advances to girl; girl reciprocates; boy is nice to girl, and (sic) they live happily ever after. The stereotypic rock music videos used in the Hansen and Hansen (1986) experiment were selected because they embodied this theme and were expected to prime the script. It was hypothesized that the stereotypic prime made sex-role stereotypic schemas highly accessible and that this schema would form the basis for judgments of a subsequent male-female interaction. The interactions that subjects appraised were constructed around this script. The schematic script implied three major components: his "coming on,” her reciprocation of his advances, and his consequent praise of her. In the earlier experiment, his ”coming on” was held constant across interactions, while the remaining two components of the script were manipulated. She either did 17 or did not reciprocate his advances, and he either praised or derogated her. Some subjects, then, appraised a schema-consistent interaction (Reciprocation-Praise), while others appraised an interaction that was schema-inconsistent (Nonreciprocation-Derogation). In the two remaining, mixed-script conditions, the behavior of one of the actors was consistent, while the other's was inconsistent: the female reciprocated but the male derogated or the female did not reciprocate but the male praised. Priming had important effects on the consequences of schema-relevant behaviors for impressions. When sex-role stereotypic schemas had been primed, schema-relevant behaviors carried more weight in evaluations of both actors. The stereotypic prime exaggerated judgments of her cuddliness, timidity, and positivity. After the neutral prime, she was judged equally cuddly, timid, and positive whether she reciprocated or not. But, after the stereotypic prime, her reciprocation produced higher ratings on all three traits, while her nonreciprocation significantly reduced them. Similarly, for him, the stereotypic prime exaggerated judgments of his sexual intimidation. After the neutral prime, he was judged equally sexually intimidating regardless of whether he derogated her or praised her, but after the stereotypic prime he was judged significamly more sexually intimidating when he derogated her than when he praised her. It appeared from these polarized judgments that stereotypic priming altered impressions of the actors so that they were perceived as more similar to the primed sex-role stereotypes when their behavior was schema-consistent but less similar when their behavior was schema-inconsistent. In addition, the stereotypic prime increased the overall favorablility of impressions of both actors when both actors engaged in schema-consistent behaviors, but decreased favorability when both actors engaged in schema-inconsistent behaviors. (See Figure 1.) Prime Condition High \\' 3. HSex-Role Stereotypic E HNeutral 3 o > a v- a c /I— '2 ¥ 3 I— 2 n. E \D 5 s 3,3 3 o f l l l l Praise Derogation Praise Derogation Reciprocation Nonreciprocation Figure 1 . The Effects of Prime, Reciprocation, and Counteraction on Overall Impression Favorability (Hansen 8 Hansen, 1986). Although a priming effect was demonstrated in the video prime experiment, the mechanism of action was largely unexplained. How can these polarization effects be accounted for theoretically? Hansen and Hansen used social information processing theory to explain findings but did not test the theory directly. An information processing approach, as suggested by Hastie (1981) and Smll (1981), among others, would separate the effects into those obtained for schema-consistent behaviors and those obtained for schema-inconsistent behaviors. Within this theory, the higher levels of schematic trait judgments after the stereotypic prime would simply represent schematic encoding of schema-consistent information. The actors' traits were appraised as more stereotypic because schema-consistent behaviors were encoded when a sex-role stereotypic schema had been primed, and, therefore, trait judgments greatly resembled 19 schematic content. On the other hand, the exaggeration of trait judgments after stereotypic priming for conditions in which the actors' behaviors were schema-inconsistent, would be expected to occur because schema-inconsistent behaviors can not be encoded easily within the accessible schema. Appraisers may need to devote more attention to schema-inconsistent information in order to continue to integrate information into a coherent impression. Thus, the appraiser would process inconsistent information more deeply, producing a greater weight to schema-inconsistent information in impressions (see also Crocker, Hannah, 8 Weber, 1983; Erber 8 Fiske, 1984; Hastie, 1980: Kulik, 1983). While the social information processing approach has much empirical support and fits the pattern of obtained data well, schema-inconsistency was confounded in this study with negativity. Schema-inconsistent behaviors in the social interaction may also have been appraised as negative behaviors. For example, after stereotypic priming, her nonreciprocation behaviors and his derogation behaviors reasonably could have been viewed as socially disharmonious, and, thus, negative. Hence, the stereotypic prime may have exaggerated the weight of behaviors which were perceived as negative or socially disharmonious (Fiske, 1980) and not merely schema-inconsistent. It has been amply demonstrated that negative social information does receive more of the perceivers attention (Dreben, Fiske, 8 Hastie, 1979; Fiske, 1980), is given greater weight in impressions (Fulero 8 Rothbart, 1980; Kanouse 8 Hanson, 1972; Wyer 8 Gordon, 1982), and results in more confident impressions (Hamilton, 1980, 1981) than positive information. In Fiske's (1980) study, for example, subjects were asked to forrn an impression of a man after viewing photos of him engaging in negative, slightly negative, slightly positive, or positive social behaviors. When the target had engaged in negative behaviors, the negative impact on his likability (using Fiske's weighting lomfula) was great-significantly greater than the impact on likability from positive behaviors. So, if schema-inconsistent behaviors in the video study were perceived as more negative due to priming of a traditionally socially harmonious 20 schema, greater weight given to negative information could account for the pattern of obtained trait and favorability ratings as plausibly as schema-inconsistency. Two general explanations have been offered by researchers for the greater weight of negative social information in impression formation. First, negative social information has been argued to contain greater motivational significance for appraisers because of a strong wish to avoid negative interpersonal consequences. Second, negative information has been posited to provide greater information value because of its infrequency or rarity in the population. Additional results from Fiske's (1980) experiment suggest that both explanations have merit. As noted earlier, two levels of positive and negative behaviors were presented to subjects. The results from analyses of likability ratings showed that the "extremity" of the positive and negative behaviors viewed was also an important determinant of weighting In likability judgments. Subjects appeared to give more weight to both negative and extreme information over positive and moderate information. In fact, when both negativity and extremity were considered together, they accounted for over ninety percent of the pooled likability variance. Fulero and Rothbart (1980) also tested the two explanations by orthogonally manipulating both the valence and papulation base rates (i.e. expected frequency) of various trait words. Regardless of the judged population freqency of the traits, undesirable traits were more likely to be used in subsequent personality judgments than were positive traits. Thus, while the results are mixed for the information value hypothesis, Kanouse (1972, Kanouse 8 Hanson, 1972) has suggested an explanation which can nicely account for both negativity and extremity effects. His hypothesis is based, surprisingly, on the positivity effect (e.g. Matlin 8 Slang, 1978; Sears, 1976). Because perceivers do not generally expect negative events, he argues, when they do occur they are striking. He seems to view the potent effects of negative social information, then, as a contrast effect to people's generally positive expectations for social behavior. It can be seen that Kanouse's explanation can describe the trait effects in the earlier video experiment if one assumes that the stereotypic prime produced an expectancy for positive behaviors, so that when 21 negative behaviors occurred, they were more striking and were given greater weight in trait judgments. Without question, the behavioral script in the schema-consistent (Reciprocation-Praise) social interaction was inherently more traditionally socially harmonious than either of the other interaction conditions. In the 'boy-meets-girl" script, behaviors of both actors were complementary, and the interaction flowed smoothly (although traditionally). In the remaining interaction conditions, one or both actors' behaviors were essentially antagonistic as well as inconsistent with the primed traditional, sex-role stereotyped schema. Thus, while prime undoubtedly affected the appraisal of the social interaction, it could reasonably be argued that the impact of disharmonious behaviors was exaggerated, rather than schema-inconsistent behaviors, per s9, because the disharmonious interaction followed a prime for social harmony. And, in general, in Hansen and Hansen's study, negative traits (his sexual intimidation, her timidity and negativity) were impacted by the prime to a greater extent than positive traits (only her cuddliness could possibly be construed as a positive trait), also suggesting that negativity can not be ignored when explaining the findings. While Kanouse's explanation does not speak directly to the impact of positive information, in general, it implies (as does the social information processing approach discussed above) that positive information, it expected, would not be salient to an appraiser. If positive information were unexpected, however, it, too, might be striking. A review of effects from priming experiments follows. S E l' | C I E . . Simply stated, priming is the activation of conceptual categories through recent use of the concept. Recent activation (i.e. priming) makes the category accessrble and more lately to be used to encode subsequent stimuli. Priming is considered a nonconsclous process (Bargh, 1984; Fazio, Sanbonmatsu, Powell, 8 Kardes, 1986; Hansen, 1985; Hastie, Park, 8 Weber, 1984; Logan, 1980). Although concepts may be activated either above (e.g., Cohen, 1981b; Ferguson 8 Wells, 1980: Martin, 1986) or below (Bargh 8 Pietromonaco, 1982) an appraiser's 22 level of conscious awareness, the person is unaware of the priming process itself when it occurs. Regardless of whether the conceptual category has been activated by an experimenters exogenous manipulation or by an appraiser in the course of forming an inpression, the result is the same; the person is more likely to use the activated category to encode subsequent behavioral information, and the appraisal seems veridical. Thus, even though a schema has been primed by an accidental course of events, the accessible category will be used to encode upcoming information, whenever possible, but the appraiser will be unaware that he or she is, in a sense, predisposed to interpret the new information in a predetermined way. In the Bargh and Pietromonaco (1982) experiment, for example, subjects were presented serially with a list of trait words that was either 0%, 20%, or 80% hostile. All words were presented below subjects' awareness such that subjects could not report having seen them. Even so, the proportion of hostile words to which they had been exposed was directly related to the hostility of a subsequent impression formed of an ambiguous person. Here, even though subjects were apparently not consciously aware of the priming manipulation, the influence of priming hostile traits was felt in their subsequent interpretation of behavioral information. Much of the social priming research has limited itself to priming individual trait concepts (e.g., Higgins, King, 8 Mavin, 1982; Higgins, Rholes, 8 Jones, 1977; Srull 8 Wyer, 1979, 1980; Wyer, Srull, 8 Gordon, 1984). The results of priming studies seem to suggest that as long as the information to be processed is schema-consistent or ambiguous, judgments about the information to be interpreted, say in forming an impression, resemble the content of primed schemas. This phenomenon may be considered an assimilatjgn effect (Hastie, Park, 8 Weber, 1984; Herr, Sherman, 8 Fazio, 1983; Sherif, Taub, 8 Hovland, 1958). Ambiguous or schema-consistent information is assimilated to the primed schema (just as in the social information approach, judgments about schema-consistent information resemble schematic content). In order for social information to be assimilated, it seems that the lnforrnation must be 23 at least moderately congment. When social information is schema-inconsistent, 99.011851 effects occur (similar to those predicted by the social information processing theory). A recent priming study by Herr, Sherman, and Fazio (1983) demonstrated these assimilation and contrast effects and suggested certain boundary conditions for them. Under the guise of a color perception experiment, Herr et al. used animal names to prime four levels of ”ferocity" (e.g., grizzly bear and shark for extremely ferocious, dove and kitten for extremely unferocious). When subjects were subsequently asked to rate the ferocity of unreal (and, therefore, ambiguous) animals, such as a "jabo" and a 'lemphor" in a supposedly unrelated task, assimilation priming effects were obtained for the primed moderately ferocious schema (or, in the authors' terms, category). Similar effects were obtained in a second experiment priming animal size. Again, the ambiguous animals were assimilated to the primed moderately extreme (size) category. Priming produced assimilation only when the ambiguous stimuli were paired with categories that were moderate. Priming very extreme categories (either ferocity or size), on the other hand, produced contrast effects. The ambiguous animals were judged smaller or less ferocious after an extremely ferocious or extremely large category prime than after more moderate primes, and they were judged more ferocious or larger after an extremely low ferocious or extremely small category prime. Only when ambiguous information is judged to plausibly fit the primed category does assimilation occur. Assimilation to a primed category, then, appears to fail when the primed category is so extreme that the target to be appraised is unlikely to fit. Herr et al.'s two experiments pointed out an additional important limitation to assimilation. When unambiguous stimuli (real animals) were presented, contrast effects occurred when noncongruent categories had been primed. Priming a noncongruent size or ferocity category resulted in more extreme judgments (I.e. contrast) than priming a congruent category. There are several theoretical arguments in the literature for the kinds of contrast effects that Herr et al. (1983) demonstrated (e.g. Bargh, 1984; Graesser 8 Nakamura, 1982; Hansen, 1985; Hastie, 1980, 1981; Posner 8 Snyder, 1975; Rothbart, Evans, 8 Fulero, 1979; SI'UII 8 Wyer, 24 1979, 1980), but there is general agreement that contrast effects result from conscious processing. If a target stimulus nan not be assimilated to a primed category, more mental effort is probably required to encode it. There is much agreement that when a conceptual category or schema is activated through priming, the “central features” of the concept will be most accessible--not necessarily all the features. Thus, consistency between the information to be appraised and the important or central features of the primed concept could provide the lower boundary for assimilation (Herr, Sherman, 8 Fazio, 1983; Higgins 8 Chaires, 1980: Higgins 8 King, 1981; Rothbart, Evans, 8 Fulero, 1979). There is a suggestion in the literature, however, that the criteria for consistency might be at least partially affective as well. For example, there is evidence that encoding is facilitated by evaluative consistency between a target and a primed category (Martin, 1986; Smll 8 Wyer, 1979). In one study (Wyer 8 Gordon, 1982), even when no effects were obtained on recall as a function of category-consistency or -inconsistency of the information, behaviors which were evaluatively inconsistent with impressions were better recalled than were evaluatively consistent behaviors, an effect which has been interpreted as a reflection of greater difficulty in processing evaluatively inconsistent information. Also, an experiment by Martin (1986) showed that evaluative consistency can actually form the basis for choosing which of several available categories to use for interpreting behavioral information about a person. Given present knowledge and theoretical formulation, it is probably safe to say that both evaluative consistency and central descriptive feature consistency are likely to be necessary for priming assimilation effects to occur. In a recent schema priming experiment, Bargh and his collegues (Bargh, Bond, Lombardi, 8 Tota, 1986) suggested that priming produces a temporary transient in the cognitive accessibility of schema-consistent information. Accessbility transients, in this view, account for differential irrrpressions formed under different priming conditions. Not dissimilar from Hastie's schema-inconsistency hypothesis, Bargh asserts that social observations conforming to accessible information (in our case, accessible schematic behaviors) are more efficiently 25 processed than nonconforming observations. In addition, he believes that schema-inconsistent behaviors can become salient under priming conditions (Bargh, 1984), drawing the perceivers attention to them because they can not be automatically processed. The accessibility construct becomes extremely useful to describe priming effects when an accessible schema is viewed as an expectancy for schema-consistent behaviors. Under priming conditions, schema-inconsistent behaviors, then, become unexpected behaviors directing the perceivers conscious attention to them (cl. Hastie, Park, 8 Weber, 1984). Not surprisingly, attention drawn to schema-inconsistent behaviors may produce liner and deeper processing, influencing appraisal. The accessibility construct can subsume Kanouse's negativity hypothesis, as well, by positing that negative or socially disharmonious behaviors are typically unexpected and, thus, normally salient (cf. Bargh 8 Pratto, 1986). If true, it should be possible to reduce the salience by making negative behaviors expected (i.e. accessible or schema-consistent). Ihefiurrentfixoerimem The experiment reported here provided a test of the schema-inconsistency hypothesis while mling out the possible social disharmony explanation of the previous research (Hansen 8 Hansen, 1986). The design of the experiment built on Hansen and Hansen's (1986) previous work to include rock music videos that primed a socially disharmonious sex-role stereotypic script--the 'boy-dumps-girl" script--in which the man "comes on" sexually, the woman refuses his advances, and the man subsequently derogates her. Using this rock video prime, the socially harmonious, reciprocation-praise interaction can be made schema-inconsistent and the socially disharmonious, nonreciprocation-derogation interaction can be rendered schema-consistent. In short, the design of the present study permitted interaction script to be fully crossed with schema-consistency. This design allowed for a test of the schema-inconsistency hypothesis unconfounded by social harmony of the interaction, since I expected both socially harmonious 26 and disharmonious behaviors to carry more weight when they were schema-inconsistent than when they were schema-consistent. If schema-inconsistency alone (rather than social dishamiony) produced priming effects on appraisal, priming would be expected to influence both types of schema-inconsistent conditions in the same manner, regardless of whether schema-inconsistent behaviors were socially harmonious or antagonistic. For example, the overly negative impressions formed of the actors in the Nonreciprocation-Derogation interaction when it followed the traditional, harmonious, ”boy-meets-girl" prime would be expected to become more positive when the disharmonious social interaction follows a congnrent "boy-dumps-girl" video prime. If, on the other hand, the effects in the earlier experiment derived solely from social disharmony, prime effects should be minimal. That is, actors in the Nonreciprocation-Derogation script condition would be judged more negatively than actors in the harmonious Reciprocation-Praise script condition regardless of Prime. However, lithe effects in the first study reflected an advantage to schema-inconsistent events, prime effects should be evident In both script conditions of the present experiment. A Script X Prime interaction should reveal that (relative to an inconsistent prime) priming a schema which is mnsjstsm with the disharmonious Nonreciprocation-Derogation interaction should attenuate negative impressions formed of both actors, while priming a schema which is jnggnsjstsm with this disharmonious interaction would be expected to exaggerate negative appraisal of the actors corrpared to a consistent prime. Likewise, priming a schema which is consistent with the harmonious Reciprocation-Praise script should attenuate positive impressions of the actors, while priming a schema inconsistent with the harmonious script should exaggerate positive impressions. Rock music videos were used to prime two types of schemas. In the first priming condition, the woman in the videos acquiesces sexually to the man's advances, he treats her well, and they go all together into the sunset (the “boy-meets-girl' schema). In the second priming condition videos 27 (the socially antagonistic prime), she does not satisfy him sexually, so he derogates her and goes off to greener pastures (the ”boy-dumps-girl" schema). Once again, neutral videos were used as a schema-irrelevant prime (or control) condition. After the priming videos, subjects observed either the Reciprocation-Praise or the Nonreciprocation-Derogation interaction sequence, as in the earlier (Hansen 8 Hansen, 1986) study. Behaviors in the Reciprocation-Praise interaction condition were carefully constructed to be schema-consistent with behaviors portrayed in the 'boy-meets-girl" prime but schema-inconsistent with behaviors portrayed in the “boy-dumps-girl' prime. Conversely, behaviors in the Nonreciprocation-Derogation interaction condition were constmcted to be consistent with the ”boy-dumps-girl" prime and inconsistent with the 'boy-meets-girl" prime. The behaviors in both interaction sequences were neither consistent nor inconsistent with the neutral prime. E I'IIEIII [EH I'III Several specific predictions were made for priming effects on impression formation. First, interactions which are schema-consistent due to priming were expected to result in schema-relevant trait judgments which resemble schematic content (assimilation). Trait judgments made about the actors, when their behavior was schema-inconsistent, were expected to show a contrast effect. In the previous study (Hansen 8 Hansen, 1986), when the male derogated the female, he was appraised as more sexually intimidating after the stereotypic (boy-meets-girl) prime than after the irrelevant prime-—a contrast effect. In the current experiment, I predicted an assimilation effect after consistent (boy-dumps-girl) prime conditions, which I expected would make the male actor seem less sexually intimidating than after an inconsistent (boy-meets-girl) prime. Similarly, in the first study, compared to effects after the irrelevant prime, the female actor's nonreciprocation produced a contrast effect on her cuddliness, timidity, and negativity after the stereotypic prime. When she did not reciprocate, she was judged as less cuddly and timid and as less positive when her behavior was schema-inconsistent than when it was schema-irrelevant. In the current experiment, when her nonreciprocation was made 28 schema-consistent by priming the boy-dumps-girl schema, her cuddliness and timidity were not expected to be as striking, and judgments on these traits should be attenuated-she should not be seen as strikingly less cuddly or less timid. Moreover, the female actors's failure to reciprocate his sexual advances should not result in her being appraised as negatively when such nonreciprocation is schema-consistent (boy-dumps-girl prime) as when it is schema-inconsistent. To the extent that the neutral prime can be considered a baseline (or control) condition, then trait judgments after this prime should fall neatly between the schema-consistent and schema—inconsistent conditions. In reality, however, this was not the expected pattern of results. It is known from Hansen and Hansen's study that 'shifts-of-meaning" occurred due to prime condition. The interaction sequences appeared to take on somewhat different meanings after the stereotypic and neutral primes. Subjects were favorably impressed when the woman did not reciprocate after the neutral prime, but they appraised her negatively for the same behaviors after exposure to the boy-meets-girl videos. This pattern of results suggests that subjects used very different schemas to encode this interaction, depending on the prime they had received. Thus, the neutral prime conditions were included in the present study primarily to look for these shifts-of meaning—rather than to serve as a baseline against which schema-consistency and schema-inconsistency could be compared. The statistical comparisons of most interest were between the two schema-relevant prime conditions. For instance, in derogation conditions, I predicted that subjects would perceive the male actor as more sexually intimidating after the schema-inconsistent (boy-meets-girl) prime than after the schema-consistent (boy-dumps-girl) prime--a contrast effect. I also expected that when the female actor does not reciprocate, subjects would perceive her as less cuddly after the boy-meets-girl prime (a contrast effect) than after the boy-dumps-girl prime. Conversely, I predicted that the male's praise would be assimilated to the boy-meets-girl schema but would produce a contrast effect against the boy-dumps—girl schema, making him appear more positive in the schema-inconsistent condition. And, since I expected that the female's reciprocation would 29 be assimilated to the boy-meets-girf schema but contrasted against the boy-dumps-girl schema, I predicted that she would appear more positive when her reciprocation was schema-inconsistent than when it was consistent. If negative behaviors are more striking to appraisers, though, it could be that schema-inconsistency would produce larger contrast effects on negative traits. E I'IIEIII ”2.. III" III II I' There is a great deal of convergent evidence that schematic processesing affects both retrieval (recall) and encoding (e.g. Alba 8 Hasher, 1983; Markus 8 Smith, 1981; Rothbart, Evans, 8 Fulero, 1979; Snyder 8 Uranowitz, 1978; Zadny 8 Gerard, 1974), although some researchers argue that the most important differences take place at the time of encoding (e.g. Baron, 1980; Weary, Swanson, Harvey, 8 Yarkin, 1980). No distinction between these two processes was made in the previous music video study. Instead, exaggerated trait judgments were hypothesized to have resulted from selective attention to schema-inconsistent behavoral information, but selective encoding of information was not tested directly. In the current experiment, a unitizing procedure in which subjects press a button at the end of each meaningful behavior they observe (Newtson, 1973) was used to assess schematic processing of information at exposure. According to Newtson (1973, 1976, 1980) and others (e.g., Neisser, 1976; Markus 8 Smith, 1981; Markus, Smith, 8 Moreland, 1985), when people process information schematically, they chunk a stream of behavior they are observing into larger perceptual units than when processing aschematically. Large units seem to be reserved for processing predictable or schema-consistent information; when the behaviors become unpredictable (Newtson, 1973) or schema-inconsistent (Markus, Smith, 8 Moreland, 1985; Sentis 8 Bumstein, 1979), perceivers chunk them into smaller units. Newtson (1980) views schemas as an expectancy for certain kinds of information. As long as a behavior is occuring parallel to the expectancy, the perceiver efficiemly processes large units. If the information is inconsistent with the expectancy, the perceiver is forced to chunk behaviors into smaller units. Smaller units may indicate that more effort or processing time is required because the incoming information does 30 not have the advantage of a behavior expectancy which is provided by an accessible schema. Several lines of evidence suggest that small units represent greater conscious, attentional processing. It is known, for example, that when people are trying to understand novel behaviors, learn a new task, or judge unfamiliar situations, they will generate smaller units, probably because unfamiliarity or aschematicity produces closer attention to small details, something that is not required of experts or schematics (Chase 8 Simon, 1973; Newtson, Rindner, 8 Campbell, 1979; Reitman, 1976; Wegner, Vallacher, Macomber, Wood, 8 Arps, 1984). Unitizing might, therefore, be related to attentional processing capacity (Bargh, 1984; Shiffrin 8 Schneider, 1980; Logan, 1980). An unpublished experiment by Newtson, Lightner, 8 Pennebaker (1980) provided some evidence consistent with this speculation. In their experiment, unitization of a predictable behavior sequence was not affected by a distractor task, but an unpredictable sequence was greatly affected, resulting in many more, smaller units. These results imply that unit size was sensitive to attentional processing capacity. A shift from large to small units suggests that the unpredictable behavior sequence is sufficiently difficult for the individual to encode that it exceeds processing capacity, forcing the reduction. Smaller units could reflect a lack of automatic processing, but at the very least, smaller units appear to indicate more effortful encoding of an ongoing behavior sequence. And, particularly relevant to the proposed research, smaller units appear to be related to lrnpression formation. In Newtson's original (1973) experiments, finer units were reliably related to (1) increased dispositional attributions and (2) more confident but more extreme (differentiated) trait judgments. Similar results were obtained in an experiment by Vlfilder (1981). He had subjects unitize the behavior of a man who was either part of a 4-person group, one of four individuals, or alone. The size of the units generated by subjects was reliably associated with impressions formed from his behavior. When the man was part of a group, subjects generated finer units of behavior and made more dispositional attributions. I expected that in the current 31 experiment smaller units generated when encoding schema-inconsistent behaviors would be related to impression formation. The unitization work by Cohen and her colleagues (Cohen, 1981a,b; Cohen 8 Ebbesen, 1979; Ebbesen, Cohen, 8 Allen, 1979) provides some insight into the potential interrelationship between unitizing and impression formation. She found that unitizing was not disruptive of in'pression formation to the extent that subjects who unitized the behaviors of a man and woman interacting (Ebbesen, et al., 1979) formed coherent impressions that seemed to be schematically guided (cf. Cantor 8 Mischel, 1979). However, when subjects' unitization was oriented toward memory for behavioral details, the schematic organization of impressions seemed weaker. Also, unitization was different when subjects were encoding the behavior for details rather than for impression formation; the units were finer (Cohen 8 Ebbesen, 1979). It is Cohen's contention (1981a,b) that impression formation unitization follows schematically-prescribed, meaningful breakpoints in a behavior sequence. When a behavior sequence can not be processed schematically-if, for example, it is unexpected (Newtson, Lightner, 8 Pennebaker, 1980)--a finer pattern of unitizing would be predicted that is similar to the finer units generated when encoding for behavioral details. In the current experiment, the ease with which an interaction could be encoded schematically (and, hence, unit size) was expected to be related to the consistency of the interaction with the primed schema. It should be noted that previous unitizing research has offered some encouragement for priming effects on unitization (cf. Cohen, 1981b). In an experiment reported by Massad, Hubbard, and Newtson (1979), subjects were shown an ambiguous "geometrical cartoon” (first used by Holder 8 Simmel, 1944) consisting of activity by three geometric shapes: a large triangle, a small triangle, and a circle. The shapes engage in a variety of ”behaviors"—chasing each other, fighting, etc. (Heider 8 Simmel, 1944). Before the film was shown, Massad et al.'s subjects were given one of two instructional sets (i.e. primes). They were told that the cartoon depicted a bully trying to steal a treasure or a rapist and two innocent passersby. Depending on their prime 32 condition, subjects unitized the behavior of the circle and two triangles very differently and formed different impressions of the ”characters.“ Unitization in the current experiment was expected to differ across the three prime conditions. Priming was expected to direct encoding of the actors' behaviors along primed schematically-relevant trait dimensions, differentially affecting unitizing. As already described, this pattern was expected to result in impressions which resemble primed schematic content when the behaviors were schemaconsistem and impressions polarized away from schematic content when appraisers were encoding schema-inconsistent behaviors. E II IEIII [EU I! I III' III I' Because unitization is thought to represent meaningful, attentional breakpoints in action (Cohen, 1981a, b; Fiske 8 Taylor, 1984; Hastie, Park 8 Weber, 1984; Newtson, 1973, 1976), researchers have investigated the link between these perceptual units and memory. Several studies have reported such a linkage (e.g., Cohen, 1981a,b; Cohen 8 Ebbesen, 1979; Newtson, 1976; Newtson 8 Engquist, 1976). In these studies, subjects better recognized behaviors that had occurred near unitized breakpoints than behaviors which did not occur near breakpoints. While better memory for information located near unitization points seems to be a consistent finding, Newtson's hypothesized encoding mechanism appears problematic. While the number of units or breakpoints can discriminate between subjects who are forming an impression of an actor and subjects who are attempting to remember each small behavior that was performed, when within-condition analyses have been performed (Cohen 8 Ebbesen, 1979), no systematic relationship between the number of units and memory for encoded behaviors was obtained. Ebbesen (1980), therefore, has argued that unitizing may be an epiphenomenon (due, for instance, to higher distinctiveness of the behaviors that are being unitized), and unitization should not be construed as necessarily representing the exact pieces of behavioral information that are being represented in memory. Researchers do seem to agree, however, that information extracted near the locations of unitization are better remembered. In the experiment 33 reported here, the behavior sequences contained several critical manipulated behaviors by both actors which make the sequences schema-consistent or inconsistent with the priming videos. From subjects' unitization during these critical behaviors, especially, I predicted that finer units would be related to better recall of highly specific details occurring in the critical parts of the interaction sequences. Specifically, I predicted that when the interaction sequences were schema-consistent, perceivers would chunk the behavior stream into larger units because encoding would be schematic and, therefore, more efficient. The critical manipulated behaviors, however, because they are easily processed in the course of forming a coherent impression, were not expected to be as precisely recalled as when they were schema-inconsistent. When these behaviors were schema-inwnsistent, l hypothesized that they would require additional conscious attentional processing, which would result in better recall for specific details of the manipulated behavior parts of the interaction sequences. Under free recall, I expected that while there would be no difference in the total number of items recalled, there would be differences in the kinds of actions that were remembered. In schema-consistent conditions, there was no reason to expect that the critical behaviors would be remembered to a greater extent than any other actions. In the schema-inconsistent conditions, however, I expected the critical (schema-inconsistent) behaviors to be overrepresented in free recall, because more attention should have been devoted to them during encoding. Also, in a measure of structured recall where precise details of the critical behaviors are required, I predicted that subjects in the schema-inconsistent conditions would be at an advantage. When schemas are used to encode behavioral information, memory for behaviors could be guided by the schemas. Not only do schemas help in recalling schema-relevant information (Cohen, 1981; Snyder 8 Uranowitz, 1978), but schematic information in memory is sometimes confusable with actually observed behaviors (Taylor 8 Crocker, 1981). People sometimes falsely report remembering information because the information is contained in schematic memory 34 (Cantor 8 Mischel, 1979; Rothbart, Evans, 8 Fulero, 1979). When perceivers process a behavior sequence schematically, errors in memory in the form of schema-consistent ”intmsions" (as in Rothbart, Evans, 8 Fulero, 1979) sometimes occur. Schematic processing biases what we recall about peeple's behavior (Cohen, 1981a) as well as the trait judgments we make when asked to form an impression (Langer 8 Abelson, 1974; Taylor, Fiske, Etcoff, 8 Ruderrnan, 1978). As discussed earlier in this dissertation, I expected that recall of behavioral information which has been made schema-inconsistent by priming would be better than recall of schema-consistent or schema-irrelevant information (Hastie, Park, 8 Weber, 1984; Srull, 1984; Wyer 8 Gordon, 1984). Thus appraisers were expected to have better recall of the critical manipulated behaviors occurring in the interaction sequence when they were inconsistent with the priming videos than when they were consistent. In the previous video study (Hansen 8 Hansen, 1986), for example, when the female actor was behaving in a schema-consistent way, she was judged more timid after the stereotypic than after the neutral prime, perhaps because female submissiveness is an important component of the stereotypic schema. Whether subjects also would have recalled more timid female behaviors after the stereotypic prime, however, can not be known because only trait ratings were collected in that experiment. In the current study, two types of recall measures were included to assess this possibility (as well as to test for the link between unitization and recall). In the first recall measure (minimally structured recall), subjects were asked to list as many behaviors as they could remember from the social interaction they observed and to indicate which actor (the male or female) had performed the behavior. This measure seemed to be a reasonable index of whether recall was better for the experimentally manipulated schema-inconsistent behaviors or schema-consistent behaviors within the interaction sequence. It also could be used to determine whether the male's or the female's behavior was more memorable. The second recall measure (stmctured recall) assesed how well subjects remembered the specific content of the experimentally manipulated behaviors in the interaction sequences, again reflecting whether 35 schema-inconsistent or schema—consistent manipulated behaviors were better remembered by subjects in the two interaction segments. In addition to these measures, the impact of schema-congruence on subjects' impressions of the actors was evaluated using trait and impression favorability scales. In summary, then, the proposed experiment tested several hypotheses about schema-inconsistent behaviors. First, I expected that schema-inconsistent behavior would attract more attention and require more effortful processing than schema-consistent behavior. Increased attention to schema-inconsistent behaviors was expected to result in (a) smaller behavioral units being chunked, (b) a recall advantage, and (c) more weight in trait judgments and impressions for schema-inconsistent than schema-consistent behaviors. Thus, when subjects were asked to observe a social interaction under priming conditions which made the observed behaviors either consistent or inconsistent with the prime, I expected that subjects' processing of the interaction behaviors would be influenced by the type of prime. Behaviors which had been made schema-inconsistent by priming were expected to require more effortful conscious processing than when the same behaviors were consistent with a primed schema. I predicted that behaviors made schema-inconsistent by priming would generate finer unitization, better recall, more extreme trait judgments, and altered impressions than would the same behaviors when made schema-consistent. Method E" El I" S I l' [II'I From previously piloted videos (Hansen 8 Hansen, 1985), seven videos were selected for use in the three prime conditions (see Appendix A). To control for the possibility that liking or disliking for certain videos might influence appraisals of the actors, care was taken to select only videos that subjects liked. The seven videos previously had been judged by approximately 100 undergraduates who found them to be equally pleasing, E (6,103) - 1.22, p < .27, M - 6.51 on an 8-point scale anchored at 0 (”negative”) and at 7 ("positive"). Originally, in the pilot study, five 36 undergraduate assistants attempted to analyze the videos for sex-role related content, but some raters failed to notice even blatantly sexist content, so a collegue and I selected the videos for the three priming conditions in the currently proposed experiment. Each prime consisted of three videos. In the 'boy-meets-girl" prime, videos were selected because the words and music portray traditional, socially harmonious sex-role stereotypic behaviors. In "boy-meets-girl" videos, women respond positively to male sexual advances and are subsequently treated well by the male. Two examples of this type were selected for use. In one, scantily-clad women take turns displaying their well-endowed forms to the delight and actuation of male onlookers. In the other, a male makes sexual advances to a pretty young woman at a party, she returns his interest, and they leave together for an implied sexual liaison. A neutral video was shown between the two 'boy-meets-girl" videos, to make the manipulation less obvious to subjects. Videos for the second ("boy-dumps-girl") prime condition were selected to prime a socially dishamlonious, but also sex-role stereotyped, schema. In the first “boy-dumps-girl" video, a man courts and marries a sexy, young woman. After they are married, she changes into a fat, lazy, 'hausfrau" who spends all her time eating bonbons in front of the television. He becomes disgusted and jumps out the window to chase after a new, sexy, young woman. In the second video, the male gets tired of his girl and finds another. While proclaiming he does not want her any more, he verbally abuses the old girlfriend, and in the last scene he is seen approaching his (prettier) new girl. The same neutral video separating the two 'boy-meets-girl' videos was also inserted between the two "boy-durrps-girl videos. Neutral prime videos were selected because they did not portray any of the behaviors described above. While it would have been interesting to include a pn‘ming condition in which a nontraditional schema would be primed, the choice of neutral videos was forced by the lack of rock music videos that portray women and men in nontraditional roles. The few videos that portray male-female equality or female dominance typically contain scenes of violence and could 37 not be used. Thus, neutral videos were negative instances of sex-role stereotypic priming, rather than positive instances of a nontraditional stereotype. One neutral video describes the life of a lead singer ”on the road”; another shows two performers who play and sing amid fantasy musical images; the third tells the story of people from all over a large city coming together to sing with a popular rock band in a recording session. W The videotaped interaction sequences showed a young man and a young woman supposedly auditioning for video jockey (VJ) positions on a local cable channel. The interaction was comprised of two separate segments: a "getting acquainted segment” and an "audition” segment. Two undergraduate confederates acted as the job candidates, and their scripted performances were videotaped. These tapes had been constructed for the previous rock music video study (Hansen 8 Hansen, 1986), so they were known to produce substantial effects on traits and impressions. Twelve undergraduate students helped to construct the scripts that the "applicants” used. For the ”getting acquainted“ segment, the students first generated a list of statements indicating sexual interest that a male might make to a female. Then they generated two sets of responses that the female might make, one set reciprocating sexual interest and the other ignoring or deflecting it. Next, they collaborated on generating filler conversation, following instmctions that the conversation was not to suggest either actor's knowledge of rock music videos, their feelings about one another, or their sex-role identity. For the ”audition" segment, these same undergraduates viewed a popular (neutral) rock music video ad lib (”Born in the U.S.A. by Bruce Springsteen) and constmcted filler conversation that was oriented toward it. During the ”audition“ segment, the actors are seen commenting on this video. The students also generated a list of comments one VJ candidate might make about the video and two sets of oounterresponses the other candidate could make to each of the comments-- one set praising and one set derogating the first person's comments. The lists of statements, responses, and filler conversation were combined into the experimental segments. In the “getting acquainted" 38 segment, two scripts were constmcted: one in which the female reciprocated the male's sexual advances and one in which she did not. In the two ”audition” segment scripts, the male either praised the female's comments about the videos or he derogated them. The two scripts in each segment were parallel, except for the female's reciprocation vs. nonreciprocation behaviors in the "getting acquainted” scene and the male's derogation vs. praise in the ”audition" scene. The confederates were given the final version of the scripts and were coached regarding their speech styles and nonverbal behaviors. After extensive practice in both segments, several takes of each were color videotaped. The best performances of the two "getting acquainted” and two "audition” segments were edited together to form the Reciprocation-Praise and Nonreciprocation-Derogation interactions. WW. During this ”getting acquainted" segment, only the female's behavior was manipulated; the male's was the same in both reciprocation and nonreciprocation conditions (see Appendix B). At the beginning of the segment, the seated male actor rises to greet the arriving female. They introduce themselves and sit down next to each other and begin getting acquainted. (This segment was supposedly being recorded without the actors' awareness.) While the actors carried on the scripted conversation, they watched the Springsteen video on a color monitor. Except for the reciprocation vs. nonreciprocation statements made by the female, the actors' conversation was identical in both conditions. During the majority of the conversation, they talked about neutral things (such as where they lived, travel, other people, etc.) and made general comments about the video (such as the name of the song, the performer, where he had appeared in concert, etc.). At various points in the segment, the female actor either reciprocated or deflected sexual advances which were made by the male actor. For example, at one point the male touched the female's blouse and said that it felt ”soft and sexy.” Her response was either a friendly, “Yeah, but it gets wrinkled when you touch it, sometimes.” in the nonreciprocation condition (accompanied by a shift away from him in posture) or a sexy, 'It feels nice to wear it,“ (with no 39 movement away) in the reciprocation condition. At another point, after the female makes a comment about the cold room, and the male replies “If I knew where your, ah, the thermostat was, l'd turn it up for you," she either responds with a coy, "I'd like that,” (reciprocation) or an indifferent, "That's OK, I'll be fine," (nonreciprocation). When, at the end of the segment, the male says that he is looking forward to going out with her if they get the jobs, she either reciprocates with, "I'd like that; I already feel that I know you better than most of the guys I've dated,” or deflects his advances with, "Well, I suppose so, but I don‘t know you that well yet.“ Both segments end with his, "Oh, you'll get to know me better!" W. In the "audition“ segment, in which the actors were supposedly formally auditioning for the VJ jobs in front of a camera (see Appendix B), a digital timing display was inserted across the bottom of the tape to reinforce the audition cover story and to distinguish it from the “getting acquainted“ segment. In this segment, the actors comment on a video that supposedly had just been shown (”Born in the U.S.A.”). In this scene, the female's behavior was the same in both scripts, but the male's was manipulated so that he either derogated or praised her comments about the video. In response to the female's comment about a quality of the video, for example, the male says either, "Well, that's not really true, but I suppose you had to look at it more closely than you did in order to evaluate that,” (derogation) or “Well, that's certainly true, but I suppose you had to look at it as carefully as you did in order to evaluate that” (praise). After another of her comments, his response was either, ”You are certainly right to point that out to our audience, the point is...," (praise) or, ”I'm sure our audience is aware of that, the point is..." (derogation). At another point, when she mentions that “It's what you hear that counts,” he counters with derogation, ”Sure, and people go to concerts and keep their eyes closed,” or praise, ”Boy, I love the way you go to the heart of it.” A three-part preliminary experiment was conducted to test whether the experimental design adequately captured the prime-script relationships; that is, to determine whether the 40 Reciprocation-Praise script was consistent with the boy-meets-girl video prime, inconsistent with the boy-durnps-girl video prime, and irrelevant to the neutral video prime, and whether the Nonreciprocation-Derogation script was inconsistent with the boy-meets-girl video prime, consistent with the boy-dumps-girl video prime, and irrelevant to the neutral video prime. In the first part, a number of subjects watched one of the two scripted conversations or the videos constituting one of the video primes and were asked to extract the ”theme" of the material in the form of a short, written story. In the second part, each of three judges further distilled all of the stories written about each presentation to a common theme in the form of a short story. In the last part, another group of subjects evaluated the relationship-on a scale ranging from consistent through irrelevant to inconsistent-of each judge's script stories with the same judge's video prime stories in a series of 18 pair-wise comparisons-six for each of the three judges. Forty-four subjects participated in the first part of the preliminary experiment. Approximately equal numbers of women and men participated in each of five experimental sessions. Each group was assigned to view either one of the two interactions or videos constituting one of the three rock video prime manipulations. Before showing the videotape, the experimenter explained the task and introduced the concept of a theme. The idea of a theme was illustrated with the story of "T he Tortoise and the Hare," selected because it had little or no sex-role connotation and because the theme of the story was representative of some contemporary motion picture themes. The theme was described in terms of a contest between an ordinary hard-working character and a flashy know-it-all braggart. "Plain and ordinary“ just keeps plodding along, suffering the derision of the braggart. who laughs. "Plain and ordinary“ doesn't give up, and, in the end, beats the cocky braggart. The experimenter explained that themes such as this are simple plots or scripts that can often be found in fairy tales like "The Tortoise and the Hare" and in books, movies, and real life: “the characters and the setting may change, the story may be much more elaborate in details, but the theme remains the sameuthe theme is the basic plot or structure around which the story is built.” The experimenter asked subjects to think of an 41 example of a contemporary movie based on the theme of “The Tortoise and the Hare.” (Subjects commonly offered the theme from the movie “Rocky” as an example.) After this exercise, the experimenter explained that they would be seeing a videotape: either a conversation between two people or rock music videos, depending on condition assignment. They were instructed to write out the theme of the video material in the form of a short story as soon as the presentation was completed. The experimenter then showed the videotape and, when subjects had finished writing, collected their stories. The stories, as already noted, were the materials used in the second part of the experiment. AW. Rather than dismissing the subjects at this point, the experiment actually was continued. Subjects went on to complete the same procedure for each of the four remaining videotape presentations, so all subjects saw and wrote stories for all three video primes and both scripted conversations. Each group viewed the interactions and video primes in a different order as determined by a latin square. After the procedure for all five presentations had been completed, the experimenter distributed materials for an additional task. Half the subjects were randomly assigned to consider the Reciprocation-Praise conversation while the remaining half were asked to consider the Nonreciprocation-Derogation conversation. They were asked to indicate which of the three rock music video presentations they had seen was most like the target conversation. They were told that they could use more than one rock video presentation if they wished, although none did. After this task had been completed, the experimenter explained the investigation to the subjects, offered them the opportunity to withdraw their stories (one did), and then thanked and dismissed them. memes. Because of concerns about reactivity, only the first story written by each subject in part one of the experiment was systematically examined. The stories generated in part one of the experiment were edited for spelling and punctuation errors. If characters were mentioned by name, sex-appropriate nouns (boy or girl) and pronouns (he, she, etc.) were substituted. All of the stories for one condition (Boy-Meets-Girl, Boy-Durrps-Girf, and Neutral Prime and the 42 Reciprocation-Praise and Nonreciprocation-Derogation scripts) were typed on the same page in random order. The five pages of stories were given to each of three undergraduate judges who were naive to the research. The order in which judges received the five pages, each representing a different condition, was randomized across judges. The introduction to the concept of themes and the instmctions for the task--reducing the stories on each page to a common theme—were the same as those used in the first part of the study (except for modifications reflecting the paper-and-pencil raw materials). The judges worked independently. Each of the three judge's five stories were corrected for spelling and punctuation and edited into the stimulus materials used in the third part of the experiment (see Appendix C). Twenty-eight subjects participated in the third part of the experiment which was designed as a 2 (Script) X 3 (Prime) X 3 (Judge) within-subjects factorial. Each subject made 18 pair-wise comparisons. For each judge, subjects were asked to evaluate the relationship of each script-story with each prime-story. For each script-prime comparison, subjects were asked to indicate on a 5-point bipolar scale the extent to which the script-story was ”very compatible" (+2), “unrelated” (0), or ”very incompatible” (-2) with the prime-story. The order in which any given subject made the comparisons was randomly determined. W W. The initial recall measure asked subjects to Ist as many behaviors as they could remember from the two interaction segments (see Appendix D). On the first page, the instructions requested that subjects list the behaviors in the order in which they came to mind. Each line contained a space to write in the person who performed the behavior and the behavior recalled. Seventeen blank lines were provided on the first page of the measure. At the bottom of this page was the direction, “Use the next page if needed.” There were an additional twenty-two blank lines on the next page. This recall measure was always placed first in the questionnaire booklet (see Taylor 8 Fiske, 1981, for a discussion of placement of recall measures). 43 13mm. One (as-item) page of scales was used to rate the female actor's traits and one page for the male. Except for instructions at the top of the page telling subjects to rate the male or the female, the pages were identical (see Appendix E). All scales were bipolar adjective scales, presented in 9-point semantic differential format. Subjects were instmcted to place a check at the point which represented their impression of the actor on that trait. Order of scales and position of anchors was random, except for the first three scales (unappealing-appealing, weak-strong, and passive-active) which were designed to tap three primary trait dimensions (Osgood, Suci, 8 Tannenbaum, 1957; cf. Rosenberg 8 Sedlak, 1972). Order of trait measures was counterbalanced, with half the subjects rating the female first, and half rating the male first. Trait scales followed the first recall measure in the questionnaire booklet. W. Three scaled items assessed different aspects of the overall impressions created by the actors (see Appendix F). On 11-point scales anchored at 0 (”definitely no") and 10 ("definitely yes”), subjects were asked to make a “general evaluation” of both actors on (1) whether they should be hired based on how well they came across in the “getting acquainted” segment, (2) whether they should be hired based on how well they came across in the ”audition" segment, and (3) whether the candidates seemed like someone the subject would Iflte personally, based on everything subjects observed. All subjects received the three questions in the same order, but for each question, half the subjects evaluated the male first and then the female on each question; the remaining subjects evaluated the female first. W. The last measure assessed recall of the manipulated lama] statements in both interaction segments (see Appendix G). Subjects were given the (nonmanipulated) statement which preceded the manipulation and were asked to recall what the other candidate's response was. For example, in the "getting acquainted” segment, subjects were given his (nonmanipulated) statement that her blouse was “really soft and sexy." Subjects were asked to write in what her response was. There were four manipulated verbal statements requested from the "getting acquainted” segment and six from the "audition" segment. The first 44 three manipulated behaviors in the ”getting acquainted“ segment consisted of body language only, so recall of them was not measured. BMW The cover story and all instructions were presented via videotape (see Appendix H). The same cover story and basic instructions were used as in the previous music video experiment (Hansen 8 Hansen, 1986). Subjects were told that they would be watching two people who had applied for jobs as video jockeys on a local, cable version of MTV (Music Television) which the investigator was producing for a local cable channel. Several important points about the actors were mentioned: (1) the male and female had not met before, (2) they had been paired up randomly from their employment applications, (3) they were not being shown because of any experimenter preference, but instead (4) they were only two of a number of applicants being shown at random to various aggregates of subjects. Subjects were told that the candidates were amateurs, and that the successful candidates would receive training as VJs. It was emphasized that the experimenter was only interested in evaluations of the "personalities“ of the two candidates--"how they came across...” Next, the instructions explained what subjects would be watching on the videotape. They were told that they would be shown some rock music videos first, "to set the stage...and to acquaint people who may be unfamiliar with rock music videos with the format." Next, they were told that they would see the VJ candidates at two different times. First, they would see the candidates when they were getting acquainted and previewing the video they would be “commenting on during their audition.” During this first segment, the candidates were said to have been unaware that they were being observed or videotaped. And, second, they were told that they would see them during their formal audition for the jobs, in which “they make comments on a preceding video and bridge into an introduction of the next one.” Subjects were led to believe that all conversations, behaviors, and comments were spontaneous. 45 After these introductory comments, the procedure was summarized, and subjects were reminded that they were to “form an impression“ of the candidates. They were told to watch the videotape and that further instmctions would be given at the end of the audition tape. After the audition tape, subjects were instructed to answer the questions in the questionnaire booklet, following the written instmctions on each page. When subjects finished the booklet, the experimenter reentered the room and partially debriefed them. Full debriefing was accomplished by a follow-up letter, which included a detailed explanation of the hypotheses and procedures and preliminary results. ! I 'I' . Instructions similar to those used by Newtson 8 Engquist (1976) and Markus, Smith, 8 Moreland (1985) described the unitizing task: "As you are watching and listening to the the candidates, what we would like you to do is to press the button in front of you whenever you see what to you is a meaningful unit of action. For example, if you see one of the candidates turn their head, look at someone, raise their arm, wave, bring their arm back down, and turn their head back, you can either say that you saw six meaningful actions (head tum, eye movement, raised arm, waving, moving arm, head turn) or one meaningful action (waving at someone). Or suppose the candidates are talking. You might consider several sentences in a row to be a meaningful action, or each sentence, or even each separate thought. Any number of meaningful actions you decide on is the answer we want. There are no right or wrong answers. Just press the button whenever you feel that you have seen a meaningful unit of action.” At the end of the videotaped, verbal instructions, after the entire procedure was summarized, subjects were briefly reinstructed on the unitizing task: “Remember that you will be pushing the button in front of you every time you see the candidates perform a meaningful unit of behavior. I would like you to get the feel of it, so imagine that you just observed the candidates performing a meaningful behavior. Push the button finnly--and release. The light should go on while the button is pressed and go off when you take your finger off the button. Do not begin pushing the button to measure what you feel are meaningful units of action until the tape begins in which the two candidates are getting acquainted.“ Wotan. A response key and an indicator light were mounted on the surface of the table in front of the subject. A green indicator light went on whenever the 46 key was depressed to give subjects a positive signal that their key presses were being recorded. The response key closed a normally open circuit gating a constantly available 1 kHz tone to one channel of a two channel audio recorder. The sound track of the video presentation as it was played to the subject was recorded on the other channel of the audio recorder. In this way, unitizing was synchronized with the auditory content of the stimulus material being unitized. Bilatflud! A pilot study was conducted to assess if the videos and interaction segments produced the anticipated effects on impressions and judgments. In the pilot study, subjects were not given the unitizing task, but simply viewed the primes and interaction sequences (as in the previous video experiment) and completed the trait scales and overall impression measures. The pilot study was designed as a 3 (Prime: Schema-consistent, Schema-irrelevant, Schema-inconsistent) X 2 (Interaction Script: Reciprocation-Praise, Nonreciprocation- Derogation) between-subjects factorial. One hundred and twenty-two undergraduates participated for partial course credit. An attempt was made to recruit one male and one female for each session, but subjects participated alone in adjacent experimental rooms and did not interact at all. Two male and two female experimenters conducted the fifty-minute sessions. The procedure for the pilot study was the same as for the experiment described below. E . | | D . | E l The full experiment was a 3 (Prime: Schema-consistent, Schema-irrelevant, Schema-inconsistent) X 2 (Interaction Script: Reciprocation-Praise, Nonreciprocation-Derogation) between-subjects factorial. After escorting subjects to the experimental room, the experimenter briefly explained the procedures and obtained informed consent (see Appendix I). Next, the experimenter (who remained blind to experimental condition while interacting with subjects) went to the control room, randomly assigned the subject to condition, and selected, loaded, and started the appropriate session videotape. After receiving their instmctions and the cover story by videotape, subjects viewed a rock music video prime and 47 a two-part interaction sequence (a ”getting acquainted” segment and an ”audition” segment). While subjects watched the interaction, they recorded what to them were meaningful behavior units by pressing a button on the table in front of them. After the interaction sequence, subjects were instructed over the videotape to begin filling out the questionnaire booklet which was in a sealed, manila envelope on the table. All subjects completed the set of dependent measures described previously: an unstmctured cued recall task, two semantic-differential trait scales, three scaled questions assessing general impressions, and a stmctured recall measure. The trait scales and impression questions were counterbalanced for sex of actor, but the four measures were always given in the same order. When subjects finished the questionnaire booklet, they were partially debriefed by the experimenter, thanked, asked not to discuss the experiment, and given an information sheet to take with them (see Appendix J). One hundred seventy-eight undergraduate subjects participated in the experiment for course credit. They were run in two adjacent experimental rooms. The experiment lasted slightly under an hour and was conducted by several male and female experimenters. Judoesandfiodino Two judges, undergraduates who were naive to the experimental hypotheses, coded both recall measures. Judges remained blind to the Prime condition of subjects, and during coding of the minimally structured measure, judges were also blind to Interaction condition. Coding the stmctured recall measure required that judges make inferences about semantic accuracy and strength of response, necessitating awareness of the subjects Script condition in order to oon'pare responses to the actual script statements. Wan. Recalled behaviors were coded into one of twelve categories. There were six major behavioral categories: his coming on to her, her reciprocation, her nonreciprocation, his praise, his derogation, and all others (residual behaviors), and each of these six categories was coded for whether or not they actually were present in either of the two interaction scripts. Last, the number of male and female behaviors was totalled. Scoring was 43 done on coding sheets. The full scoring procedure and coding sheets may be found in Appendix K. W. Recalled statements were judged along two dimensions: accuracy and strength. To score accuracy, judges were trained in a coding system designed to weight accuracy of meaning as well as verbatim recall. Using guidelines provided by the researcher, the judges were trained in the procedure; practice examples were discussed and scored at several practice sessions before judges began scoring actual data. Periodically, random checks were performed to establish that judges were maintaining coding guidelines. (Refer to Appendix L for the scoring protocol and coding sheet.) After accuracy was coded, judges assessed the strength of each recalled statement according to whether it approximated the strength of the actual script statement or indicated a stronger or weaker reply than the one in the script. After one training session, judges were able to score statements along this dimension, but their actual coding was checked periodically. I 1' ID I' fill" B Unitizing responses were manually transcribed from each subject's audiotape. By playing both channels of a subject's tape simultaneously (recall that one channel contained the unitizing responses and the other the contemporaneous interaction conversation), the unitized conversation points were located and marked on a coding sheet containing the entire scripted conversation. Next, units within twenty-three conversation segments were counted. Units were tallied within nine (verbal and nonverbal) ”he came on to her" segments, eight (verbal and nonverbal) ”reciprocation or nonreciprocation” segments, and six (verbal) ”praise or derogation” segments. Total units during the "getting acquainted" and “audition“ sequences were also counted and recorded. The coded segments are shown in Appendix M. 49 Results El" E . I'S'I-E' E 'l Injnauest, Recall that subjects in the first part of this preliminary experiment were asked to match up the priming videos with the two social interactions. The results were analyzed in a 2 (Interaction) X 3 (Video Presentation) contingency table (see Table 1). That judgments made about the match between the videos and interactions were not independent can be seen by the significant, X 2. Looking at the first row of the contingency table, it is obvious that the boy-meets-girl videos were judged to be most like the Reciprocation-Praise interaction script at a level far greater than chance (the expected frequency is only 7.3 for all cells). Neither of the other two types of videos approached this expected frequency for being judged ”most like" this interaction. Similarly, looking at row two of the table, the boy-dumps-girl videos were judged ”most like" the Nonreciprocation-Derogation interaction at a greater than chance level, while matches between this interaction and the other types of videos fell far below expected frequencies. Although not a particularly stringent test of the schema-script relationships, the data did indicate that the video primes were judged to be consistent with the appropriate interaction scripts. Imam. A3 (Prime: Boy-Meets-Girl, Boy-Dun‘ps-Girl, Neutral) X 2 (Script: Reciprocation-Praise, Nonreciprocation-Derogation) X 3 (Judge) repeated measures ANOVA was conducted on subjects' (pairwise) ratings of the consistency, inconsistency, or irrelevancy of the script themes to the prime themes. The ANOVA produced the anticipated Prime X Script interaction, E (2,26) = 63.65, p < .001. It was not qualified by interaction with Judge, E (4,24) = 2.03, p < .13. The means underlying the significant Prime X Script interaction are shown in Table 2. A simple effects analysis (Wrner, 1971) indicated that the Script effect was significant in Boy-Meets—Girl, E (1,27) - 113.63, p < .001, and in Boy-Dunps-Girl prime conditions, E (1,27) = 55.59.51 < .001, but not in Neutral prime conditions, E (1,27) < 1, p < .92. The Boy-Meets-Girl video prime was more compatible with the Reciprocation-Praise than the 50 Table 1 Consistency of Scripted Conversations with Rock Music Video Primes Observed Frequency of “most lite” Responses Video Interaction Boy—Meefs—Girl Boy-Dums-Girf Neutr'ci Reciprocation-Praise 18 1 2 Nonreciprocation-Derogation 2 18 2 x2 (2) =soss, p< .001 Table 2 Summary of Conscious Appraisals of the Relationships of Prime-Stories to Script-Stories Video Interaction Boy—MeetsGirl Boy-nmps-Girl Neutrai Reciprocation-Praise 1 .06a -.63c “25b Nonreciprocation-Derogation -.87c .62a "24b Note. Evaluations of the relationship of each script-story with each prime-story were made on a 5-point scale anchored at -2 (“very incompatible"), at 0 ("irrelevant”), and at +2 (”very compatible"). Row means not sharing a common subscript are significantly (p < .05) different. 51 Nonreciprocation-Derogation script. Conversely, the Boy-Dumps-Girl video prime was more compatible with the Nonreciprocation-Derogation than the Reciprocation-Praise script. The Neutral prime bore nearly the same relationship to both scripts. In passing, it should be noted that the contrasts with Script induced by the Boy-Meets-Girl prime (i.e. consistency with Reciprocation-Praise and inconsistency with Nonreciprocation-Derogation) appeared somewhat more powerful than the reversed contrasts induced by the Boy-Dumps-Girl prime (i.e. inconsistency with Reciprocation-Praise and consistency with Nonreciprocation-Derogation). The asymmetry was sufficient to produce a Script main effect, E (1,27) = 9.16, p < .005. It remained to be seen whether this would result in stronger experimental effects in the Boy-Meets-Girl prime conditions. But, this one comparison aside, the data fit the anticipated pattern very well. Sirmle effects analyses also indicated that the Prime effect was significant in both Reciprocation-Praise, E (2,26) - 124.75, p < .001, and Nonreciprocation-Derogation script conditions, E (2,26) . 60.01, p < .001. Specific comparisons indicated that these effects matched expectations. The Reciprocation-Praise script was judged more consistent with the Boy-Meets-Girl prime than with the Neutral prime, E (1,27) = 121.36, p < .001. It also was judged more inconsistent with the Boy-Dumps-Girl prime than with the Neutral prime, E (1 ,27) - 6.12, p < .02. The Nonreciprocation-Derogation script was judged more consistent with the Boy-Dun'ps-Gin prime than with the Neutral prime, E (1,27) - 67.08, p < .001. It also was judged more inconsistent with the Boy-Meets-Girl prime than with the Neutral prime, E (1,27) - 11.74, p < .002. These results strongly supported the contention that the 2 X 3 experimental design adequately captured the conceptual prime-script relationships; the Reciprocation-Praise script was consistent with the Boy-Meets-Girl video prime, inconsistent with the Boy-Dumps-Girl video prime, and irrelevant to the Neutral video prime, while the Nonreciprocation-Derogation script was inconsistent with the Boy-Meets-Girl video prime, consistent with the Boy-Dumps-Girl video prime, and irrelevant to the Neutral video prime. 52 ll 'I'i E'IISI I As described in the methods section of the dissertation, a pilot study was conducted, without unitizing, to examine if the priming videos would produce effects on trait and favorability judgments prior to the introduction of the additional unitizing task into the design. Indeed they did. The only difference between the results for the pilot study and the unitizing experiment was that both trait and favorability effects were stronger when subjects did not have the added task of unitizing the interaction while they were watching it. Aside from that one difference in procedure, the pattern of effects was quite similar in both experiments, and, as hoped, the priming videos produced striking effects on impressions of the actors. The complete results are reported elsewhere (Hansen 8 Hansen, 1987), but the general findings are described briefly here. Subjects rated the actors more favorably after seeing the (harmonious) Reciprocation-Praise interaction than the (disharmonious) Nonreciprocation-Derogation interaction. Wrthin each type of interaction, however, both actors were better liked when the social interaction was consistent with the priming videos (i.e. schema-consistent) than when it was inconsistent (i.e. schema-inconsistent). An identical pattern was found for trait judgments. Both actors were appraised as having more favorable traits after the Reciprocation-Praise interaction than after the Nonreciprocation-Derogation interaction, but these ratings were also affected by the type of priming videos to which subjects had been exposed. The negativity of impressions after the Nonreciprocation-Derogation interaction was significamly attenuated when the behavior was schema-consistent (when subjects had seen the boy-dumps-girl videos) as predicted. The positivity of impressions after the Reciprocation-Praise interaction, though, was exaggerated in a positive direction when it was schema-consistent (when subjects had seen the boy—meets-girl priming videos) contrary to predictions from the information processing hypothesis. The neutral videos produced a somewhat different pattern of effects on the two actors. The male was less well-liked and was given more negative trait ratings after neutral videos than after either stereotypic prime. For the female, on the other hand, neutral videos caused her to be 53 appraised less favorably only after the Reciprocation-Praise interaction. After the Nonreciprocation-Derogation interaction (in which she deflected his sexual advances), neutral videos produced a more favorable appraisal. Results from the pilot investigation--while there were departures from effects which would be predicted by the social information processing hypothesis as it stands--were certainly encouraging enough to conduct the full study. As shown in the balance of the results, the pattern of effects found in this preliminary experiment was replicated and extended by findings for unitizing, but the social information processing hypothesis was not totally supported. ll'l" E . l'I'lll | 1mm. Responses on the two 33-item trait scales were transformed to 9-point scales, anchored at 9 with positive instances of a concept and at 1 with negative instances. On the dominant-submissive scale, for example, dominant would be considered a positive instance. Because the masculine-feminine dimension is somewhat different from the other scales, masculine was arbitrarily used to anchor the scale at 9. W. Transformed trait ratings of the male actor and the female actor were submitted to factor analysis. Oblique rotation was selected (T abachnick 8 Fiddell, 1983) to permit nonorthogonal factors to emerge. Eight factors were extracted for her and six factors for him. The degree of correlation between the individual factors for each actor is presented in Table 3. Factor loadings are shown in Tables 4 and 5. For the female, ten traits showed no appreciable loadings on any factor. These traits (colorful, prosocial, sensitive, sympathetic, positive, extraverted, happy, pleasant, sophisticated, and intelligent) were used as variates, along with the eight obtained factors. Forthe male, nine traits (colorful, clumsy, friendly, intelligent, positive, happy, masculine, active, and competent) were treated as variates, along with the six obtained factors. To maintain equivalent scaling between factors and individual trait variates, summed ratings for all factors were divided by the nurrber of constituent scales, and these 54 average ratings were the data submitted to analysis. The factors were tested for coherence across experimental conditions before they were analyzed further. As a test of stability across conditions, or coherence, for each actor, the constituent measures from each obtained factor were regressed on factor scores from that factor, and the resulting regression lines were compared (tested for equality) across Prime X Script conditions. Among the eight female factors, all but two (ability and mastery) met the equality of slopes criterion (all Es approximately equal to 1, ns.). Inspection of the ability constituents suggested that the coherence of the two components (skill and competence) was differentially affected in the Irrelevant Prime condition, E (10,157) = 2.36, p < .02. For mastery, the slope for the Nonreciprocation-Derogation condition appeared to be decreased significantly more by the Consistent prime than by the other two primes, E (10, 157) :- 2.36, p < .01. For the male factors, all but one (social power) evidenced good coherence across conditions (Es approximately equal to 1, ns.). The slopes for the extraversion constituent of social power appeared more nearly parallel across conditions than did the other components, E (20, 147) . 2.05, p < .01. Significant effects produced from these three factors, then, will need to be interpreted conservatively, but with these exceptions, the factors appeared stable and all were used in subsequent analyses. W. A 3 (Prime: Schema-consistent, Schema-irrelevant, Schema-inconsistent) X 2 (Script: Reciprocation-Praise, Nonreciprocation-Derogation) MANOVA was conducted on the female trait variates. (In this analysis, and all subsequent analyses, significant univariate effects are reported only when the multivariate E was significant.) As anticipated, very different impressions of the female actor were formed in the two social interaction conditions. Script produced significant effects on several variates, indicating that impressions of the female were much more favorable in the hamronious Reciprocation-Praise condition (see Table 6). When she reciprocated, she was judged higher in charm, E (1,169) - 17.10, p < .001, more sexy, E (1,169) . 26.85. p < .001, more sophisticated, E (1,169) . 7.00, p < .01, more colorful, E (1,169) - 23.71, p < .001, more extraverted, E (1,169) . 21.59, p < .001, Intercorrelations Among Factors for Male and Female Impressions 55 Table 3 Factor Intercorrelations of Female Factors Social Mean ad ”0" Charm Mastery Sexy Openness AblTrty Assertion Nasty Strung Social Assertion 1.00 Mean and Nasty .16 1.00 Charm .19 .10 1.00 Mastery .33 .05 .26 1 .00 Sexy .29 .10 .11 .22 1.00 Openness .09 .24 .22 .23 .15 1 .00 Ability .08 .08 .18 .15 .16 .17 1.00 High Strung .05 .10 .16 .08 .01 .13 .07 1.00 Factor Intercorrelations of Male Factors Charisma Nice Guy Social Power Social Energy Truthful Aloof Charisma 1.00 Nice Guy .29 1.00 Social Power .21 .06 1.00 Social Energy .29 .09 .28 1.00 Tmthful .23 .30 .06 .20 1.00 Aloof .10 .05 .07 .03 .03 1 .00 56 Table 4 Pattern of Factor Loadings for Individual Female Trait Judgments Factors Traits Social Meanand Assertion Nasty Charm Mastery Sexy Openness Ability outgoing assertive talkative active strong clever dominant masculine threatening nice aggressive attractive appealing clumsy independent sexual warm honest friendly corrpetent skilled emotional excited .682 .678 .649 .579 .577 .506 .808 .691 -.564 .504 .767 .733 -.763 .706 .767 .552 .733 .524 .758 .691 .808 .579 VP 3.57 2.64 2.22 2.17 2.11 2.04 1.73 1.68 57 Table 5 Pattern of Factor Loadings for Individual Male Trait Judgments Fm Charisma Nice Guy Social Power Social Energy anthful Aloof attractive appealing clever strong skill sympathetic sensitive threatening nice pleasant warm assertive dominant extraverted aggressive excited prosocial outgoing talkative emotional honest independent sexual sophisticated .884 .706 .647 .585 .515 .826 .766 -.652 .629 .556 .521 .778 .637 .615 .520 .715 .657 .635 .547 .713 .606 .559 -.413 .426 VP 3.91 3.75 2.88 2.64 2.12 1.22 58 and more socially assertive, E (1 .169) - 30.65.51 < .001. Also, she was perceived to be happier. E (1 .169) - 61.74. p < .001. more sensitive, E (1.169) a 37.80. p < .001. more sympathetic. E (1.169) . 17.52. p < .001. more prosocial. E (1 .169) . 20.95. p < .001. and more pleasant, E (1 .169) = 12.64. p < .001. Finally. she was judged as less high-strung. E (1 .169) - 15.48. p < .001, more positive. E (1 .169) = 17.34.12 < .001. and higher in ability. E (1 .169) a 15.27. p < .001 . and mastery. E (1.169) = 17.75. p < .001. The two Script conditions. then. produced the anticipated differences in the female's general positivity. social efficacy. and certainly nurturance. The idea that the Reciprocation-Praise Script condition was perceived to be more socially harmonious than the Nonreciprocation-Derogation condition was strongly supported by subjects' evaluations of the female's traits. But of even greater importance for purposes of the experimental predictions. Prime also produced effects. Recall that it had been predicted that relative to an inconsistent prime. a consistent prime would attenuate negative impressions in the Nonreciprocation-Derogation conditions and attenuate positive impressions in the Reciprocation-Praise conditions. This hypothesis was partially supported. In the Nonreciprocation-Derogation condition. the schema—consistent video prime made impressions of the female less negative than they were in the schema-inconsistent condition. supporting the hypothesis. For the Reciprocation-Praise conditions. however. schema-inconsistency failed to increase the positivity of judgments of her; in fact, inconsistency appeared to reduce them. For both Script conditions. then. inconsistency produced less favorable evaluations of the female's traits. 59 Table 6 Means for Script Effects on the Female's Traits Scr'n Condition Trait Reciprocation-Praise Nonreciprocation-Derogation Charm 6.13 4.29 Sexy 5.37 4.26 Sophisticated 5.69 4.84 Colorful 5.83 4.29 Extraverted 4.33 3.05 Socially Assertive 3.84 2.77 Happy 6.84 4.95 Sensitive 7.71 6.22 Sympathetic 6.97 6.02 Prosocial 5.48 4.14 Pleasant 7.62 6.77 High Strung 5.42 6.49 Positive 6.56 5.40 Ability 6.26 5.31 Mastery 5.64 4.66 60 Table 7 Main Effects of Video Prime on Judgments of the Female's Traits Video Pr'rne Condiion Trait Consistert Irrelevart Inconsistert Socially Assertive 3.03a 3253b 3sz Nasty Girl 2.51 ab 2.17b 3.568 Sexy 5.56a 4.81b 4.21b Ability 6.29a 5'72a,b 5.44b Sensitive 7.95.. 6.86b 6.27c Sympathetic 6.87a 7'05a 57% Positive 6.64a 6.04b 5.40c N919- Row means not sharing a common subscript are significame (p < .05) different. Significant prime effects are reported in Table 7. When the female's behavior was schema-consistent. rather than schema-inconsistent. she was judged sexier, E (2.169) . 13.15. p < .001, more socially assertive. E (2.169) - 3.19. n < .05, more sensitive, E (2.169) - 16.22. p < .001. and more positive. E (2.169) = 7.22. p < .01 . She was also judged higher in ability. E (2.169) . 4.15. p < .02. and lower on the mean and nasty factor. E (2,169) - 33.56. p < .001 . Inspecting the tabled means. it becomes clear that the inconsistent prime decreased the favorability of some of the female's trait ratings. Also of interest are the effects for the Irrelevant Prime conditions. In general, the neutral videos produced trait ratings which fell between the consistent and inconsistent conditions. Prime X Script interactions were obtained on two of the female actor's trait factors: openness. E (2.169) = 10.41. p < .001. and mean and nasty. E (2.169) s 9.95.31 < .001. While she had been judged nastier in inconsistent than consistent conditions, simple effects tests 61 indicated that this was not an accurate picture. Simple effects tests showed that the Prime effect was significant at both levels of Script (see Table 8). but multiple range tests performed within each script condition indicated that the effects were somewhat different in the two contexts. For Nonreciprocation-Derogation. the female was judged nastier in the schema-inconsistent than in the schema-consistent or -irrelevant conditions. following the previous pattern. For Reciprocation-Praise, however. she was judged somewhat more mean and nasty in consistent and inconsistent conditions than in the irrelevant prime condition. It should be kept in mind. though. that the means are very low in all but the Nonreciprocation-Derogation/Schema-inconsistent condition, and the small differences at the extreme end might not be meaningful. As should become clear later. within-script differences probably reflect subjects' reaction after nonstereotypic videos that the female's reciprocation causes her to come across as something of a “wallflower.” but her nonreciprocation is viewed more favorably than after stereotypic videos. The Prime X Script interaction effect on openness reflected a different pattern. The Script effect was found to have been entirely produced by differences in the consistent prime conditions. Only in the consistent prime condition was the female judged as more open when she reciprocated than when she did not; she was judged equally open in irrelevant and inconsistent conditions. regardless of script condition. Wrthin Reciprocation-Praise conditions. she was judged more open following the consistent prime than following the irrelevant or inconsistent prime. From Table 9 which presents the relevant condition means. it can be seen that the typical pattern of effects was repeated for the Reciprocation-Praise script. The female's reciprocation caused her to be seen as more open in schema-consistent than schema-inconsistent conditions. but her openness after nonreciprocation was not affected. 62 Table 8 Prime X Script Interaction on Ratings of "Mean and Nasty“ for the Female Condion Consistert Imelevart lnconsisterl Reciprocation-Praise 2.80b 2-18a 3-08b Nonreciprocation-Derogation 2.19a 2.16a 4.08,, N913. Row means not sharing a common subscript are significantly (p < .05) different. Table 9 Prime X Script Interaction on Ratings of ”Openness" for the Female Condition Consistent Irrelevan lnconsistert Reciprocation-Praise 7.50a 6.61b 6°33b.c Nonreciprocation-Derogation 5.48c 6.74b 6-37b N916. Means not sharing a common subscript are significame (p < .05) different. 63 Table 10 Attenuation and Reversal of Script Effects Produced by Consistent and Inconsistent Video Primes on Trait Ratings of the Female's Positivity and Ability Scr'pt PrineCondtion Condition Consister't Irrelevart Inoonsistert PositMty Trat Ratings Reciprocation-Praise 7.35a 5'93b.c 6.39b Nonreciprocation-Derogation 5.85c 6.14b 4.30d Abiity Trai Ratings Reciprocation-Praise 7.17a 4°98c.d 6.55“, Noreciprocation-Derogation 5.31 c 6.43b 4.23., Note. Positivity and ability means not sharing a common subscript are significantly (p < .05) different. The two remaining interactions provide some insight into how the female was perceived after the irrelevant prime. For judgments of her positivity. E (2.169) a 6.58. p < .01. and ability. E (2.169) = 26.01. p < .001. the pattern of effects was different for neutral videos than the pattern obtained on previous trait ratings. Recall that trait ratings after neutral videos generally fell somewhere between those that followed consistent and inconsistent videos. From that pattern, it seemed that the impact of priming was relatively simple. but the interactions of prime and script on the female's positivity and ability suggested that priming effects may have been more complex (see Table 10). Replicating effects found in the earlier video experiment, the usual pattern of script effects was overturned in Neutral prime conditions on both these judgments. When subjects had watched neutral videos. the female's reciprocation did not increase her positivity over that of nonreciprocation. Subjects did not value her reciprocation as highly when they had 34 not seen the stereotypic videos. In addition. after neutral videos they rated her as significantly lower in ability when she reciprocated than when she did not. contrary to the pattern that was found in the other prime conditions. The pattern of effects suggests that the meaning of the female's behavior probably shifted after neutral videos. Referring back to Table 7. one can see that the female's trait ratings at times resembled the consistent condition. at other times the inconsistent condition, or sometimes fell midway between. Adding to this variation the effects on her positivity and ability. it seems likely that her behavior was phenomenologically different after neutral videos. not simply more "neutral." W. A 3 (Prime: Schema-consistent. Schema-irrelevant. Schema-inconsistent) X 2 (Script: Reciprocation-Praise. Nonreciprocation-Derogation) MANOVA was performed on the male trait variates. As expected. impressions of the male were different in the two social interaction conditions. When the male praised rather than derogated the female. he was judged more positive. E (1,171) =- 5.42. p < .03 (Ms = 6.68 for Reciprocation-Praise and 5.89 for Nonreciprocation-Derogation) and much higher on the nice guy factor. E (1 .171) - 22.43. p < .001 (M3 - 5.64 and 4.52). When he derogated her. though. he was appraised as more masculine, E (1 .171) - 8.26, p < .01 (ML: . 7.84 for Nonreciprocation-Derogation and 7.25 for Reciprocation-Praise). and higher in both social power. E (1 .171) = 5.72. p < .02 (M3 = 7.89 and 7.46). and social energy. E (1 .171) = 6.32. p < .02 (M15 =- 8.09 and 7.67). Thus. the male was perceived to be more forceful when he derogated the female. but at the same time. he also was judged much less nice. Prime impacted significame only on the nice guy factor, E (2.171) . 4.86, p < .01. and. marginally, on positive. E (2.171) - 2.48.]: < .09. Simple effects analyses indicated that the male was seen to be more of a nice guy in the consistent prime conditions than in the inconsistent or irrelevant prime conditions (see Table 11). Whether he praised or derogated the female. then. when his behavior was consistent with the priming videos. the male gained points here. Because he had been judged less forceful in the Reciprocation-Praise condition in general. though. there 65 is a suggestion that in the inconsistent, Reciprocation-Praise condition. the male's behavior may have made him seem more like a "wimp“ than in the other conditions. After the disharmonious, boy-dumps-girl videos. subjects may have been slightly put off by his praise of her; it may have seemed gratuitous. Table 11 Prime X Script Interaction on Ratings of "Nice Guy" for the Male Scr'pt PrImeCondtion Concfion Consistert lrrelevart Inconsistert Reciprocation-Praise 6.16a 5.28:, 5'44b Nonreciprocation-Derogation 5.02a 4.39b 4.19b Note. Row means not sharing a common subscript are significamly (n < .05) different. Similar patterns of effects were obtained on trait judgments of the two actors. although differences were much stronger for the female. Her reciprocation and his praise led to more positive traits after the consistent, boy-meets-girl videos than after the inconsistent. boy-dumpsgirl videos--a pattern that does not support the information-processing hypothesis. Thus. for the socially harmonious interaction. positive impressions were attenuated by schema-inconsistency. In addition. the female's nonreciprocation and the male's derogation led to a less negative impression of both actors when their behavior was schema-consistent than when it was schema-inconsistent. Negativity of impressions of both actors during the socially disharmonious interaction. then. was attenuated when their behavior was congruent with the priming videos. as predicted by the infonnation-processing hypothesis. E | II General usability. The six favorability items (three for each actor) were entered into 3 (Prime: Schema-consistent. Schema-irrelevant. Schema-inconsistent) X 2 (Script: Reciprocation-Praise, 66 Nonreciprocation-Derogation) MANOVAs. For both actors. the ”like personally“ item produced Prime main effects. E (2.172) = 9.14, p < .001 (for him). and E (2,172) = 2.99. p - .053 (for her). As suggested by the trait ratings. both actors were liked better when their behavior was schema-consistent than when it was schema-inconsistent. The means are presented in Table 12. A Prime X Script interaction qualified the female's general likability. E (2.172) = 4.62. p < .01. but not the male's. E (2.172) <1. 9 = .48. Table 12 Prime Main Effects on Overall Favorability of Impressions of the Two Actors Prine Conrition Actor Consistert lrrelevart Inconsistert Female (“like [her] personally") 7.14a 4.78b 5.06b Male (”like [him] personally”) 5.79a 3.25c 4.68b N913. Row means not sharing a common subscript are significantly (p < .05) different. Mama. Simple effects tests indicated that the Prime effect was significant for the Reciprocation-Praise script condition. E (2.172) = 28.50, p < .001, but not for the Nonreciprocation-Derogation condition. E (2. 172) . 2.39. p <.10. From the relevant condition means reported in Table 13, It is clear that the female was exceptionally well-liked when her reciprocation was schema-consistent--much more than in the other two prime conditions. When she did not reciprocate, however. prime had no overall effect. When subjects saw neutral videos, their liability ratings reversed in the two script conditions. In the consistent condition. Reciprocation-Praise was significantly higher than Nonreciprocation-Derogation, E (1 .171) = 4.71. p < .03. and in the inconsistent condition. Reciprocation-Praise was slightly but not significantly higher than Nonreciprocation-Derogation. E (1 .171) - 1.22. p < .27; however, in the irrelevant prime condition. Nonreciprocation-Derogation 67 was significantly higher than Reciprocation-Praise. E (1 .171) = 3.93. p < .05. It appears that the female's general favorability was greatly reduced when she reciprocated after neutral videos. This reversal parallels the effects reported earlier for the female's ability and positivity. Thus. it seems that when subjects were not exposed to either of the stereotypic primes. her reciprocation made her appear less likable as well as less capable. Table 13 Prime X Script Interactions on Overall Favorability of Impressions of the Female Actor Condition Consistert Inelevari Inconsistent "Ike [heft W Reciprocation-Praise 7.87a 4.04d 5'42b,c Nonreciprocation-Derogation 6.30b 5.48“: 4.670 Note. Favorability means not sharing a common subscript are significantly (p < .05) different. lame. East Inc analyses of the Prime effect on the male's general favorability (see Tables12) indicated that he was liked much more when his behavior was schema-consistent than when it was schema-inconsistent or schema-irrelevant. Also, note that the irrelevant prime conditions produced the least favorable ratings for both scripts. One fact seems clear from the analyses so far. When subjects saw either condition of sex-role stereotypic videos. impressions of the male were advantaged over impressions following the irrelevant prime. However. this pattern did not occur for the female. Only behaviors that were consistent with the stereotypic primes augmented a positive evaluation of her. For both actors. the largest effects on impressions consistently occurred in schema-consistent conditions. It could be that the male's behavior in both script conditions was more easily assimilated to both sex-role stereotypic schemas. affording him more positive ratings regardless of whether he praised or derogated the 68 Table 14 Main Effects of Video Prime on Favorability of Getting Acquainted and Audition Impressions of the Female and Male Actors Prirle Concition Favorabiliy Measure Cons'sterl lrrelevar! lnoorsistert Irrpressions ofthe Female Actor Getting Acquainted Favorability 6.26a 4.79b 4.86b Audition Favorability 6.86 6.21 6.00 Inpreesions of the Male Actor Getting Acquainted Favorability 5.78 4.72 4.90 Audition Favorability 7.38a 5.40b 63% Note. Row means not sharing a common subscript are significantly (p < .05) different. The means for female audition favorability and for male getting acquainted favorability are shown for corrparison purposes (the Prime X Script interactions were not significant). female. It seems likely that. for both actors. schema-inconsistent behaviors were generally devalued compared to schema-consistent ones. Also, it is interesting that neutral videos reduced the male's positivity in both social interactions, but only reduced the female's when she reciprocated the male's sexual advances. In this study. without the benefit of a sex-role stereotypic video prime, subjects devalued both types of male sex-role stereotypic behavior (praise or derogation) and stereotypic female reciprocation. but not female nonreciprocation. Wale. Ratings of the female's performance during the getting acquainted segment were similar to her general favorability ratings. Results showed a Prime main effect, E (2,172) . 5.43, p < .01 and a Prime x Script interaction. E (2.172) . 5.63. p < .01. A, multiple range test indicated that the female was given higher ratings in the schema-consistent condition than in the 69 other two prime conditions (see Table 14). Simple effects tests, however. showed the prime effect to be significant only in the Reciprocation-Praise condition. E (2.171) =- 19.96, p < .001. The means and specific group comparisons underlying the Prime X Script interaction are presented in Table 15. As in the general favorability question. the female's getting acquainted activity produced a reversal in appraised favorability after the neutral videos. In the schema-consistent condition. Reciprocation-Praise was evaluated more favorably than Nonreciprocation-Derogation. E (1,172) = 4.37. p < .04. whereas in the schema-inconsistent condition the scripts were rated equally. E (1 .172) = .07. p a .79. In the irrelevant prime condition. though. the means reversed, and the female's nonreciprocation was evaluated more positively than her reciprocation, E (1 .172) - 6.85. p < .01. Table 15 Prime X Script Interactions on Favorability of Getting Acquainted and Audition Impressions of the Female Actor Prine Condtion Scr'pt Condition Consisterl Irrelevant lnconsistert Getting Aoqrirted Favorabin Reciprocation-Praise 6.94a 3.86b 4°94b Nonreciprocation-Derogation 5-48a 5.69a 4.77b Audion Favoraafly Reciprocation-Praise 7.29 6.25 6.31 Nonreciprocation-Derogation 6.37 6.17 5.67 N918. Row means not sharing a common subscript are significantly (p < .05) different. Audition favorability means are shown for comparison purposes only. The Prime X Script interaction was not significant on the measure of audition favorability. 70 In general. it appears that whenever subjects were exposed to stereotypic videos, the female's behavior was appraised more favorably when it was schema-consistent. After they had seen the neutral videos. however. the picture was very different. As found for general favorability. the female was given a quite low rating when she reciprocated the male's sexual advances in the "getting acquainted” segment. It seems likely that when a sex-role stereotypic schema was primed, subjects liked traditional female social behavior. but that when a stereotypic schema had not been primed. her reciprocation (while applying for a job) seemed inappropriate--and perhaps diagnostic of low competence--and. hence. disliked. There were no significant Prime or Script effects for the female's behavior during the ”audition” segment (recall that her behavior was not manipulated during this segment). The means for the three prime conditions are shown in Table 14. The female's ratings are quite similar across all experimental conditions, although there appears to be a trend toward the familiar pattern of more favorable judgments when behavior is schema-consistent compared with schema-inconsistent. For purposes of comparison with ratings during the "getting acquainted” segment. the means for all six experimental conditions for the ”audition” segment are displayed in Table 15. It is interesting to note that subjects in all conditions gave her a relatively high rating for her performance in this segment. Even after a relatively unfavorable evaluation for her reciprocation of his advances after neutral videos, she was able to substantially improve her ratings by achieving a positive performance outcome (his praise) during the "audition" segment. Subjects apparently could believe that the female was performing more capably during the "audition” segment, despite her earlier poor start. Male. For the male. the picture is somewhat different. There was only a significant effect for Script. E (1 .172) - 9.90, n < .01. on his performance during the ”getting acquainted“ segment (when his behavior was not manipulated). He was always appraised more favorably in the Reciprocation-Praise than in the Nonreciprocation-Derogation condition (M s 5.87 and M = 4.34). For ratings of the male's performance during the ”audition” segment. however. there is a 71 significant Prime effect, E (1 .172) . 6.97, p < .01 (see Table 14). As in his general favorability ratings. reported earlier. the neutral videos decreased the male's favorability regardless of script condition. Positive appraisal of the male's behavior was augmented when subjects saw either stereotypic prime. For both actors. the general favorability rating appeared to be a composite of the individual segment evaluations. To test this inference, ratings for each segment were entered into a multiple linear regression to predict the general favorability ratings. Indeed, for both actors, the two segments were good predictors. For the female. both the ”getting acquainted“ and ”audition" segments were significant predictors of the overall rating, 1'5 (176) =- 4.70 and 2.39. 915 < .001 and < .02, respectively. with a total adjusted R2 of .29. p < .001. For the male. the adjusted R2 with both predictors was .40. n < .001, again with both ”getting acquainted“ . t (176) - 5.49. p < .001. and "audition" .1(176) =- 3.27. p < .001. segment ratings being significant predictors of the overall rating. A pattern in the data emerged which suggests that the video primes were directing the valences of observed behavior. Each of the stereotypic primes could have provided an expectancy for the actors' behavior. The boy-meets—girl videos increased the favorability of Reciprocation-Praise behaviors. but, interestingly, the boy-dumps-girl prime. while it reduced favorability somewhat, still produced ratings above or equal to those after neutral videos. The same pattern is true for the Nonreciprocation-Derogation script. Favorability of appraisals was increased after the consistent boy-dumps-girl prime. but the inconsistent boy-meets-girl prime only slightly reduced favorability and never reduced it as much as did seeing neutral videos. Valence of the behaviors. per as. did not seem to have much impact in the current test. Similar effects were found for the harmonious and disharmonious social interactions. What seemed to be more important was whether the behaviors matched the schematic expectancy-regardless of whether the expectancy was harmonious or disharmonious. In fact. during the socially disharmonious interaction. favorability ratings of both actors seemed to benefit as much as ratings 72 for the harmonious interaction from being consistent with the priming videos. Interactions reflected reversals of script effects in the neutral prime condition relative to schema-consistent and Inconsistent prime conditions. WW Four categories of behaviors were extracted from judges' coded responses of the subjects' minimally structured recall: the male's "coming on" and the female's reciprocation or nonreciprocation in the ”getting acquainted“ segment. the male's subsequent praise or derogation in the ”audition” segment. and residual (other behaviors). Wrthin each of the four behavior categories. the judges scored the behavior as one that was actually ”available” in the interaction script or as a nonavailable "intrusion." After coding each recalled item. judges totalled the number of responses in each of the eight categories, and these totals comprised the units of analysis. To look at lnterjudge agreement. the two judges' codings within each of the eight categories were correlated (Pearson 1). The correlation coeficients are reported in Table 16. Averages of the two judges' ratings were the data analyzed. A 3 (Prime: Schema-consistent. Schema-irrelevant. Schema-inconsistent) X 2 (Script: Reciprocation-Praise. Nonreciprocation-Derogation) MANOVA was conducted on the eight behavior categories. Recall that the social information-processing hypothesis predicted that schema-inconsistent behaviors would be better recalled than schema-consistent behaviors. The analysis indicated that this was indeed the case. W. Three of the four available categories showed Prime effects. Only the residual category was uninfluenced by priming condition. E (2.172) - 1.14, p a .32. Significant effects were found for ”coming on”. E (2,172) . 12.78. p < .001. reciprocation/nonreciprocation. E (2.172) - 12.03. p < .001. and praise/derogation. E (2.172) - 19.66. p < .001. The means for these effects can be found in Table 17. Interestingly. the male's ”coming on” behaviors. which were the same in both scripted interactions (and. of course, consistent with both stereotypic video primes), showed greater recall after either stereotypic prime than after the neutral videos. 73 These effects are an example of an instance in which schema-congruence serves to aid memory for behaviors. But, for reciprocation or nonreciprocation and praise or derogation. schema-inconsistent behaviors appear to have an even greater advantage than schema-consistent behaviors. Table 16 lnterjudge Reliability on Minimally Structured Recall Categories Coding caegory Correlation (g1 .. 176) His "coming on” Available .96 Intrusions .84 Her Reciprocation or Nonreciprocation Available .94 Intrusions .82 His Praise or Derogation Available .90 Intrusions .53 Residual Behaviors Available .97 Intrusions .89 Two of the coded categories represented scripted behaviors which were manipulated to be schema-consistent. schema-inconsistent. or schema-irrelevant. In both of these categories (reciprocation or nonreciprocation. praise or derogation). subjects recalled significantly more schema-inconsistent than schema-consistent behaviors, a pattern that is consistent with the social infonnation-processing hypothesis. For the fourth (residual) category. no difference in 74 recall would be expected because these behaviors were designed as fillers. consisting of common conversational topics; and no significant effects were found for this measure. Two additional findings were somewhat surprising. There was also a significant Script effect for the two manipulated behavior categories. More nonreciprocation than reciprocation behaviors were recalled. E (1 .172) = 23.94. p < .001 (M3 = 1.50 and .78. respectively). and more derogation than praise behaviors were recalled. E (1 .172) =- 4.87. p < .03 (M3 = .65 and .45). Thus. it appears that recall in this study was better for disharmonious (or negative) social behaviors. especially when they were schema-inconsistent. This pattern parallels the finding presented earfier. regarding the female's and male's trait judgments: higher ratings were given to her on the ”mean and nasty“ factor (Table 8) for her nonreciprocation when it was schema-inconsistent; and lower ratings were given to him on the "nice guy“ factor (Table 11) when his derogation was schema-inconsistent. At the very least. then. it can be said that there does appear to be a relationship between negative impressions and recall of socially disharmonious behaviors. The relationship might not be as strong, however. for recall of socially positive behaviors and impressions. mm. In general, intrusion errors (recall for behaviors that could be coded as “coming on.“ residual. reciprocation or nonreciprocation. and praise or derogation but that did not actually occur in the interaction) followed the same pattern as recall for available behaviors (see Table 17). Prime effects occurred in reciprocation/nonreciprocation, E (2.172) = 29.91. p < .001. and in praise/derogation. E (2.172) .. 26.44. p < .001. and. once again. they were due to greater numbers of schema-inconsistent intrusions. There were no Script effects for either ”coming on” or residual behaviors (both Es approximately equal to 1. ns.). A Script effect was obtained both for "coming on," E (1 .172) a: 5.15. p < .03 (M3 = .09 for Nonreciprocation-Derogation and .01 for Reciprocation-Praise) and reciprocation/nonreciprocation behaviors. E (1 .172) a 4.03. p < .05 (M3 - .39 and .28). There were so few intrusions. however. and so little variance. that these 75 findings should be regarded cautiously. The pattern for intrusions followed that for correct recall. In general, there were more schema-inconsistent and negative (nonreciprocation) intnlsions. Table 17 Video Prime Main Effects on Minimally Structured Recall Prine Cordtion Recall Category Consistert lrrelevart Inconsistert His “coming on" Available 3.1 0 a 1 .83b 3.23a Intrusions .02a .09a .05a Her Reciprocation or Nonreciprocation Available .78a .95a 1 .62b Intrusions .18a .18a .62b His Praise or Derogation Available .28a .33al '94b Intrusions .10a .'°7a .60., Residual Behaviors Available 7.21 a 7.36a 8.50a lntnrsions .64a .42a '04a Note. Row means not sharing a common subscript are significamly (p < .05) different. W. A Prime (Schema-consistent. Schema-irrelevant. Schema-inconsistent) X Script (Reciprocation-Praise. Nonreciprocation-Derogation) X 2 (Actor: Male. Female) mixed ANOVA was conducted on the number of behaviors recalled for each actor. Several significant effects were generated. First, a significant Prime effect was found. E (2.172) - 9.95. p < .001 (M5 .. 6.15 for consistent. 5.61 for 76 irrelevant, and 7.98 for inconsistent). Second. a significant Actor effect indicated that more male (M - 7.69) than female (M = 5.56) behaviors were recalled. E (1 .172) a 202.95. p < .001. In addition. the Prime X Actor interaction was significant. E (2.172) = 9.08. p < .001. Simple effects tests (see Table 18) indicated that the most male and female behaviors were recalled in the schema-inconsistent conditions. Table 18 Video Prime Main Effect on Minimally Stnrctured Recall for All Behaviors Attributed to the Female and to the Male PrineCondtion Actor Consistert Irrelevart Inconsistert Female 4.88a 5.01 a 6.69b’ Male 7.42b 6.21 a 9.27c Note. Means not sharing a common subscript are significamly (p < .05) different. Actor also interacted with Script to influence recall. E (1 .172) = 7.10. p < .01. A comparison of within-subject difference scores (number of male minus number of female responses) indicated that there was less difference in the number of male versus female behaviors recalled in the Nonreciprocation-Derogation condition than in the Reciprocation-Praise condition. E (1 .172) - 6.76. p < .01 (mean differences equalled 1.71 in Nonreciprocation-Derogation and 2.51 in Reciprocation-Praise). Recall that subjects recalled almost twice as many female nonreciprocation behaviors than reciprocation behaviors. a pattern which may largely account for the interaction. In general. though. subjects listed more male than female behaviors and more inconsistent than consistent behaviors. W Mag. lnterjudge agreement was assessed for coding of recall accuracy by correlating the two judges' scores within each of the ten stmctured recall statements. The correlation 77 coeficients are shown in Table 19. Averages of the two judges' ratings were used as data for these analyses. Accuracy scores on the ten statements were submitted to a 3 (Prime: Schema-consistent, Schema-irrelevant. Schema-inconsistent) X 2 Script (Reciprocation-Praise. Nonreciprocation-Derogation) MANOVA. Table 19 lnterjudge Reliability for Accuracy of Structured Recall Measures Cueing Statemerl Correlation (d - 176) “Getting Acquainted” Cueing Statements (Recall Her Reciprocation or Nonreciprocation) ”That's a nice blouse . . ." (1) .99 '. . . nice material. . ." (2) .99 ”If I knew. . thermostat. . ." (3) .97 "I'm really excited . . . concerts.“ (4) .95 “Audition" Cueing Statements (Recall His Praise or Derogation) ”Bmce is one of Americas . . ." (1) .76 '. . . better produced videos.” (2) .92 "It's not what you see . . ." (3) .99 "He . . . new mood of patriotism." (4) .96 ". . . missing in action there" (5) .95 ”Lot's of . . .about the Boss.” (6) .93 Accuracy. lnterjudge agreement was assessed for coding of recall accuracy by correlating the two judges' scores within each of the ten structured recall statements. The correlation coeficients are shown in Table 19. Averages of the two judges' ratings were used as data for these analyses. Accuracy scores on the ten statements were submitted to a 3 (Prime: Schema-consistent. 78 Schema-irrelevant. Schema-inconsistent) X 2 Script (Reciprocation-Praise. Nonreciprocation-Derogation) MANOVA. Table 20 Video Prime Main Effects on Accuracy of Structured Recall Statements Pr'meCondibn Main Effect Cueing Statement Consister! Inelevant lnconsistert E (2,172) "Getting Acquainted“ Cueing Statements (Recall Her Reciprocation or Nonreciprocation) "That's a nice blouse . . ." (1) 1.66a 1.97a 4.25b 42.44‘ ". . . nice material . . 3' (2) 1.3631 2.67b 3.60G 1698* "If I knew. . .therrnostat . . ." (3) 1.53a 2.08a 3.21b 2018‘ "I'm really excited . . . concerts." (4) 1.92a 2.07a 3.09,, 13.04: ”Audition“ Cueing Statements (Recall His Praise or Derogation) "Bnlce is one of Americas . . ." (1) .09a '07a 1.12b 308.99‘ ”. . . better produced videos.“ (2) .18a .41b 1.33G 5587‘ ”It's not what you see . . ." (3) .77a .82a 1.64b 6.04“ "He . . . new mood of patriotism." (4) .08a .35b 1.51c 51 .02‘ ". . . missing in action there" (5) .28a .36a 2.01., 97.88‘ "Lot's of . . .about the Boss." (6) .16a .14a 1.33:, 5852" N919- Row means not sharing a common subscript are significantly (p < .05) different. * p < .001 “p< .003 W. Responses to all four cueing statements produced Prime effects (see Table 20). Once again. schema-inconsistency augmented recall. this time for the specific. manipulated conversational statements. Responses in this segment were recall of the female's reciprocation or nonreciprocation statements. Her responses were. of course. different in 79 the two script conditions. There was one significant Script effect; on the first question. subjects had better recall for the nonreciprocation response than the reciprocation response, E (1 .172) . 68.80. p < .001; however. the nonreciprocation response was a simpler one. namely ”thank you .“ which may have accounted for the recall difference. Statement four generated a significant Prime X Script interaction, E (2,172) = 3.88. p < .03, but again. this effect is uninterpretable except to say that the inconsistent prime produced a greater effect in one script condition than in another. The means are shown in Table 21. The pattern suggests that the more difficult response (reciprocation) was especially benefitted by schema-inconsistency. Table 21 Prime X Script Interaction on Stnlctured Recall for the Female's Reciprocation or Nonreciprocation Statement Four (“He . . .new mood of patriotism") Script PrineCondtion Condiion Consistert lrrelevart lnconsistert Reciprocation-Praise 1 .909 1 .75d 3.44a Nonreciprocation-Derogation 1 .946 2.380 2.70b N912. Means not sharing a common subscript are significamly (p < .05) different. MW. Recall of responses in the six ”audition" segment statements, in which subjects were asked to recall the male's praise or derogation. shows the same pattern as the "getting acquainted” segment (see Table 20). Responses to the cueing statements were recalled with greater accuracy after an inconsistent prime than after a consistent or neutral prime. Four Script effects also occurred. For statement one, E (1 .172) - 19.65. p < .001. the male's derogation remarks were recalled more accurately than his praise (M's. - 3.74 and 1.69). On statement three (M15 . .52 and 1.70) and five (M15 - .78 and 1.07). there was also better recall for 80 derogation responses. E5 (1 .172) a 26.70 and 6.33. Q's < .001 and < .02. respectively). On the last statement (six). though. the male's praise was more accurately recalled than his derogation. E (1 .172) - 8.02, p < .01 (Mi: - .72 and .41). Once again, Script effects can be interpreted as differences in the inherent diffictu of one of the responses over the other. Table 22 Prime X Script Interaction on Structured Recall for the Male's Praise or Derogation Statement Five (". . . missing in action there") Condlion Consistert lnelevart lnconsistert Reciprocation-Praise .19d .39c 1 .65b Nonreciprocation-Derogation .39c .33c 2.40a N919. Means not sharing a common subscript are significame (p < .05) different. There was one Prime X Script interaction-on statement five, E (2,172) = 4.34. p < .01, for which means are shown in Table 22. Recall was most accurate in the inconsistent Nonreciprocation-Derogation condition. From the Script effect, it appeared that the Nonreciprocation-Derogation response was inherently more difficult to recall accurately. so. once again. a conservative explanation would be that schema-inconsistency augmented the accuracy of difficult-to-recall statements. Of course. a less conservative explanation would be that inconsistency aided recall of socially disharmonious (derogation) remarks (which also occurred). Wall. lnterjudge agreement for coding the strength of the recalled replies was good (see Table 23). The averages of the ratings from the two judges were used as data in the analysis. A 3 (Prime: Schema-consistent. Schema-irrelevant. Schema-inconsistent) X Script (Reciprocation-Praise. Nonreciprocation-Derogation) MANOVA was conducted on strength judgments for each of the ten stnrctured recall items. If schema-inconsistent information attracted 81 more attention (or was processed more deeply), it is logical that schema-inconsistent behaviors would also be recalled more strongly. That is. because it is more salient, the same behavior would be perceived as having been stronger (e.g. stronger reciprocation of his advances or stronger derogation) when it was made schema-inconsistent by priming. Table 23 lnterjudge Reliability for Strength of Structured Recall Measures Cueing Statement Conelation (d -176) "Getting Acquainted“ Cueing Statements (Recall Her Reciprocation or Nonreciprocation) "That's a nice blouse . . ." (1) .98 ”. . . nice material. . ." (2) .97 "If I knew. . .therrnostat . . ." (3) .89 "I'm really excited . . . concerts.” (4) .91 ”Audition“ Cueing Statements (Recall His Praise or Derogation) "Bnrce is one of Americas . . ." (1) .95 ". . . better produced videos.” (2) .95 "It's not what you see . . ." (3) .96 ”He . . . new mood of patriotism." (4) .98 ". . . missing in action there” (5) .91 ”Lot's of . . .about the Boss." (6) .94 Due to the format used for response scaling (numbers from zero to minus three for the Nonreciprocation-Derogation responses and zero to plus three for Reciprocation-Praise responses). all items produced significant Script effects. The means for the two script conditions are displayed in Table 24. There was only one significant Prime effect. for the first ”getting acquainted“ segment item. E (2.170) = 4.17. p < .02. On the first item. the tendency to recall the 82 female's nonreciprocation response as stronger than her reciprocation response was particularly pronounced in irrelevant prime conditions MS a -.23 for schema-consistent, -.55 for irrelevant. and -.10 for inconsistent). Table 24 Script Main Effects on Accuracy of Stnlctured Recall Statements Sa'ptCordtion Ma'nEfled Recbrocation-Praise Noriecbrocation-Derogation E(1.170) Cueing Statement ”Getting Acquainted“ Cueing Statements (Recall Her Reciprocation or Nonreciprocation) "That's a nice blouse . . ." (1) .88 -1.55 454.91 ”. . . nice material . . ." (2) 1.25 -.59 152.20 "If I knew. . .therrnostat . . ." (3) .76 -1.36 178.94 "I'm really excited . . . concerts.” (4) 1.28 -1.09 221.06 "Audition” Cueing Statements (Recall His Praise or Derogation) "Bmce is one of Americas . . ." (1) .48 -.25 68.54 ”. . . better produced videos." (2) .41 -.40 76.24 "It's not what you see . . ." (3) .34 -.60 70.63 ”He . . . new mood of patriotism." (4) .49 -.24 73.87 ". . . missing in action there" (5) .60 -.43 103.16 "Lot's of . . .about the Boss." (6) .42 -.15 45.12 N919. All E values are significant (p < .001). The effects that are most relevant to the schema-inconsistency hypothesis are the Prime X Script interactions. Significant interactions were obtained on three of the structured recall items. All three were reciprocation or nonreciprocation replies during the ”getting acquainted“ segment. 83 The E5. within-group means. and simple effects are shown in Table 25. From the table, it can be seen that the recalled responses were significame stronger in the schema-inconsistent conditions than in the schema-consistent conditions, with strength for the irrelevant condition falling in-between. This is a nice finding. as it indicates that schema-consistency affects not only the amount of recall. but the quality of the recall as well. Inconsistent behaviors. it seems, also appear stronger when they are contrasted with primed schemas. Table 25 Prime X Script Interactions on Stnlctured Recall for Three of the Female's Four Reciprocation and Nonreciprocation Statements Condition Consistert Inelevarl Inconsistent Cueing Statement 1: ”That's a nice blouse . . ." E (2,170) =- 40.84. p < .001 Reciprocation-Praise .39a .48a 1 .67b Nonreciprocation-Derogation -.94a -1 .61 b 2030 Cueing Statement 2: ". . . nice material . . ." E (2.170) - 19.01 . p < .001 Reciprocation-Praise .63a 1 '34b 1 .76b Nonreciprocation-Derogation -.17a -.33a -1 .20b Cueing Statement 4: "I'm really excited . . . concerts."E (2.170) - 9.44. p < .001 Reciprocation-Praise .87a .98a 1 .91 b Nonreciprocation-Derogation -.74a -1 ~19a,b -1 .32b Note. Row means not sharing a common subscript are significantly (p < .05) different. Possible scale vanes ranged from -3 (recalled nonrecbrocation response was stronger than the scripted response) through +3 (recalled rec‘procation response was stronger than the scripted response). 84 ! I 'l' . Meaningnflbejasls. Forty random subjects answered an open-ended question which was attached to the end of the questionnaire booklet: "In the instructions, you were asked to press the button when a meaningful behavior had been performed. Briefly describe what these instructions meant to you and what constituted a meaningful behavior for you." Responses were coded into four categories: (a) candidates completed one topic of conversation and began another. (b) candidates interacted socially in a positive way. (c) candidates made meaningful physical movements and gestures. and (d) other miscellaneous responses. Data from two subjects were excluded from analysis because their answer was uncodable in any single category. Table 26 Main Effect of Self-Described Unitizing Strategy on Number of Units During the "Getting Acquainted” Segment Completed Topic Socially Positive Body Movements ofConversation Behaviors andGestures Mm Unitization 18.2.0 6.603 12.63b 5.57..I N912- Means not sharing a common subscript are significantly (p < .05) different. Analym. Using the four ”meaning” categories as grouping factors, ANOVAs were performed on (1) number of'getting acquainted” segment units, (2) number of "audition" segment units, and (3) number of behaviors recalled in four coded categories: his “coming on," her reciprocation or nonreciprocation. his praise or derogation, and residual behaviors. Reported meaning produced no effects on recall or on the "audition” segment unitizing (all E5 approximately equal to 1. ns.). It did produce differences in the ”getting acquainted” unitization. however, E (3.34) a 3.23. p < .04. means for which are reported in Table 26. Subjects who 85 reported that they pressed the button to indicate topics of conversation or meaningful movements and gestures generated nrore units during the “getting acquainted" segment than did subjects in the other two categories. Table 27 Contingency Table of Self-Described Unitizing Strategy X Experimental Condition Indicating Noninterdependence of Strategy and Condition Self-Descrbed Unitizing Strategy Corrpleled Topic Socialy Positive Body Movements ExperinentalCondtion Misc. of Conversation Behaviors ar‘d Gestures Prime x 2 (6) = 4.56. p < .60 Observed Frequency Consistent 5 1 4 3 Irrelevant 2 1 6 3 Inconsistent 3 3 6 1 Script x 2 (3) = 1.54. p < .68 Observed Freqrency Reciprocation-Praise 7 3 8 3 Nonreciprocation-Derogation 3 2 8 4 To explore whether the type of priming videos or social interaction to which subjects were exposed influenced the reported meaning of unitizing. two contingency tables were analyzed. Neither Prime nor Script appeared to influence meaning (see Table 27). since. neither X 2 was significant. For this small sample, then, the meaning of the button presses gave the appearance of being unassociated with either the type of videos or the type of social interaction they watched. If subjects' insight into the meaning of unitizing is valid, these results suggest that unitizing is not linked to schema-inconsistency. Even though the instructions to subjects were the same as those in previous unitizing studies. asking subjects to chunk a rapid social 86 conversation into "meaningful units” appears to have produced varied. idiosyncratic results. The number of button presses ranged in this experiment from a single press to eighty-five presses. with a mean of 21.10 (and a standard deviation of 16). Obviously, large variance was produced by the instmctions. Data from the small subsample indicated that subjects reported having interpreted the instmctions in a number of ways. and they seemed to generate a wide range of chunking strategies. W. For each subject. unitizing responses were manually receded from the audiotapes onto verbatim, written transcriptions of the interaction scripts. This enabled a calculation of the timing of button presses throughout the two conversational segments. Unitizing intervals in twenty-three behavior periods and during the remaining portions of the interaction were calculated for each subject (nine behavior periods corresponded to his ”coming on“ behaviors. eight to her reciprocation or nonreciprocation. six to his praise or derogation. and the balance of the interaction constituted the residual period). The coded behavior periods can be found in Appendix M. As a preliminary analysis, unitizing during ”coming on" behaviors, reciprocation or nonreciprocation behaviors. and praise or derogation behaviors were used as within-subjects factors in separate 3 (Prime) X 2 (Script) ANCOVAs. Because the length of the behavior periods varied slightly between the Reciprocation-Praise and Nonreciprocation-Derogation scripts. the statement length for each behavior period was used as a covariate. Neither the nine level "coming on" factor, the eight level reciprocation or nonreciprocation factor. nor the six level praise or derogation factor entered into any interactions with either Prime or Script (all Es approximately equal to 1, ns.). Therefore. in subsequent analyses. mean unit intervals for each category of behaviors were used as the data for analysis. Wham. A 3 (Prime: Schemaconsistent. Schema-irrelevant. Schema-inconsistent) X 2 (Script: Reciprocation-Praise. Nonreciprocation-Derogation) was conducted on the four timing categories. Results strongly 87 supported the social information processing hypothesis. Only one Script effect emerged, E (1.168) a 4.40, p < .04. During parts of the interaction in which the female reciprocated or deflected the male's advances, the time between units was longer when she reciprocated than when she did not. (M's - 14.47 and 11.99 seconds). Prime produced effects on all but the residual behavior category (refer to Table 28). For residual portions of the interaction (filler behaviors that were not relevant to the priming manipulation). subjects generated units of about the same intenrnit interval in all conditions regardless of which set of priming videos they had seen (M a 29.16 seconds). For his "coming on“ behaviors. smaller units were generated by the neutral videos than by either stereotypic prime. This pattern would be predicted by the social information hypothesis, too. since within this perspective. his ”coming on" behaviors were consistent with both stereotypic primes. and. thus. could be processed in larger chunks. The means and results of multiple range tests are shown in Table 28. Table 28 Prime Effects on Unitizing During Portions of the Interaction When the Male was ”Coming On." When the Female was Reciprocating or Not Reciprocating, When the Male was Praising or Derogating. and During the Residual Prine Condiion Portion of Interaction Consistent Irrelevarl Inconsiaert E (2.168) ”Coming On" 17.1%I 13.48b 19.24.. 4.02. p < .02 Reciprocation or Nonreciprocation 15.24a 13393,!) 11.25b 3.84. p < .03 Praise or Derogation 20.04a 16:19am 14.62b 2.75. n < .07 Residual 27.95 28.19 31.23 1.00, n < .88 Note. Row means not sharing a common subscript are significantly (p < .05) different. Times are in seconds. Total times for portions were: ”Coming On" . 35.4. Reciprocation or Nonreciprocation a 21.4. Praise or Derogation . 41.2. Residual . 113.4. 88 During the remaining. manipulated behavior portions of the interaction. finer units were generated in schema-inconsistent than schema-consistent conditions, as predicted. Interunit intervals during the reciprocation or nonreciprocation behavior periods were shorter when the female's behavior was inconsistent with the priming videos than when it was consistent. The Prime effect for the male's praise or derogation behaviors was marginally significant; from Table 28 it can be seen that finer units were generated when his behavior was schema-inconsistent than when it was consistent. Overall. the data support the social information processing hypothesis well. For instance. the priming videos did not influence filler portions of the interaction. And. while they had an effect on his “coming on' behaviors. it was exactly what would be predicted-larger units for both the “schema-consistent” and "schema-inconsistent“ than irrelevant conditions. The priming condition labels are. of course. a misnomer in this behavior category. as ”coming on“ is consistent with both stereotypic primes. Also. and most critical to the hypothesis. the manipulated behaviors of the actors were chunked into finer units when they were inconsistent with the priming videos than when they were consistent. Bll' I' fill" IE II Minimally Stuntman Bacall. To test whether unitization predicted memory in the first recall measure, mean unitization times for the four coded categories (his ”coming on,” her reciprocation or nonreciprocation. his praise or derogation. and residuals) were entered into multiple linear regressions to predict recall of (1) total number of his “coming on” behaviors. (2) total of reciprocation or nonreciprocation behaviors. (3) total praise or derogation behaviors, and (4) total residual behaviors. Results of the regressions are shown in Table 29. Evidence for the link between the size of units and recall for behaviors was mixed. Unitizing of the female's reciprocation or nonreciprocation responses during the interaction did significantly predict how many behaviors subjects recalled in this category. The finer the unitizing. the greater the recall. as predicted. However. unit size during the male's praise or 89 Table 29 Multiple Linear Regressions to Predict Minimally Structured Recall from Unitizing Wrthin Portions of the Interaction Significant Predictor Standardized Regression Coefficient Simple I: t Recall for His “Corning oh“: Multiple 32 (3.169) = .06. p < .03 ”Coming On" Unitizing .18 .14 2.02. n < .05 (Non)Reciprocation Unitizing -.22 -.06 2.61. p < .01 Recall for Her Reciprocation or Nonreciprocation: Multiple 32 (3.169) = .12. n < .001 (Non)Reciprocation Unitizing -.34 -.33 4.21. n < .001 Recall for His Praise or Derogation: Multiple 32 (3.169) = .04, n < .17 No Significant Predictors Recall for Residual Behavior: Multiple 32 .- .04. p < .07 No Significant Predictors derogation and during residual behavior periods failed to predict recall in their corresponding categories. Interestingly, unitizing of his ”coming on” behaviors did predict recall for that category, but. in this case. larger units predicted greater recall. After the variance for this predictor was extracted by the regression. however. finer unitizing of her reciprocation or nonreciprocation also significantly predicted how many ”coming on” behaviors subjects would recall. Because his ”coming on” was followed in almost all instances by her reciprocation or nonreciprocation response, it may be that some subjects considered his “coming on" and her response to it as an integrated conversational unit. When unitizing occurred during her response, it may have represented encoding of the entire conversational unit and not merely her reciprocation or nonreciprocation. That recall for his "coming on" was predicted by larger units is certainly consistent with the notion that recall was enhanced by schema—consistency or schematic 90 processing in this instance. rather than by inconsistency. While support for the hypothesis that smaller units would result in greater recall received mixed support for the correspondence between unitizing and recall. unitization during the whole interaction sequence did. in fact. predict overall recall. R2 = .03, n < .04. The finer the units generated during the interaction. the more total behaviors subjects recalled. WW. Unitizing during the critical interaction behaviors was hypothesized to be predictive of accuracy of recall for those statements. Because the length of the statements subjects were asked to recall was slightly different in the two script conditions. Script effects were controlled for when testing for a link between unitization and accuracy of the structured recall responses. The unitization variance due to Script was held constant by using partial correlations between mean unitization time during each interaction statement and accuracy of recall. The (partial) correlation coeficients are shown in Table 30. The weak associations found between mean unit intervals and accuracy of recall of the individual statements suggests that finer units did not enhance recall memory for the verbatim responses. W. Similar partial correlations were conducted on strength ratings. As shown by the correlation coeficients in Table 31. the relationship between finer unitization of an actor's individual statement and how strongly the statement was recalled was not strongly supported. Only one correlation (the response for the fifth ”audition segment" item) was statistically significant. but the direction was opposite that predicted by the social information processing theory. suggesting that it may be simply a chance association. all I' “II" II'III I Mean unit intervals for the four coded categories were regressed on traits and trait factors which showed significant Prime or Script effects for both actors. Unit size was not significantly associated with the extremity of any trait judgments for the male. but it was related to judgments on several of the female's traits. Results from the significant regressions can be found in Table 32. The pattern of significant effects is the same for each of the female's traits. Smaller units 91 during her reciprocation or nonreciprocation combined with larger units during his ”coming on” did predict lower trait rating on (positive) traits of ”sensitive”. ”sympathetic”. and the “openness" trait factor. As seen in the earlier section on recall. it appeared that his ”coming on” and her reciprocation or nonreciprocation may have been treated as integrated conversational units. but that finer chunking which occurred on her responses seemed to be associated with more extreme trait ratings. While the findings are suggestive of an association between fine units and more extreme trait judgments. there were so few significant effects that these findings should be accepted with some caution (especially since there have been several failures to replicate this effect. first found in Newtson's 1973 experiment). Table 30 Partial Correlations of Unitizing During Statement with Accuracy of Structured Recall for Statement CueingStatemert Conelation(d-176) ”Getting Acquainted" Cueing Statements (Recall Her Reciprocation or Nonreciprocation) "That's a nice blouse . . ." (1) -.06 '. . . nice material. . ." (2) -.02 "If I knew. . .therrnostat . . ." (3) .10 "I'm really excited . . . concerts.” (4) .04 "Audition" Cueing Statements (Recall His Praise or Derogation) ”Bnrce is one of Americas . . ." (1) .06 ". . . better produced videos.” (2) .06 ”It's not what you see . . ." (3) .05 "He . . . new mood of patriotism." (4) -.08 ". . . missing in action there” (5) .08 "Lot's of . . .about the Boss.“ (6) .00 92 Table 31 Partial Correlations of Unitizing During Statement with Strength of Structured Recall for Statement Cueing Statemert Correlation (d = 176) ”Getting Acquainted" Cueing Statements (Recall Her Reciprocation or Nonreciprocation) “That's a nice blouse . . ." (1) -.13 ". . . nice material. . ." (2) -.10 “If I knew. . .thermostat . . ." (3) .09 ”I'm really excited . . . concerts." (4) .05 “Audition” Cueing Statements (Recall His Praise or Derogation) 'Bnlce is one of Americas . . ." (1) -.02 '. . . better produced videos.” (2) .01 ”It's not what you see . . ." (3) .09 “He . . . new mood of patriotism." (4) -.05 ". . . missing in action there” (5) .15‘ "Lot's of . . .about the Boss.“ (6) .09 'p<.05 93 Table 32 Multiple Linear Regressions to Predict the Female's Traits From Unitizing VWthin Portions of the Interaction Significant Predictor Standardized Regression Coefficient Simple I 1 Sensitive: Multiple 32 (3,171) = .12, p < .001 "Coming On" Unitizing -.03 -.14 2.17. p < .04 (Non)Reciprocation Unitizing .23 .23 4.39. p < .001 Sympathetic: Multiple 132 (3.171) = .08, p < .003 "Coming On" Unitizing -.03 -.16 -2.16, p < .04 (Non)Reciprocation Unitizing .06 .16 3.04. n < .003 Openness (Factor): Multiple 52 (3,171) = .06, p < .02 "Coming On” Unitizing -.03 -.14 2.34. p < .02 (Non)Reciprocation Unitizing .04 .15 2.68. p < .009 94 Discussion Wm Exposure to sex-role stereotypic rock videos significantly altered subjects' appraisal of a subsequent interaction between a male and a female. A harmonious social interaction (she reciprocates his sexual advances. and he praises her) generally resulted in more positive appraisal of the actors than the disharmonious interaction (she does not reciprocate. he derogates her). but these appraisals were also influenced by the type of priming videos to which subjects were exposed. When behavior during the social interaction was consistent with ailing; the boy-meets-girl videos or the boyodumps-girl videos. the favorability of impressions formed of the actors was enhanced. When their behavior was inconsistent with either stereotypic prime, trait judgments were more negative and overall impressions were less favorable than when their behavior was schema-consistent. For example, when the male derogated the female during the interaction. he was viewed as more of a “nice guy" when his behavior was consistent with the priming videos than when it was inconsistent. Similarly. when she did not reciprocate his advances. she was evaluated as less ”mean and nasty” when her behavior was consistent with the priming videos than when it was not. It was predicted from social information processing theory that when perceivers observed schema-inconsistent behaviors. those behaviors would become salient and would produce trait and favorability ratings which were exaggerated away from the direction of schema content. This pattern was expected to occur regardless of the valence of the observed behaviors. Schema-consistent behaviors. on the other hand, were predicted to be assimilated to the accessible schema and, thus. were not expected to become salient or to weigh as heavily in appraisal. The predictions were borne out for the disharmonious but not the harmonious social interaction. Subjects' more negative evaluation of the disharmonious interaction was attenuated after consistent boy-dumps-girl videos, but their positive evaluation of the harmonious interaction 95 was not augmented by being contrasted with the boy-dunps-girl videos. Instead. the positive evaluation suffered by contrast and was reduced by schema-inconsistency. Recall that videos for both stereotypic priming conditions were selected because even though they portrayed two antagonistic behavioral scripts (boy-meets-girl versus boy-dumps-girl). they were equally well-liked. This is probably an important feature of the manipulation. Observed behaviors which were inconsistent with behaviors portrayed in ljlsan videos were devalued. The hamionious script was liked less after an inconsistent. but well-liked video prime, and the same held true for the dishamionious script. Assimilation and contrast, then, may have operated in this study in a slightly different (but, perhaps. more realistic) way than results from previous studies and theoretical explanations would have predicted. Schema-consistent behaviors appeared to have been assimilated to the valence of both primed schemas (which in this case were positive). Schema-inconsistent behaviors. however. were contrasted with the positive valence of behaviors contained in the primed schema. making them seem more negative. The most positive impressions were formed when subjects were primed to like Reciprocation-Praise behaviors by seeing the boy-meets—girl videos; when the primed schema led them to expect reciprocation and praise, and they subsequemly observed it. evaluations of the actors were extremely positive. Similarly. when they were primed to expect nonreciprocation and derogation by the (liked) boy-dumps—girl videos, the disharmonious interaction did not produce evaluations which were as negative as they would have been ordinarily. The finding that schema-consistency enhances positive evaluations is consistent with other research on media effects. For example. social psychologists studying the impact of long-term exposure to erotica have found that people come to like it more over repeated exposure (Byme. 1977; Zillmann 8 Bryant. 1983). What appears to be happening in this context is that positively-valenced schemas for erotica are being built (cf. Zajonc. 1968). and that observation of schema-consistent behaviors in subsequent films becomes even more positive because the behaviors are easily assimilated to 96 the positive hedonic tone of the schema. Specifically. when subjects were massively exposed to pomography by Zillmann 8 Bryant (1983). their evaluation of schema-related. real-life sexual behaviors was influenced. In questionnaires. their attitudes toward sexual behavior and legal pornography were much more positive than the attitudes of controls. In addition, when these subjects viewed a filmed rape trial and were asked to recommend punishment for the defendant. their sentences were more lenient than those recommended by control subjects. This is an important finding, because it points out that not only are schema-consistent behaviors more positively evaluated (e.g. sexual behavior). but stimulus generalization effects also occur so that previously negative behaviors (rape). if they can be assimilated to the schema. will take on a more positive tone. For purposes of our study of sex-role stereotypic behaviors. this suggests that not only will stereotypic behaviors be assimilated to positively primed sex-role stereotypic schemas. but with repeated exposure, even sexist behavior might take on a positive valence. While the results of the current experiment support the conclusion that appraisal of schema-inconsistent social behaviors was generally less favorable than appraisal of schema-consistent behaviors. the mechanism underlying this effect remains unclear. It is possible that schema-inconsistency is inherently negative-perhaps because of the increased effort required to encode and make sense of it. Alternatively, though. the effects of inconsistency on evaluation may depend on the valence of the primed scheme. The effects found for schema-inconsistency in the current experiment might be unique to instances in which the primed schema had a positive hedonic tone. Appraisals of an interaction that is inconsistent with a disliked primed schema might become mom positive. rather than less. A subsequent experiment which compares the effects of schema-consistency and schema-inconsistency on printing videos in which the same themes are portrayed, but the videos induce either a positive or negative hedonic tone. would address this important question. Appraisals after neutral videos were substantially different from those after either sex-role stereotypic prime. The male in both social interaction conditions was less likable after neutral 97 videos than after either stereotypic prime; both stereotypic primes augmented his favorability. The same was true for the female. but only in the hamionious Interaction in which she reciprocated the male's advances. When a sex-role stereotypic schema had not been primed. subjects dislflred her (stereotypic) acquiescence. Her nonreciprocation. on the other hand. made her seem more likable. This is an interesting and hopeful finding (and replicates similar results from the previous video study) because it suggests that in the absence of a sex-role stereotypic prime. subjects liked her better and judged her more competent when she deflected the male's advances than when she reciprocated them--a somewhat surprising finding even though it is a replication of past work (Hansen 8 Hansen. 1986). It is important. though, not to lose sight of the other side of the coin. Stereotypic, even sexist, behaviors are more likely to be condoned when stereotypic schemas have been primed. Although it is tempting to conclude from the findings obtained in the Irrelevant Prime conditions that the kinds of sex-role stereotypic behaviors portrayed in the two interactions are generally disfavored unless they have been primed in a positive way (as by the stereotypic videos). the picture may not be this simple. While the neutral videos were selected because they did not prime sex-role stereotypic behaviors (and. thus. were assumed to be irrelevant to interpretation of the subsequent social interaction), it seems unlikely that the neutral prime had no effect at all on impressions or evaluation. Rather. the themes primed by the neutral videos (see Appendices C.1-C.8. theme 3) may, themselves, have contributed to the less favorable evaluations given to actors engaging in traditional stereotypic social behaviors. The neutral videos may have triggered schemas suggestive of "higher" values. such as interpersonal striving or universal harmony. And. against this background. the stereotypic (even sexist) behaviors during the social interaction may have been evaluated more negatively than they would have been without the neutral videos. Regardless of the precise role of the neutral videos. It does seem clear that the meaning of both actors' behavior was phenomenonologically different after the neutral videos than after the 98 stereotypic videos. Perhaps the male's sexual advances and the female's reciprocation were constmed as inappropriate behaviors for two job candidates after neutral videos but not after stereotypic videos. His ”coming on” might have seemed more like sexual harrassment without the stereotypic prime. There is some suggestion in the trait judgments. for example, that the impressions of the actors were somewhat different. not simply more or less favorable. depending on the type of video prime subjects had been exposed to. For instance. the male appeared to be judged weaker for prsising the female after subjects had seen the Boy-meets-girl videos than after they had seen the (consistent) Boy-dumps-girl videos. Phenomenologically. he may have seemed more like a ”wimp" when subjects had just seen a macho Image portrayed by the videos. So. not only was he less liked. but the meaning of his behavior may have been different. Other research suggests that such ”shifts of meaning“ do occur (Hamilton 8 Zanna, 1974). and future research might explore how priming impacts on differences in the meaning of social behavior for shifting the kinds of impressions formed as well as how priming affects the positive and negative evaluation of such behaviors. The effects of priming in this study are consistent with much of the social cognition literature if one incorporates the valence of the primed schema into existing models. Assimilation to a positively valenced schema (such as the script of a well-Inted rock music video) could exaggerate the positive valence of observed behaviors when they are schema-consistent. This suggests an important consideration. Social harmony and disharmony may be relative phenomena after all. Positivity and negativity may very much depend on the particular schema with which the behaviors are being encoded. This conclusion implies that behaviors which might ordinarily be perceived as negative (such as a male's derogation of a female) will not seem so bad after people have seen similar behaviors portrayed positively in the media. If this line of reasoning is carried to its logical conclusions. the negative behavioral implications are clear. In the current experiment. appraisal was altered significantly after only a short exposure to rock videos. Future research might be directed toward determining whether 99 changed appraisal would generalize to behavior toward the people being appraised. From previous research on aggressive behavior (e.g. Geen. 1976. 1983), there is every reason to expect that it would. Second. future experimentation might concern itself with the generalization of priming effects from rock music videos to other types of schemas and behaviors. A Ifltely candidate, at present. would be positive portrayal of antisocial behaviors in well-liked videos. For instance, one video currently at the top of the (MTV) chart ("You've got to fight for your right to party“ by the Beastie Boys). portrays two nerds (the Beastie Boys) who are turned into heroes by throwing a wild party against the expressed wishes of their parents. As the title inplies, all the guests fight wildly (mostly with pics) for their right to party. as the house is turned into a shambles and the boys lose their virginity to several “sex-starved nymphomaniacs" in the bathroom. The song ends with mom and dad returning . . . to pies in the face. And (unfortunately). ”Fight for your right to party” is great rock music. It has all the necessary components of a perfect prime for antisocial behavior (and it has been in the top twenty for the last couple of months and has all the signs of becoming a classicl). Perhaps more research producing effects like this will show that it is time that television begins to take some responsibility for the effects of the content of its portrayals. W? Why was unitizing not a better predictor of impressions? Why did it not predict recall for specific kinds of responses. when it did predict. overall, the number of items subjects recalled? While Newtson's early finding that more units were related to exaggerated trait judgments was not conclusively supported, similar failures to replicate have been reported by Ebbesen (1980). There are several possible explanations for these negative results. According to Newtson's (1973. 1976. 1980) model. ongoing behavior Is encoded in meaningful cognitive units. It is known that unitizing strategies are different when subjects are attempting to form an impression than when they are trying to pay close attention to an actor's precise behaviors for a later memory test (or when the goal is to learn to imitate these behaviors). When attending to an actor’s specific 100 actions, subjects generate more units and can recall more of the actor's behaviors than can subjects given instructions to form an impression of the actor. In previous experiments. when subjects unitted under these different instmctional sets. Ebbesen (Ebbesen, Cohen. 8 Allen) found that overall trait judgments did not differ--they were no more extreme for subjects who generated many units than for subjects who generated few. There did appear to be a difference, however. in how the individual trails were correlated with each other in the two conditions. Pairwise correlations of the measured traits were quite different for subjects with impression versus behavior instructional sets. Associations between the traits for impression unitizing subjects resembled those of subjects who had simply observed the interaction, without the unitizing task. but inter-trait associations for behavior unitizing subjects appeared to be different. Ebbesen (1980; Ebbesen, Cohen, 8 Allen. 1979) has suggested that trait judgments from the behavior unitizing subjects may have been based more upon implicit personality theory (Cantor 8 Mischel, 1979). In other words. when subjects are attending to actions, but are asked to appraise personality, they may rely more heavily upon normative. previously learned trait associations than impression subjects. For impression subjects, appraisal may more closely resemble actual. observed. personal attributes (see also Cohen. 1981b). " However. results for the nonunitizing pilot study. while they showed the same pattern of effects as the unitizing study. were much stronger. Trait judgments appeared to be more polarized by priming when subjects were not asked to unitize. This conclusion does suggest that irrpressions when unitizing are not always the same as when people are merely observing. even though everyone is given the same set of instructions-to form an impression. Ebbesen (1980) has also argued for a modification of Newtson's notion of encoding; he proposes a dual processing model. He feels that people ordinarily have parallel. dual encoding strategies in which they encode ongoing behavior in large. more abstract, units designed for forming impressions and in smaller. activity-oriented units, representing the specific behaviors performed by the person. He feels that these ordinary strategies may be interfered with when 101 people are given instructions to unitize actions. Because the default seems to be to the more abstract. impression level, though. it may make just as much sense to assume that unitizing small. behavioral units is an unusual strategy (of. Wegner, et al.. 1984) as it does to assume that perceivers have a secondary encoding process for ongoing actions. Somehow the dual processing model is not satisfying. especially since smaller units are reliably associated with greater recall of behaviors. A more parsimonious explanation might be that encoding ongoing behavior in small units is representative of increased attention, and that increased attention enhances recall. Like Newtson, Ebbesen believes that unitizing strategies can be primed. as attempted in this experiment. and that primed schemas should direct encoded cognitive units. but so far he and his collegues have been unsuccessful in inducing measurable appraisal differences by changing subjects' unitizing strategies. In the current experiment, all subjects were given an impression set. but looking at the wide variance in button pressing. the most parsimonious explanation for the weak support may be that some subjects were simply paying more attention to finer behaviors in the social interaction. and that these subjects showed greater recall. but that unitization m 53 does not alter the ability to form an inpression. It would be interesting to know whether the apparent ability to make personality appraisal even while unitizing fine behaviors is based on traits implied by the behaviors or on a default to normative values. If trait judgments turn out to be based on normative prior probabilities. it would support the idea that fine behavior unitization Is debilitating of forming an impression rather than the notion that people ordinarily use a dual encoding strategy. There is another theoretical approach that might extricate us from the dilemma. Markus. Smith. and Moreland (1985) found that when subjects schematic for masculinity were asked to generate fine, behavioral units in a schema-relevant behavior sequence. they were better able to perform the impression and the behavior recall tasks in parallel than were subjects aschematlc for masculinity. The authors explained this pattern by assuming that schematics have a great deal of 102 stored trait information which they can retrieve in making appraisals. Asking schematic subjects to unitize in fine units did not seem to affect their performance on trait judgments. For the present experiment. it is assumed that all subjects were schematic for the stereotypic schemas which were primed. If true. then it is not entirely surprising that regardless of their idiosyncratic unitizing strategies, unitizing. itself. did not alter trait judgments. Markus et al. (1985) did find. however. that even for schematics. recall was enhanced by instructions to generate very fine units. Until further experimentation is done. then. the failure of unitizing to uniformly predict trait extremity is probably best explained as an attentional phenomenon in which the unitizing task enhances memory for observed behaviors. In some ways, the current findings raised more questions about the interrelationships among impression judgments, memory for behavior. and unitization than they answered. Greater recall for behavior was reliably associated with finer unitization. Both appeared to be forced by the schema-inconsistency of behavior. Generating finer units may simply represent more effort devoted to the encoding of schema-inconsistent behavior. Of course, if schema-inconsistent behavior is more effortfully or more deeply processed. a recall advantage would be anticipated (as suggested by Hastie. 1980,1981). Similarly. if one were to argue that schematically-consistent behavior can simply be integrated over larger units and still be understood. while schema-inconsistent behavior can not (cf. Wegner, et al., 1984), the same findings would be obtained. But. either of these approaches would seem to conflict with the results obtained on impression judgments if impressions were solely. or even largely. guided by what was remembered. For example. his praise might be expected to result in a particularly favorable impression of him when praise was schema-inconsistent, because it was well-remembered. But, it did not. Indeed. the pattern of effects obtained on impressions. recall. and unitizing suggested that they may not be strongly linked. The data certainly fell short of suggesting that schemas influence impressions by determining which behaviors are remembered or how well they are 103 l Hedonic . Tone Behavior Stream I A primed Input Schema i Behavior Consistent with Schematic Representation? (NO) I A I (YES) 7 _ IReclprocal Hedonic Tone Integral Hedonlc Tone I 7 Elaborated Processing: Minimal Processing: I Shift Meaning or Contrast Assimilation Fine Gross Unitizing Unitizing Encode Not Encode ‘— Behavioral Behavioral —‘ Detail Detail | Understanding I [Reciprocal Evaluation I I Schematic Evaluation V Lfil Impression [‘J ' Recall Advantage Recall Disadvantage For Behavioral For Behavioral Details Details Figure 2. Process Model Showing Relationships Among Recall. Unitizing. and Impressions 104 remembered. Results from the current study would best be characterized as follows: The more subjects recalled of the behavior and the finer the detail of the recall and unitizing. the less favorable were the impression judgments they made. All of these things. of course. resulted when behaviors were schema-inconsistent. The model depicted in Figure 2 was constnrcted to illustrate these findings in a coherent framework. As shown. the model implies that a behavior sequence which is consistent with (matches) a primed schema can be understood with only minimal processing. Schema-consistent behavior can be encoded (understood) at a relatively abstract level and can be integrated (unitized) over rather long periods. As a result. of course. behavioral details are lost to memory. and impressions are based. largely. on schematic content and hedonic tone. A behavioral sequence that is inconsistent (mismatched) with a primed schema, however. must be additionally processed before it can be understood (cf. Schank. 1977). Schema-inconsistent behavior must be encoded at a relatively concrete level and can be integrated (unitized) only over short periods. As a result. behavioral details are more accessble to memory. Further. the model implies that the additional processing involves reconstruing behavior in such a way that it is compatible with the schematic context. Thus. the meaning of behavior could be phenomenologically different when behaviors are being encoded with different primed schemas. For example. the male's praise in the context of the primed boy-dumps-girl schema, rather than enhancing his image (i.e. a contrast effect). may have made him look like a 'wimp‘lor not seeing through her sexual ploy. In other words. schema-priming could exert an influence beyond altering memory for behaviors on which impressions are based. Primed schemas might also influence the way in which what is remembered is understood. It could be argued that the positively evaluated rock music videos primed positively-toned schematic content. lnpressions of schema-consistent interactions would be based primarily on schema content, rather than attention to actual behavioral details and would be anticipated to be favorable. And. they were. 105 But. impressions based more on behavioral details that. presumably, were less positive than those embodied in the rock music videos, would be expected to reduce the favorability of impressions-which they did. Whether this model is a reasonable constmction of the effects obtained in the current experiment requires further research. As an empirical model. it offers the hypothesis that impressions are multiply influenced by primed schemas and challenges the notion that schemas influence impressions exclusively by mediating what is remembered. LIST OF REFERENCES Alba. J. W.. 8 Hasher, L. (1983). Is memory schematic? Esygnglngjgalfiullaljn. 23. 203-231. Anderson. J., 8 Bower. G. (1973). W. Washington. DC: Winston. Andison. S. F. (1977). TV violence and viewer aggression: a cumulation of study results 1956-1976. W 41. 314-331. Ashmore. R. D.. 8 Del Boca, F. K. (1979). Sex stereotypes and implicit personality theory: Toward a cognitive-social psychological conceptualization. ms, 5, 219-248. Ashmore. R. D.. 8 Del Boca. F. K. (1981). Conceptual approaches to stereotypes and stereotyping. In D. L. Hamilton (Ed.), WWW nanaylgr. Hillsdale, NJ: L. Erlbaum. Atkin. C.. 8 Miller. M. (1975). The effects of television advertising on children: Experimental evidence. Paper presented at the annual meeting of the International Communication Association. As cited in Eisenstock (1984). Bandura, A. (1965). Influence of models' reinforcement contingencies on the acquisition of imitative responses. WW. 1, 589-595. Bandura, A. (1973). W. Englewood Cliffs. NJ: Prentice-Hall. Bargh. J. A. (1984). Automatic and conscious processing of social information. In R. S. Wyer. Jr. 8 T. K. SI'UII (Eds). WW (Vol. 3). Hillsdale, NJ: L. Erlbaum. Bargh. J. A. Bond. R. N.. Lombardi. W. J.. 8 Tota. M. E. (1986). The additive nature of chronic and temporary sources of constnlct accessibility. JnumaLQLEarsnnaljtyandfiogial mm 5.0. 869-878- 106 107 Bargh. J. A.. 8 Pietromonaco. P. Automatic information processing and social perception: The influence of trait information presented outside of conscious awareness on impression formation. WW 4.3. 437-449. Bargh. J. A.. 8 Pratto, F. (1986). Individual constnlct accessibility and perceptual selection. MW 22. 293-311- Bargh. J. A.. 8 Thein, . R. D. (1985). Individual construct accessibility. person memory. and the recall-judgment link: The case of information overload. WW Esyglnlggy, 49,1129-1146. Baron. R. M. (1980). Contrasting approaches to social knowing: An ecological perspective. WW 4. 591-600. Berkowitz. L. (1965). The concept of aggressive drive: Some additional considerations. In L. Berkowitz (Ed.), WWW. (Vol. 2). New York: Academic Press. Berkowitz. L. 8 LePage, A. (1967). Weapons as aggression-eliciting stimuli. ,lmmalgf W1. 202-207. Berkowitz. L.. 8 Rogers. K. H. (1986). A priming effect analysis of media influences. In J. Bryant 8 D. Zillmann (Eds). WM. Hillsdale, NJ: L. ErIbaum. Brewer. M. B.. Dull. V.. 8 Lui, L. (1981). Perceptions of the elderly: Stereotypes as prototypes. W 5.1, 656-670. Bower. G. H.. 8 Cohen, P. R. (1982). Emotional influences in memory and thinking: Data and theory. In M. S. Clark 8 S. T. Fiske (1593-).an WW. Hillsdale, NJ: L. Erlbaum. Byrne. D. (1977). The imagery of sex. In J. Money 8 H. Musaph (Eds). tlandhgnkgLsamlggy. Amsterdam: Elsevier/Norlh-Holland Biomedical Press. Cantor. N. 8 Mischel, W. (1977). Traits as prototypes: Effects on recognition memory. .lgumalgt W 35. 38-48. 108 Carver, C. S.. Ganellen. R. J.. Froming. W. J.. 8 Chambers. W. (1983). Modeling: An analysis in terms of category accessibility. W. 19. 403-421. Chase, W. G.. 8 Simon. H. A. (1973). The mind's eye in chess. In W. G. Chase (Ed.), flaual lnl9nnatl9n_9:99939jng. New York: Academic Press. Coates, B., Pusser. H. E.. 8 Goodman (1976). The influence of ”Sesame Street” and "Mister Roger's Neighborhood" on children's social behavior In the preschool. W. 51. 138-144. Cobb. N. J.. Stevens-Long. J.. 8 Goldstein, S. (1982). The influence of televised models on toy preference in children. SELBQIQS. 8. 1075-1080. Cohen, C. E. (1981). Goals and schemas in person perception: Making sense out of the stream of behavior. In N. Cantor 8 J. Kihlstrom (Eds.).E_a:§9nallly._9999ltj9n_an9_§99(aL99l1ayi9:. Hillsdale, NJ: L. Erlbaum. (a) Cohen. C. E. (1981). Person categories and social perception: Testing some boundaries of the processing effects of prior knowledge. .l99:na|_9LEa:s99allty_an9_S99iaLEsy9l19l9,gy. 59, 441452. (b) Comstock, G. (1980). New emphases in research in effects of television and film violence. In E. L- Palmer 81 A Dorr (Eds). W New York: Academic Press. Comstock, G.. Chaffee. S.. Katzman, N.. McCombs, M.. 8 Roberts. D. (1979). Ialayi9i9n_an9 W. New York: Colunbia University Press. Crocker. J.. Hannah. D. B.. 8 Weber, R. (1983). Person memory and causal attributions. mm W a. 55-66. Dohrmann, R. (1975). Agender profile of children's educational TV. .l99:naL91_Q9mnmnj9ati9n, 25, 56-65. 109 Donnerstein, E. (1983). Erotica and human aggression. In R. Geen 8 E. Donnerstein (Eds), W. New York: Academic Press. Ebbesen, E. B. (1980). Cognitive processes in understanding ongoing behavior. In R. Hastie. T. M. Ostrom. E. B. Ebbesen. R. S. Wyer, Jr.. D. L. Hamilton. 8 D. E. Cariston (Eds), 39:59n WW. Hillsdale, NJ: L. Erlbaum. Ebbesen. E. B.. Cohen. C. E., 8 Allen. R. B. (1979). Cognitive processes in person perception: Behavior scanning and semantic memory. Unpublished manuscript. University of California at San Diego. Eisenstock. B. (1984). Sex-role differences in children's identification with counterstereotypical televised portrayals. W. 19, 417-430. Eron. L. D.. 8 Huesmann, L. R. (1980). Adolescent aggression and television. Annals_9l_ma W341. 319-331. Eron, L. D.. Huesmann, R. L. Lefkowitz, M. M., 8 Walder, L. O.. (1972). Does television violence cause aggression? Am9n9an£sycn9l99m 21. 253-263. Erber. R.. 8 Fiske. S. T. (1984). Outcome dependency and attention to inconsistent information. W .41. 709-726. Fazio, R. H.. Sanbonmatsu, D. M.. Powell. M. C.. 8 Kardes, F. R. (1986). On the automatic activation of attitudes. W 50. 229-238. Feldstein, J. H.. 8 Feldstein, S. F. (1982). Sex differences on televised toy commercials. $9): Boles. 3. 581 -586. Ferguson. T. J.. 8 Wells. G. L. (1980). Priming of mediators in causal attribution. .l99:naL91 W 38. 461-470. Fiske, S. T. (1980). Attention and weight in person perception: The impact of negative and extreme behavior. W. 38. 889-906. Fiske. S. T.. 8Taylor. S. E. (1984). S991aLQ9gnitl9n. Reading. MA: Addison-Wesley. 110 Friedrich. L. K.. 8 Stein. A. H. (1973). Aggressive and prosocial television programs and the natural behavior of preschool children. WWW Willem. 38. (4. Serial No. 151). Fulero. S.. 8 Rothbart. M. (1980). Encoding individual traits into group stereotypes: The effects of frequency and desirability of trait descriptors. As cited in Rothbart. M. (1981). Memory processes and social beliefs. In D. L. Hamilton (Ed.), WW 991W. Hillsdale, NJ: L. Erlbaum. Geen. R. G. (1976). Observing violence in the mass media: Implications of basic research. In R. G. Geen 8 E. C. O'Neal (Eds). W. New York: Academic Press. Geen. R. G. (1981). Behavioral and physiological reactions to observed violence: Effects of prior exposure to aggressive stimuli. W 4.0. 868-875. Geen, R. G. (1983). Aggression and television violence. In R. Geen 8 E. Donnerstein (Eds). W. New York: Academic Press. Geis. F. L.. Brown. V.. Jennings. J.. 8 Porter. N. (1984). TV commercials as achievement scripts for women. 592989199, 1Q, 513-524. Gerbner. G. (1978). The dynamics of cultural resistance. In G. Tuchman. A. K. Daniels. 8 J. Benét(Eds.). .:. = '. New York: Oxford University Press. Gerbner. G.. 8 Gross. L. (1976). Living with television: The violence profile. .l9un1aL9: Q9mmun19a119n, 25, 173-199. Gerbner. 6., Gross. L.. Morgan. M., 8 Signiorelli, N. (1980). The “mainstreaming” of America: Violence profile no. 11. J99:nal_9l_Q9:n:n9nj9ali99. 39. 10-29. Gerbner. 6., Gross, L.. Morgan. M.. 8 Signiorelli, N. (1986). Living with television: The dynamics of the cultivation process. In J. Bryant 8 D. Zillmann (Eds.),Ea:s9e91iyas_9n_:n991a_afl9919. Hillsdale, NJ: L Erlbaum. Goffman, E. (1979). mm. New York: Harper8 Row. 111 Graesser. A. C.. 8 Nakamura, G. V. (1982). The impact of a schema on comprehension and memory. In G. H. Bower (Ed.), WNM. 16). New York: Academic Press. Gross. L.. 8 Jeffries-Fox. S. (1978). “What do you want to be when you grow up. little girl?" In G. Tuchman. A. K. Daniels. 8 J. Benet (Eds.). W55 mania. New York: Oxford University Press. Hamilton. D. L. (1980). Person perception. In L. Berkowitz (Ed.), W99! (2nd ed.). New York: Holt. Rinehart, 8 Winston. Hamilton. D. L. (1981). Cognitive representations of persons. In E. T. Higgins. C. P. Herman, 8 M. P. Zanna (Eds.). WW. Hillsdale, NJ: L. Erlbaum. Hamilton. D. L.. 8Zanna. M. P. (1974). Context effects in inpression formation: Changes in connotative meaning. J9un1al9Ll39:59nalltyan9_S99ial_E5yan9l99y. 29. 649-654. Hansen. C. H.. 8 Hansen. R. D. (1985). Discriminating affective and cognitive differentiation of complex visual stimuli: Rock music videos. Unpublished manuscript Michigan State University. Hansen. C. H.. 8 Hansen. R. D. (1986). Priming stereotypic appraisal of social interactions: How rock music videos can change what's seen when boy meets girl. Submitted for publication. Hansen. 0. H.. 8 Hansen, R. D. (1987). The hidden message of rock music videos: sex role stereotypes. Presented at the meeting of the American Psychological Association. New York. Hansen. R. D. (1985). Cognitive economy and commonsense attribution processing. In J. H. Harvey & G- Weary (Eds.). WWW Orlando. FL: Academic Press. 112 Hastie. R. (1980). Memory for behavioral information that confirms or contradicts a personality inpression. In R. Hastie. T. M. Ostrom. E. B. Ebbesen. R. S. Wyer, D. L Hamilton, 8 D. O. Canmon(Eds.).E9§99m9m9m;19999gmm9a5§9L5991aL9am9m9n.Hillsdale. NJ: L Erlbaum. Hastie. R. (1981). Schematic principles in human memory. In E. T. Higgins. C. P. Herman. 8 M. P. Zanna (Eds.). WWW (Vol. 1). Hillsdale, NJ: L. Erlbaum. Hastie, R., 8 Kumar, P. A. (1979). Person memory: Personality traits as organizing principles in memory for behavior. W 31, 25-38. Hastie. R.. Park. 8., 8 Weber, R. (1984). Social memory. In R. S. Wyer. Jr. 8 T. K. Smll (Eds.). WW (Vol. 2). Hillsdale, NJ: L. Erlbaum. Heider, F.. 8 Simmel, M. (1944). An experimental study of apparent behavior. Aman9an_.l99:nal my. 51. 243-259. Herr. P. M.. Sherman, S. J., 8 Fazio. R. H. (1983). On the consequences of priming: Assimilation and contrast effects. J99:nal_9l_Ex99:lm9nlal_S99iaLflsyan9l99y. 19, 323-340. Higgins, E. T.. Bond. R. N.. Klein. R.. 8 Strausman, T. (1986). Self-discrepancies and emotional vulnerability: How magnitude. accessibility. and type of discrepancy influence affect. .l99:nal oLEarsonalitundfiociaLEsxcboloox. 51. 5-15. Higgins. E. T.. 8 Chaires, W. M. (1980). Accessibility of interrelational constnrcts: Implications of stimulus encoding and creativity. W, 16, 348-361. Higgins. E. T.. 8 King. G. (1981). Accessrblity of social constructs: Information processing consequences of individual and contextual variability. In N. Cantor 8 J. F. Kihlstrom (Eds.). WW. Hillsdale, N.J.: Lawrence Erlbaum Associates. Higgins. E. T.. King. G. A.. 8 Mavin, G. H. (1982). Individual construct accessibility and subjective impressions and recall. W .43. 35-17. 113 Higgins. E. T.. Rholes, W. S.. 8 Jones. C. F. (1977). Category accessibility and impression formation. W 13. 141-153. Howard. J. W.. 8 Rothbart. M. (1980). Social categorization and memory for ingroup and outgrouo behavior. W 38. 301-310. Huesmann, L. (1982). Violence and aggression. In National Institute of Mental Health, WW (Vol- 1). Washington. DO: U. 8. Government Printing Office. Jennings. J., Geis. F. L., 8 Brown. V. (1980). The influence of television commercials on women's self-confidence and independence of judgment. WNW E5y999|9gy, 99, 203-210. Johnston. W. A.. 8 Dark. V. J. (1986). Selective attention. Anmaljayimflfisygnm. 31. 43-75. Kanouse. D. E. (1972). Language. labeling, and attribution. In E. E. Jones. et al. (Eds.). W. Morristown. NJ: General Learning Press. Kanouse. D. E. (1972). Negativity in evaluation. In E. E. Jones. et al. (Eds.).Annhmm WW. Morristown. NJ: General Learning Press. Kulflt. J. A. (1983). Confinnatory attribution and the perpetuation of social beliefs. .l99:nal_9l W 44. 1171-1181. Langer, E. J., 8 Abelson. R. P. (1974). A patient by any other name...: Clinician group difference in labeling bias. J99mal_9l_Q995ulling_an9_Qlini9al_Esygn9l99y. 4.2. 4-9. Lefkowitz, M. M., Eron. L. D.. Walder, L. O.. 8 Huesmann, L. R. (1977). W. New York: Pergamon. Linvllle, P. W. (1982). Affective consequences of complexity regarding the self and others. In M. S. Clark 8 S. T. Fiske (Eds.). : r :.or . WM. Hillsdale, NJ: L. Erlbaum. 114 Logan. G. D. (1980). Attention and automaticity in Stroop and priming tasks: Theory and data. WM. 12. 523-553. Malamuth. N. 8 Donnerstein. E. (1982). The effects of aggressive-pomegraphic mass media stimuli. In L. Berkowitz (Ed.), WWW (Vol. 15). New York: Academic Press. Markus, H.. 8 Smith. J. (1981). The influence of self-schemata on the perception of others. In N. Cantor8 J. Kihlstrom(Eds.),Ea:59nalily_999nill9n_an9_5m|aLin19:am9n.Hillsdale, NJ: L. Erlbaum. Markus, H.. Smith, J., 8 Moreland, R. L. (1985). Role of the self-concept in the perception of others. W. 49. 1494-1512. Martin. L. L. (1986). Set/reset: Use and disuse of concepts in impression lonnation. J99:naL9[ W. .51. 493-504. Massad. C. M.. Hubbard. M., 8 Newtson. D. (1979). Selective perception of events. J99:naL91 ExnaflmantaLSociaLEsxcboloox. 15. 513-532. Matlin, M., 8 Stang. D. (1978). (999991. Cambridge. MA: Schenkman. McArthur, L. Z. (1975). The portrayal of men and women in American television commercials. .IoumaLoLSociaLEsxcbolocx.91 209-220. McNeil, J. C. (1975). Feminism. femininity. and the television series: A content analysis. J99:nal Wm. 19. 259-69. X Meyer. T. P. (1972). The effects of sexually arousing and violent films on aggressive behavior. WW 8. 324-333. Miloram. S. (1963). Behavioral study of obedience. W. 61. 371-378. 115 fis Mohr. D.. 8 Zanna. M. P. (1987). Treating women as sexual objects: Look to the (gender schematic) male who has viewed pomography. Unpublished manuscript University of Waterloo. xMorgan, M. (1982). Television and adolescents' sex role stereotypes: A longitudinal study. WW1. 43. 947-1062. Newtson. D. (1973). Attribution and the unit of perception of ongoing behavior. ,l99:nal_91 MW. 28. 28-38. Newtson. D. (1976). Foundations of attribution: The perception of ongoing behavior. In J. H. Harvey. W. J. lckes, 8 R. F. Kidd (Eds.). MW (Vol. 1). Hillsdale, NJ: L. Erlbaum. Newtson. D. (1980). An interactionist perspective on social knowing. Ea:59nality_and_§99ial EsmIJQImLBuIIetin. 6. 520-531. Newtson, 0.. 8 Engquist. G. (1976). The perceptual organization of ongoing behavior. J99:nal oLEmflmamaLSociaLEsxcbolocx. 12. 847-862. Newtson, 0.. Lightner, J., 8 Pennebaker. J. Effects of cognitive interference on the perception of ongoing behavior. Unpublished Manuscript. University of Virginia, 1980. Newtson. 0.. Rindner, R.. 8 Campbell. R. Observer skill and the perceptual organization of task performance. Unpublished manuscript, University of Virginia. 1979. Neisser, U. (1976). W. Englewood Cliffs. NJ: Prentice-Hall. O'Bryant. S., 8 Corder-Bolz, C. (1978). The effects of television on children's stereotyping of women's work roles. WM 12. 233-244. O'Leary, V. E. (1977). W. Monterey. CA: Brooks/Cole. Osgood. C. E.. Suci, G. J., 8 Tannenbaum, P. H. (1957). W. Urbana: University of Illinois Press. 116 Piliavin, J. A.. 8 Unger, R. K. (1985). The helpful but helpless female: Myth or reality? In V. E. O'Leary, R. K. Unger, 8 B. S. Wallston (Eds.).W9m9n._99n:l9:._an9_5991aL95an9l99y. Hillsdale. NJ: L. Erlbaum. Posner. M. I.. 8 Snyder. C. R. R. (1975). Attention and cognitive control. In R. L. Solso (Ed.), WWW. Hillsdale. N.J.: Lawrence Erlbaum Associates. Reitman, J. S. (1976). Skilled perception in GO: Deducing memory stnrctures from inter-response times. Q99nitiy9E5ycn9|99y 8. 336-356. Rogosa, D. (1979). Causal models in longitudinal research: Rationale. formulation. and interpretation. In J. R. Nesselroad, 8 P. B. Baltes (Eds.). WW9! 9Lbenayi9Lan999yel99m9nt. New York, NY: Academic Press. Rosenberg, S., 8 Sedlak, A. (1972). Stnrctural representations of implicit personality theory. In L. Berkowitz (Ed.), WWW (Vol. 6). New York: Academic Press. Rosene, J. M. (1971 ). The effects of violent and sexually arousing film content: An experimental study. Unpublished doctoral dissertation, Ohio University. 1971. As cited in Zillmann. 1984. Rossi. S. R.. 8 Rossi. J. S. (1985). Gender differences in the perception of women in magazine advertising. $92:_B9|9_5, 12. 1033-1038. Rothbart. M. (1981). Memory processes and social beliefs. In D. L. Hamilton (Ed.), 9999M WW. Hillsdale. NJ: L. Erlbaum. Rothbart, M., Evans. M.. 8 Fulero. S. (1979). Recall for confirming events: Memory processes and the maintenance of social stereotyping. J9umaL9LExpanm9n1aLS99iaL35yan9l99y, 15. 343-355. Schank. R. C. (1975). The structure of episodes in memory. In D. G. Bobrow. 8 A. Collins (Eds.). W New York: Academic Press. 117 Schank. R. C..8Abelson. R. P. (1977). W. Hillsdale. NJ: L. Erlbaum. Sears. D. O. (1976). WW5. Paper presented at annual meeting of the American Psychological Association, Washington. DC. Sentis. K. P., 8 Bumstein, E. (1979). Remembering schema-consistent information: Effects of a balance schema on recognition memory. .l9u:naL9l_E_9:59nality_an9_$9¢;iaLE5y9n9|9gy, 92. 2200-221 1 . Shotland. R. L.. 8 Straw. M. K. (1976). Bystander response to an assault: When a man attacks a woman. W 34. 990-999. Singer. J. L.. 8 Singer, D. G. (1979). Television viewing, family style. ang aggressive behavior in preschool children. In M. Green (Ed.), W. Washington. DC: American Association for the Advancement of Science. Singer. J. L.. 8 Singer. D. G. (1980). Television viewing and aggressive behavior in preschool children: afield study. W. 3:12. 289-303. Singer, J. L.. Singer. D. G.. 8 Rapaczynski. W. S. (1984). Family patterns and television viewing as predictors of children's beliefs and aggression. ,l9umaL9LQ9mmni9ali99, 34. 73-89. Sherif, M.. Taub, D.. 8 Hovland, C. I. (1958). Assimilation and contrast effects of anchoring stimuli in judgments. .l9_u:naL9l_9x99:imanlaL95yan9l9,gy. 55. 150-155. Shiffrin, R. M.. 8 Schneider, W. (1980). Controlled and automatic human information processing: ll. Perceptual Ieaming. automatic attending. and a general theory. W 94, 127-190.‘ Snyder. M.. 8 Uranowitz, S. W. (1978). Reconstructing the past: Some cognitive consequences 0' person perception. WNW 33.6. 941 -950. 118 Sprafkin. J. N.. 8 Liebert, R. M. (1978). Sex-typing and children's television preferences. In G. Tuchman. A. K. Daniels. 8 J. Benet (Eds.). W55 :nadja. New York: Oxford University Press. Snlll. T. K. (1981). Person memory: Some test of associative storage and retrieval models. WW 1. 440-462. Srull, T. K.. 8 Wyer. R. S. (1979). The role of category accessibility in the interpretation of information about persons: Some determinants and implications. J99:nal_9l_Ee:59nality_an9 5.9913135112119199!» 31. 1660-1672. Snlll. T. K.. 8 Wyer. R. S. (1980). Category accessibility and social perception: Some implications for the study of person memory and interpersonal judgment. J99:naL9( MW. 38. 841 -856. Stein. A. H.. 8 Friedrich. L. K. (1972). Television content and young children's behavior. In J. P. Murray, E. A. Rubinstein. 8 G. A. Comstock (Eds.).Ialayi5i9n_an9_5991al_99nayi9:._19L_ll; I9l9yi5i9n_an9_5991aLI9a:nin9. Washington. DO: U. S. Government Printing Office. Stemglanz. S. H.. 8 Serbin, L. A. (1974). Sex role stereotyping in children's television programs. WWW. 19. 710-715. Tabachnick. B. G.. 8 Fiddell, L. S. (1983). u5in9m9lliya:iala_5tati5ti95. New York: Harper8 Row. Taylor. S. E.. 8 Crocker. J. (1981). Schematic bases of social information processing. In E. T. Higgins, C. P. Herman. 8 M. P. Zanna (Eds.). WWW. Hillsdale. NJ: L. Erlbaum. Taylor. S. E., 8 Fiske, S. T. (1981). Getting inside the head: Methodologies for process analysis. In J. Harvey. W. lckes, 8 R. Kidd (Eds.). WWW (Vol. 3). Hillsdale. NJ: L. Erlbaum. Taylor. S. E.. Fiske. S. T.. Etcoff, N. L., 8 Rudennan. A. J. (1978). Categorical bases of person memory and stereotyping. W 36. 778-793. 119 Thomas. M. H.. Horton, R. W., Lippincott, E. C.. 8 Drabman. R. S. (1977). Desensitization to portrayals of real-life aggression as a function of exposure to television violence. J9u:naL91 MW. 35. 450-458. Tuchman. G. (1978). The symbolic annihilation of women. In G. Tuchman. A. K. Daniels. 8 J. Benet (Eds.). W. New York: Oxford University Press. Waite. B. M.. 8 Paludi. M. A. Sex-role stereotyping in popular music videos. Presented at the meeting of the Midwestem Psychological Association. Chicago. 1987. Weary, G.. Swanson. H.. Harvey. J. H.. 8 Yarkin, K. L. (1980). A molar approach to social knowing. WW 6. 574-581. Wegner, D. M.. Vallacher, R. R.. Macomber. 6.. Wood. R.. 8 Arps, K. (1984). The emergence of action. WWW. 45. 269-279. \Mlder, D. A. (1981. Perceiving persons as a group: Categorization and intergroup relations. In D. L. Hamilton (Ed). WWW Hillsdale. NJ: L. Erlbaum. Wlner. B. J. (1971). W. New York: McGraw-Hill. Wolf. T. M. (1973). Effects of live modeled sex-inappropriate play behavior in a naturalistic selling. WW. 9. 120-123. Wyer, R. 8.. Jr.. Srull. T. K.. 8 Gordon. S. (1984). The effects of predicting a person's behavior on subsequent trail ludcments. JoumaLoLExnarimantaLSociaLEsxcnclcox. 20. 29-46. Wyer. R. 8.. Jr.. Srull. T. K.. Gordon. S. E.. 8 Hartwick, J. (1982). Effects of processing oblectives on the recall of prose material. WWW. 43. 674-688. Zadny.J..8Gerard. H. B. (1974). Attributed intentions and informational selectivity. J99maL9l W. 1.0. 34-52. 120 Zajonc. R. B. (1968). Attitudinal effects of mere exposure. WW WW9. 1-27. Zillmann. D. (1984). Wm. Hillsdale. NJ: L. Erlbaum. Zillmann. D.. 8 Bryant. J. (1983). Effects of massive exposure to pornography. In N. M. Malamuth 8 E. Donnerstein (Eds.). EQm99:a9ny_an9_59xuaLa99:955i9n. New York: Academic Press. APPENDICES APPENDIX A Priming Videos in the Order in Which They Are Shown Witt 'Califomia Girls" by David Lee Roth "Out of Touch" (neutral) by Hall and Oates ”Heart 'n Soul“ by Huey Lewis and the News BQL-dummid ”Do We Diddy" by the Fools "Out of Touch” (neutral) by Hall and Oates "New Girl Now" by Honeymoon Suite Neutral "Faithfully" by Journey ”Out of Touch" by Hall and Oates ”You've Got to Know What Love Is” by Foreigner 121 APPENDIX 3.1 Interaction Script: "Getting Acquainted" Segment (Reciprocation) Parts of the Script Written to Contain the Manipulated Behaviors Insert One: Male: (Stares at woman. tracking with eyes. looking up and down, smiles.) ”Hello.” Female: (While walking on camera.) ”Hi." Male: (Continues to smile, maintains eye contact. reaches up to shake hand.) "Hi. are you here for the VJ job?“ (Holds her hand longer than is necessary.) Female: (Sitting in chair. maintaining eye contact and smiling when he holds her hand too long.) “Yeah. I guess we're going to be partners.” (Manipulation 1 - nonverbal 8 speech tone) Male: (As she sits. his eyes move down her body and back up. ) ”Great.” (In a subtly lecherous way.) Insert Two: Male: ”Oh (pause) Where do you live?" Female: "I just moved from Ypsilanti. and I got an apartment over on Walton." (Manipulation 2 -nonverbal) Male: "Do you live alone.“ (Again, lecherous voice. with eyebrows moving, turns more toward her while speaking.) Female: (Looking down his body.) 'I do right now. Do you?” (Manipulation 3 - nonverbal) Male: "I live with a guy who just got divorced, but he's out so much it's almost as if I live alone." (Again, lecherous voice. implying more than is said.) 122 Insert Three: Male: (Looking at her blouse.) "That's a nice blouse you're wearing.” Female: (Smiling at him.) "I'm glad you like it.” (Manipulation 4 - verbal 8 nonverbal) Male: (Same as last. feeling the material of the blouse. just slightly above the waist. Removes hand after saying ) ”Nice material. Really soft and sexy.” Female: (Watching his hand as he does this then looking into his face.) "It feels nice to wear it.” (Manipulation 5 - verbal 8 nonverbal) Insert Four: Female: (Crossing her arms tightly across her chest hugging herself--exaggerating movement.) ”It's kind of cold in here. I wish they'd turn up the heat." Male: (Looking down across the top of her folded arms.) "Do you get cold easily?” Female: ”Not too easily.” (Looks down his body.) (Manipulation 6 - nonverbal) Male: ”Well. if I knew where your (out off in mid word), ah. the thermostat was. I'd turn it up for you." Female: ”I'd like that.” (Manipulation 7 - verbal) Insert Five: Male: ”I'm really excited about getting this job. (Say next sentence slowly with emphasis.) If we do get it. we'd have to spend a lot of time together. and we'd probably get to go to some concerts." Female: ”I'd love it; it'd be great. I feel like I know you better than some of the guys I've dated.” (88 inviting.) (Manipulation 8 - verbal 8 nonverbal) Male: “You'll get to know me better." (Again the lecherous voice.) 123 APPENDIX 8.2 Interaction Script: ”Getting Acquainted” Segment (Nonreciprocation) Parts of the Script Written to Contain the Manipulated Behaviors Insert One: Male: (Stares at woman, tracking with eyes, looking up and down. smiles.) ”Hello.” Female: (While walking on camera.) "Hi." Male: (Continues to smile, maintains eye contact, reaches up to shake hand.) ”Hi. are you here for the VJ job?" (Holds her hand longer than is necessary.) Female: (Pulls hand away, nanows eyes as if she doesn't like it. Then without enthusiasm. while directing her attention to the TV monitor.) "Yeah, I guess we're going to be partners.” (Manipulation 1 - nonverbal 8 speech tone) Male: (As she sits. his eyes move down her body and back up.) “Great." (In a subtly lecherous way.) Insert Two: Male: “Well, (pause) where do you live?” Female: (Shrugs his hand off her as she leans away with mild disgust on her face. ) "I just moved from Ypsilanti over to an apartment over on Walton.“ (Manipulation 2 - nonverbal) Male: "Really. Do you live alone." (Again. lecherous voice. with eyebrows moving. turns more toward her while speaking.) Female: (Briefly shift eyes from TV to him as she says ) "Yeah, well I do right now. How about you?” (Friendly but no sexual interest at all.) (Manipulation 3 - nonverbal 8 speech tone) Male: "I live with a guy who just got divorced, but he's out so much it's like living alone.” (Again. lecherous voice. implying more than is said.) 124 Insert Three: Male: (Looking at her blouse.) "That's a nice blouse you're wearing." Female: (Smiling at him slightly.) "Thank you.“ (Manipulation 4 - verbal 8 nonverbal) Male: (Same as last, feeling the material of the blouse, just slightly above the waist.) Removes hand after saying "Nice material. Really soft and kind of sexy." Female: (Pulls away from his touch and saying in a not too harsh way. Look at place touched while delivering line.) "It gets wrinkled when you touch it, sometimes.“ (Manipulation 5 - verbal 8 nonverbal) Insert Four: Female: (Crossing her arms tightly across her chest hugging herself.) “It's kind of cold in here. I wish they'd turn up the heat.” Male: (Looking down across the top of her folded arms.) ”Do you get cold easily?” Female: "Well. Not too easily.“ (Keep arms in position until beyond and of insert. Look around the room as if to find a thermostat.) (Manipulation 6 - nonverbal) Male: "If I knew where your (cutoff in mid word), ah. the thermostat was. I'd turn it up for you.“ Female: “That's OK. I'll be fine“ (Again not too unfriendly. Ignore innuendo. but not nasty. ) (Manipulation 7 - verbal) 125 Insert Five: Male: "I'm really excited about getting this job. (Say next sentence slowly with emphasis.) If we get it. we'd have to spend a lot of time together. and we'd probably get to go to some concerts.” Female: "I suppose so. but I don't feel like I know you that well yet.” (Be friendly and cautious but not inviting.) (Manipulation 8 - verbal 8 nonverbal) Male: "You'll get to know me better.” (Again the lecherous voice.) 126 APPENDIX 3.3 Interaction Script: “Audition" Segment (Praise) Complete Audition Script Female: Without a doubt. Bruce is one of America's hottest performers. Male: You're right, looking back on his 84-85 concert tour. it's true; he really appeals to a wide range of people. (Manipulation 1 - verbal) Female: For a live stage performance. this is definitely one of his better produced videos. Male: That's an excellent point. Some people felt that the bad lip sync was distracting. but. small technicalities aside, it was one of the great performances. Super video. (Manipulation 2 - verbal 8 speech tone) Female: Right. it's what you hear that's important. not what you see. Male: Boy, I love the way you go to the heart of it. (Manipulation 3 - verbal 8 speech tone) Female: Well, he sure feels the pulse of the country: The video portrays hard-working, middle-class America as Bruce sees it. A powerful point-of-view direct from his upbringing. And he sure anticipated the new mood of patriotism. Male: Well. that's an interesting observation. I hadn't thought of that. You have a lot of good insight. (Manipulation 4 - verbal 8 speech tone) Female: Bruce reveals a lot about himself in this video. The Vietnam theme surfaced because he had a brother who was missing in action there. Male: Yeah, they publicized that aspect quite heavily when they released the video, but you're right to remind the audience of that because it makes the video that much more personal. (Manipulation 5 - verbal) 127 Female: Well. there are lots of different things to like about the Boss in this video. Male: That's for sure. And, like you. he puts it all together. (Manipulation 6 - verbal 8 speech tone) Female: OK. We'll be back after the next video with some concert news of local interest. 128 APPENDIX 8.4 Interaction Script: “Audition” Segment (Derogation) Complete Audition Script Female: Without a doubt. Bruce is one of America's hottest performers. Male: Well. looking back on his 84-85 concert tour. it's obvious that he really appeals to a wide range of people. (Manipulation 1 - verbal) Female: For a live stage performance. this is definitely one of his better produced videos. Male: Well I don't know about that. A lot of people felt that the bad lip sync was distracting, but I guess you'd have to look at it more closely than you did to really evaluate it. But. technicalities aside. that was one of the great performances. Super video. (Manipulation 2 - verbal 8 speech tone) Female: Right. it's what you hear that's important. not what you see. Male: Sure. and I suppose when people go to concerts they keep their eyes closed. (Manipulation 3 - verbal 8 speech tone) Female: Well. he sure feels the pulse of the country: The video portrays hard-working. middle-class America as Bruce sees it. A powerful point-of-view direct from his upbringing. And he sure anticipated the new mood of patriotism. Male: Well, that goes without saying. I'm sure that won't be news to our audience. (Manipulation 4 - verbal 8 speech tone) Female: He sure reveals a lot about himself in this video. He had a brother who was missing in action in Vietnam. Male: Yeah. they publicized that aspect heavily when they released the video but it really only makes a small contribution compared to the tremendous impact of his energy in the video. (Manipulation 5 - verbal) 129 Female: Well, there are lots of different things to like about the Boss in this video. Male: That's for sure. But. the point is that he puts it all together. (Manipulation 6 - verbal 8 nonverbal) Female: OK. After this next video. we'll be back with some concert news of local interest. 130 APPENDIX C.1 Prime and Script Stories READ AND CONSIDER THE FOLLOWING THREE THEMES OR STORIES CAREFULLY THEME 1: Boy sees girl. Boy and girl are both attractive. Finally they are able to get together. They live happily ever after. THEME 2: Boy wants beautiful girl. Boy gets girl. Boy finds girl not so terrific. Boy looks for another girl. THEME 3: Musicians are lonely. They are trying to get a message across in their work. Love and family are things that they want. but are very difficult to maintain because of their lifestyle. HOW COMPATIBLE IS THE STORY BELOW WITH EACH OF THE THEMES ABOVE? IN OTHER WORDS. TO WHAT EXTENT IS THE STORY BELOW A VARIATION ON EACH OF THE THEMES ABOVE? INDICATE YOUR ANSWERS BY CIRCUNG A NUMBER ON EACH OF THE SCALES BELOW. STORY: Boy wants girl. He tries by talking and complimenting her. She responds, but she doesn't know if he's putting her on or if he's serious. She tests him. She finds him OK. They go out together. To what extent is this story a variation on theme one? That is. how compatible is this story with theme 1? +2 Story is very compatible with theme onefit is almost the same as theme one +1 Story is compatble with theme oneflt is a variation on theme one 0 Story seems relatively unrelated to theme one -1 Story is not compatible with theme one/rt is not a variation on theme one -2 Story is very incompatible with theme one/rt seems different from the theme 131 To what extent is this story a variation on theme two? That is. how compatible is this story with theme 2? +2 Story is very compatible with theme twofit is almost the same as theme two +1 Story is compatible with theme twofit is a variation on theme two 0 Story seems relatively unrelated to theme two -1 Story is not compatible with theme two/rt is not a variation on theme two -2 Story is very incompatible with theme twofit seems different from the theme To what extent is this story a variation on theme three? That is. how compatible is this story with theme 3? +2 Story is very compatible with theme threefit is almost the same as theme three +1 Story is compatible with theme three/rt is a variation on theme three 0 Story seems relatively unrelated to theme three -1 Story is not compatible with theme threefit is not a variation on theme three -2 Story is very incompatible with theme three/it seems different from the theme 132 APPENDIX 02 Prime and Script Stories READ AND CONSIDER THE FOLLOWING THREE THEMES OR STORIES CAREFULLY THEME 1: Boy sees girl. Boy and girl are both attractive. Finally they are able to get together. They live happily ever after. THEME 2: Boy wants beautiful girl. Boy gets girl. Boy finds girl not so terrific. Boy looks for another girl. THEME 3: Musicians are lonely. They are trying to get a message across in their work. Love and family are things that they want. but are very difficult to maintain because of their lifestyle. HOW COMPATIBLE IS THE STORY BELOW WITH EACH OF THE THEMES ABOVE? IN OTHER WORDS. TO WHAT EXTENT IS THE STORY BELOW A VARIATION ON EACH OF THE THEMES ABOVE? INDICATE YOUR ANSWERS BY CIRCUNG A NUMBER ON EACH OF THE SCALES BELOW. STORY: Boy meets girl. Boy hits on girl. Girl rejects boy. Boy is crushed. Then boy gets angry. Boy tries to get back at girl by making her look foolish. To what extent is this story a variation on theme one? That is. how compatible is this story with theme 1? +2 Story is very compatible with theme oneflt is almost the same as theme one +1 Story is compatible with theme onefit is a variation on theme one 0 Story seems relatively unrelated to theme one -1 Story is not compatible with theme one/it is not a variation on theme one -2 Story is very incompatible with theme onefrt seems different from the theme To what extent is this story a variation on theme two? That is. how compatible is this story with theme 2? +2 Story is very compatible with theme twofit is almost the same as theme two +1 Story is compatible with theme twofit is a variation on theme two 0 Story seems relatively unrelated to theme two -1 Story is not compatible with theme two/rt is not a variation on theme two -2 Story is very incompatible with theme twofit seems different from the theme 133 To what extent is this story a variation on theme three? That is, how compatible is this story with theme 3? +2 Story is very compatible with theme three/it is almost the same as theme three +1 Story is compatible with theme three/it is a variation on theme three 0 Story seems relatively unrelated to theme three -1 Story is not compatible with theme threeflt is not a variation on theme three -2 Story is very incompatible with theme threefit seems different from the theme 134 APPENDIX C.3 Prime and Script Stories READ AND CONSIDER THE FOILOWING THREE THEMES OR STORIES CAREFUIJ.Y THEME 1: Boy sees attractive girl all alone. Girl sees boy and they become mutually attracted. Some obstacles occur but both overcome these and they live happily ever after. THEME 2: Boy meets girl of his dreams. Boy and girl eventually marry. After marriage. boy finds girl is a complainer and grows tired of girl. Boy finds new girl and wants to dispose of first girl. THEME 3: On the outside things may look glamorous, but underneath these things are filled with loneliness and isolation. HOW COMPATIBLE IS THE STORY BELOW WITH EACH OF THE THEMES ABOVE? IN OTHER WORDS, TO WHAT EXTENT IS THE STORY BELOW A VARIATION ON EACH OF THE THEMES ABOVE? INDICATE YOUR ANSWERS BY CIRCUNG A NUMBER ON EACH OF THE SCALES BELOW. STORY: Boy meets girl. Boy flatters girl. Girl stays submissive. Boy continues flattery. Girl accepts and is interested in boy. To what extent is this story a variation on theme one? That is. how compatible is this story with theme 1? +2 Story is very compatible with theme one/it is almost the same as theme one +1 Story is compatible with theme one/rt Is a variation on theme one 0 Story seems relatively unrelated to theme one -1 Story is not compatible with theme one/rt is not a variation on theme one -2 Story is very incompatible with theme one/rt seems different from the theme To what extent is this story a variation on theme two? That is, how compatible is this story with theme 2? +2 Story is very compatmle with theme twoflt is almost the same as theme two +1 Story is compatible with theme twofit is a variation on theme two 0 Story seems relatively unrelated to theme two -1 Story is not compatible with theme twofit is not a variation on theme two -2 Story is very incompatible with theme twofit seems different from the theme 135 To what extent is this story a variation on theme three? That is. how compatible is this story with theme 3? +2 Story is very compatfiale with theme threefit is almost the same as theme three +1 Story is compatible with theme three/rt is a variation on theme three 0 Story seems relatively unrelated to theme three -1 Story is not compatible with theme three/it is not a variation on theme three -2 Story is very incompatible with theme three/rt seems different from the theme 136 APPENDIX C.4 Prime and Script Stories READ AND CONSIDER THE FOILOWING THREE THEMES OR STORIES CAREFULLY THEME 1: Boy sees attractive girl all alone. Girl sees boy and they become mutually attracted. Some obstacles occur but both overcome these and they live happily ever after. THEME 2: Boy meets girl of his dreams. Boy finds girl is a complainer and grows tired of girl. Boy finds new girl and wants to dispose of first girl. THEME 3: On the outside things may look glamorous. but underneath these things are filled with loneliness and isolation. CIRCUNG A NUMBER ON EACH OF THE SCALES BELOW. HOW COMPATIBLE IS THE STORY BELOW WITH EACH OF THE THEMES ABOVE? IN OTHER WORDS. TO WHAT EXTENT IS THE STORY BELOW A VARIATION ON EACH OF THE THEMES ABOVE? INDICATE YOUR ANSWERS BY CIRCUNG A NUMBER ON EACH OF THE SCALES BELOW. STORY: Boy meets girl. whereupon he makes advances toward her. She refuses. Boy feels offended and tries to make the girl look foolish. To what extent is this story a variation on theme one? That is. how compatible is this story with theme 1? +2 Story is very compatible with theme one/it is almost the same as theme one +1 Story is compatible with theme one/it is a variation on theme one 0 Story seems relatively unrelated to theme one -1 Story is not compatible with theme oneflt is not a variation on theme one -2 Story is very incompatible with theme one/it seems different from the theme To what extent is this story a variation on theme two? That is. how compatible is this story with theme 2? +2 Story is very compatible with theme two/rt is almost the same as theme two +1 Story is compatible with theme two/it is a variation on theme two 0 Story seems relatively unrelated to theme two -1 Story is not compatible with theme twofit is not a variation on theme two -2 Story is very incompatible with theme twofit seems different from the theme 137 To what extent is this story a variation on theme three? That is. how compatible is this story with theme 3? +2 Story is very compatible with theme three/it is almost the same as theme three +1 Story is compatible with theme three/It is a variation on theme three 0 Story seems relatively unrelated to theme three -1 Story is not compatible with theme three/it is not a variation on theme three -2 Story is very incompatible with theme three/rt seems different from the theme 138 APPENDIX 05 READ AND CONSIDER THE FOLLOWING THREE THEMES OR STORIES CAREFULLY THEME 1: Boy sees girl. Girl sees boy. They are attracted to one another. They go off together. THEME 2: Boy meets girl. Girl is attractive. But, boy sees another girl who he likes more. Boy leaves first girl. THEME 3: Boy is on the road away from family. He is lonely and misses them. He is reflecting on his loneliness while travelling and doing his job. He questions whether it is worth doing the job. HOW COMPATIBLE IS THE STORY BELOW WITH EACH OF THE THEMES ABOVE? IN OTHER WORDS. TO WHAT EXTENT IS THE STORY BELOW A VARIATION ON EACH OF THE THEMES ABOVE? INDICATE YOUR ANSWERS BY CIRCUNG A NUMBER ON EACH OF THE SCALES BELOW. STORY: Boy meets girl. Boy tries to pick up girl with compliments. Girl becomes interested. To what extent is this story a variation on theme one? That is. how compatible is this story with theme 1? +2 Story is very compatible with theme onefit is almost the same as theme one +1 Story is compatible with theme one/it is a variation on theme one 0 Story seems relatively unrelated to theme one -1 Story is not compatible with theme one/it is not a variation on theme one -2 Story is very incompatible with theme onefit seems different from the theme To what extent is this story a variation on theme two? That is. how compatible is this story with theme 2? +2 Story is very compatible with theme twofit is almost the same as theme two +1 Story is compatible with theme twofit is a variation on theme two 0 Story seems relatively unrelated to theme two -1 Story is not compatible with theme twofit is not a variation on theme two -2 Story is very incompatible with theme twofit seems different from the theme 139 To what extent is this story a variation on theme three? That is, how compatible is this story with theme 3? +2 Story is very compatible with theme three/it is almost the same as theme three +1 Story is compatible with theme three/it is a variation on theme three 0 Story seems relatively unrelated to theme three -1 Story is not compatible with theme threefrt is not a variation on theme three -2 Story is very incompatible with theme three/it seems different from the theme 140 APPENDIX C.6 Prime and Script Stories READ AND CONSIDER THE FOLLOWING THREE THEMES OR STORIES CAREFULLY THEME 1: Boy sees girl. Girl sees boy. They are attracted to one another. They go off together. THEME 2: Boy meets girl. Girl is attractive. But. boy sees another girl who he likes more. Boy leaves first girl. THEME 3: Boy is on the road away from family. He is lonely and misses them. He is reflecting on his loneliness while travelling and doing his job. He questions whether it is worth doing the job. HOW COMPATIBLE IS THE STORY BELOW WITH EACH OF THE THEMES ABOVE? IN OTHER WORDS, TO WHAT EXTENT IS THE STORY BELOW A VARIATION ON EACH OF THE THEMES ABOVE? INDICATE YOUR ANSWERS BY CIRCUNG A NUMBER ON EACH OF THE SCALES BELOW. STORY: Boy meets girl. He hits on her. She rejects him. His is critical of her and cuts her down. To what extent is this story a variation on theme one? That is. how compatible is this story with theme 1? +2 Story is very compatible with theme onefit is almost the same as theme one +1 Story is compatible with theme one/it is a variation on theme one 0 Story seems relatively unrelated to theme one -1 Story is not compatible with theme onefit is not a variation on theme one -2 Story is very incompatible with theme one/rt seems different from the theme To what extent is this story a variation on theme two? That is. how compatible is this story with theme 2? +2 Story is very compatible with theme two/rt is almost the same as theme two +1 Story is compatible with theme twofit is a variation on theme two 0 Story seems relatively unrelated to theme two -1 Story is not compatible with theme twofit is not a variation on theme two -2 Story is very incompatible with theme twofit seems different from the theme 141 To what extent is this story a variation on theme three? That is. how compatible is this story with theme 3? +2 Story is very compatible with theme threefit is almost the same as theme three +1 Story is compatible with theme three/it is a variation on theme three 0 Story seems relatively unrelated to theme three -1 Story is not compatible with theme threefit is not a variation on theme three -2 Story is very incompatible with theme three/it seems different from the theme 142 APPENDIX C.7 Prime and Script Stories READ AND CONSIDER THE FOLLOWING THREE THEMES OR STORIES CAREFULLY THEME 1: Boy sees girl. Girl sees boy. They are attracted to each other and eventually get together. They fall in love and go off together. THEME 2: Boy meets girl. Boy pursues girl and they eventually fall in love. Boy becomes bored with girl. Girl becomes a complainer to boy. Boy meets new girl. THEME 3: Musicians (groups/bands) perform in front of large audiences and tour. They get lonely and miss people they love. The lifestyle can be glamorous but has disadvantages. HOW COMPATIBLE IS THE STORY BELOW WTTH EACH OF THE THEMES ABOVE? IN OTHER WORDS, TO WHAT EXTENT IS THE STORY BELOW A VARIATION ON EACH OF THE THEMES ABOVE? INDICATE YOUR ANSWERS BY CIRCUNG A NUMBER ON EACH OF THE SCALES BELOW. STORY: Boy meets girl. Boy gives attention to girl. Girl acts shy and naive. Girl becomes more confident. Boy gives her more attention and treats her on a more equal level. To what extent is this story a variation on theme one? That is. how compatible is this story with theme 1? +2 Story is very compatible with theme one/it is almost the same as theme one +1 Story is compatible with theme one/it is a variation on theme one 0 Story seems relatively unrelated to theme one -1 Story is not compatible with theme one/it is not a variation on theme one -2 Story is very incompatible with theme one/it seems different from the theme To what extent is this story a variation on theme two? That is. how compatible is this story with theme 2? +2 Story is very compatible with theme twofit is almost the same as theme two +1 Story is compatible with theme two/it is a variation on theme two 0 Story seems relatively unrelated to theme two -1 Story is not compatible with theme two/it is not a variation on theme two -2 Story is very incompatible with theme twofit seems different from the theme 143 To what extent is this story a variation on theme three? That is, how compatible is this story with theme 3? +2 Story is very compatible with theme threefit is almost the same as theme three +1 Story is compatible with theme threefit is a variation on theme three 0 Story seems relatively unrelated to theme three -1 Story is not compatible with theme threefit is not a variation on theme three -2 Story is very incompatible with theme three/it seems different from the theme 144 APPENDIX C.8 Prime and Script Stories READ AND CONSIDER THE FOLLOWING THREE THEMES OR STORIES CAREFULLY THEME 1: Boy sees girl. Girl sees boy. They are attracted to each other and eventually get together. They fall in love and go off together. THEME 2: Boy meets girl. Boy pursues girl and they eventually fall in love. Boy becomes bored with girl. Girl becomes a complainer to boy. Boy meets new girl. THEME 3: Musicians (groups/bands) perform in front of large audiences and tour. They get lonely and miss people they love. The lifestyle can be glamorous but has disadvantages. HOW COMPATIBLE IS THE STORY BELOW WITH EACH OF THE THEMES ABOVE? IN OTHER WORDS, TO WHAT EXTENT IS THE STORY BELOW A VARIATION ON EACH OF THE THEMES ABOVE? INDICATE YOUR ANSWERS BY CIRCUNG A NUMBER ON EACH OF THE SCALES BELOW. STORY: Boy makes advances toward girl. Girl is not interested and rejects the boy. Boy reacts by trying to hurt girl and make her look foolish. To what extent is this story a variation on theme one? That is, how compatible is this story with theme 1? +2 Story is very compatible with theme onefit is almost the same as theme one +1 Story is compatible with theme one/it is a variation on theme one 0 Story seems relatively unrelated to theme one -1 Story is not compatible with theme onefit is not a variation on theme one -2 Story is very incompatible with theme oneflt seems different from the theme To what extent is this story a variation on theme two? That is, how compatible is this story with theme 2? +2 Story is very compatible with theme two/it is almost the same as theme two +1 Story is compatible with theme IWO/It is a variation on theme two 0 Story seems relatively unrelated to theme two -1 Story is not compatible with theme twofit is not a variation on theme two -2 Story is very incompatible with theme twofit seems different from the theme 145 To what extent is this story a variation on theme three? That is, how compatible is this story with theme 3? +2 Story is very compatible with theme threefit is almost the same as theme three +1 Story is compatible with theme threefit is a variation on theme three 0 Story seems relatively unrelated to theme three -1 Story is not compatible with theme three/it is not a variation on theme three -2 Story is very incompatible with theme three/it seems different from the theme 146 APPENDIX D Minimally Structured Recall Measure Please list all the separate behaviors or actions you can remember. Put each separate behavior on a single line. List the behaviors or actions in the order in which they come to your mind. In the Person column, write in who performed the behavior or action, and in the Behavior/Action column, write in what the person did. List as many behaviors as you can. Be sure to write them down in the order that you think about them. List only one behavior or action per line. EEBSQNEEHAMIQBZAQIIQN USE THE NEXT SHEET IF NEEDED 147 (page 2) EEBSQN BEHAMIQBLAQIIQN 148 APPENDIX E.1 Trait Scales (female) MUSIC VIDEO EXPERIMENT - IMPRESSION SCALE We are interested in your overall impression of the female VJ candidate. Indicate your impression of the female VJ by placing a check on each scale at the point describing your impression of her. Make sure you place a check (‘1) on each line. unappealing _________ appealing weak _________ strong passive _________ active emotional _________ unemotional intelligent _________ unintelligent nonassertive _________ assertive nice _________ nasty threatening _________ nonthreatening pleasant _________ unpleasant skilled _________ unskilled nonaggressive _________ aggressive shy _________ outgoing attractive _________ unattractive saphisticated _________ naive dull _________ clever calm _________ nervous clumsy _________ graceful sensitive _________ insensitive competent _________ incompetent extraverted _________ introverted sexual _________ nonsexual talkative _________ quiet honest _________ dishonest colorful _________ colorless antisocial _________ prosocial dominant _________ submissive friendly _________ unfriendly sympathetic _________ hostile negative _________ positive happy _________ sad masculine _________ feminine cold _________ warm dependent _________ independent 149 APPENDIX E.2 Trait Scales (male) MUSIC VIDEO EXPERIMENT - IMPRESSION SCALE We are interested in your overall impression of the male VJ candidate. Indicate your impression of the male VJ by placing a check on each scale at the point describing your impression of him. Make sure you place a check (‘1) on each line. unappealing _________ appealing weak _________ strong passive _________ active emotional _________ unemotional intelligent _________ unintelligent nonassertive _________ assertive nice _________ nasty threatening _________ nonthreatening pleasant _________ unpleasant skilled _________ unskilled nonaggressive _________ aggressive shy _________ outgoing attractive _________ unattractive sophisticated _________ naive dull _________ clever calm _________ nervous clumsy _________ graceful sensitive _________ insensitive competent _________ incompetent extraverted _________ introverted sexual _________ nonsexual talkative _________ quiet honest _________ dishonest colorful _________ colorless antisocial _________ prosocial dominant _________ submissive friendly _________ unfriendly sympathetic _________ hostile negative _________ positive happy _________ sad masculine _________ feminine cold _________ warm dependent _________ independent 150 APPENDIX F.1 Impression Favorability Scales (order 1) MUSIC VIDEO 9(PERIMENT - GENERAL EVALUATION (1) Based on your observations of the getting acquainted period and predicting how people will react when they meet these VJs in person, (a) should the male be considered for the VJ job? definitely no 0 1 2 3 4 5 6 7 8 9 10 definitely yes (b) should the female be considered for the VJ job? definitely no 0 1 2 3 4 5 6 7 8 9 10 definitely yes (2) Based on your observations of the audition and predicting how people will react when the see the VJs on television, (a) should the male be considered for the VJ job? definitely no 0 1 2 3 4 5 6 7 8 9 1O definitely yes (b) should the female be considered for the VJ job? definitely no 0 1 2 3 4 5 6 7 8 9 1O definitely yes (3) Based on your observations of both the getting acquainted period and the audition, what is your own personal reaction to the two candidates? (a) The male seems like someone I would like personally. definitely no 0 1 2 3 4 5 6 7 8 9 10 definitely yes (b) The female seems like someone I would like personally. definitelyno O 1 2 3 4 5 6 7 8 9 10 definitely yes 151 APPENDIX F.2 Impression Favorability Scales (order 2) MUSIC VIDEO BfPERIMENT - GENERAL EVALUA TIOIV (1) Based on your observations of the getting acquainted period and predicting how people will react when they meet these VJs in person, (a) should the female be considered for the VJ job? definitely no 0 1 2 3 4 5 6 7 8 9 1O definitely yes (b) should the male be considered for the VJ job? definitely no 0 1 2 3 4 5 6 7 8 9 10 definitely yes (2) Based on your observations of the audition and predicting how people will react when the see the VJs on television, (a) should the female be considered for the VJ job? definitely no 0 1 2 3 4 5 6 7 8 9 10 definitely yes (b) should the male be considered for the VJ job? definitely no 0 1 2 3 4 5 6 7 8 9 10 definitely yes (3) Based on your observations of both the getting acquainted period and the audition, what is your own personal reaction to the two candidates? (a) The female seems like someone I would like personally. definitely no 0 1 2 3 4 5 6 7 8 9 10 definitely yes (b) The male seems like someone I would like personally. definitely no 0 1 2 3 4 5 6 7 8 9 10 definitely yes 152 APPENDIX G.1 Structured Recall Measure: “Getting Acquainted" Segment We want to find out how memorable the speaking style of the two candidates is. Try to recall exaQIIlehat the female candidate said to the male after he made each of the following statements while they were getting acquainted. Write down exactly what you remember her saying in the spaces below. 1. MALE: "That's a nice blouse you're wearing." FEMALE: 2. MALE: "Nice material, too. Really soft and sort of sexy.” FEMALE: 3. MALE: "If I knew where your...the...thermostat was, I'd turn it up for you.” FEMALE: 153 4. MALE: “I'm really excited about getting this job. If we get it, we'd have to spend a lot of time together.” FEMALE: 154 APPENDIX 62 Structured Recall Measure: “Audition“ Segment Now, try to recall exactly what the male candidate said to the female during the audition. Write down exactly what he said to her after she made the statements below. 1. FEMALE: "Without a doubt, Bruce is one of America's hottest performers.“ MALE: 2. FEMALE: “You know, for a live stage performance, this is definitely one of the better produced videos." MALE: 3. FEMALE: "Rightl It's what you hear that's important-- not what you see.” MALE: 4. FEMALE: "The video protrays hard-working, middle-class America as Bruce sees it--a powerful point-of-view direct from his upbringing. And he sure anticipated the new mood of patriotism.” MALE: 155 5. FEMALE: “Bruce really reveals a lot about himself in this video. The Vietnam theme surfaced because he had a brother who was missing in action there.” 6. FEMALE: ”Well, there are lots of different things to like about the Boss in this video." MALE: 156 APPENDIX H Cover Story and Instructions Welcome to the Music Video Experiment, a joint project of researchers in the Departments of Psychology at Oakland University and Michigan State University. Please listen carefully as we explain the project and tell you how what you will be doing fits into the research. Thank you in advance for your attention. Pause The goal of the project is to explore a problem of community access programming on cable television circuits. In many communities, cable companies are required to provide local access to a cable channel for the presentation of locally produced programs of community interest. It was hoped that the programming on the channel could be integrated into community life--giving people a sense of community involvement and belonging--and could extend education from local classrooms right into the home. But, it hasn't worked out that way. Very frequently, these local access channels are underused and generate very little community interest. Part of this problem, we believe, results from poor or boring productions. Certainly, relative to cable channels providing well produced programming such as MTV and HBO, local programming has little chance. But, another approach to increasing the appeal of local cable programs is available and it is the purpose of this project to investigate that approach. Pause One big advantage that local programming could have over national programming is the availability of local television personalities to the public. Research has shown that local television personalities--almost always local newscasters--have tremendous appeal, in part, because the audience feels that they can identify with the person. In short, when people meet and feel that they know the personalities they see on television they are much more likely to tune in and to become involved in the programming. This is the approach that the project is designed to test. Let me tell you how it's going to work. Pause In a later stage in the project, a weekly television program that we produce will be broadcast on a number of local access channels in southern Michigan on the same night for five consecutive weeks. The program will be a simuluated MTV-format show in which two amateur videojockeys will show rock music videos at the top of the local music charts and offer commentary and information about local music events. A large number of people in the 17 to 25 year old age range from households receiving the cable signal have been identified to participate in this project. People of this age were targeted for this research for two reasons. First, people in this age range constitute a large portion of the viewing audience. Second, when we surveyed a large number of 17 to 25 year olds, we found that music video programs are very popular. So the people who take part in the research will be familiar and comfortable with the MTV-type format of the simulated tv program. Pause Everyone participating in the project will watch the program in their home. Every viewer will fill out a questionnaire each week after watching the program. Each person will be asked what they liked and didn't like and to make requests for the following week.'s program. The idea is to get each viewer involved in the program and to feel that they are a part of the program. About 30% of the viewers participating in this research will become involved in the program in an additional way. These people will meet the two VJs. These people have volunteered to participate in a series of 157 person-to-person contact social gatherings scheduled to meet in local schools in the areas that will be receiving the program on their cables. These people will get to talk to the television personalities face to face and to offer their suggestions and comments. We are particularly interested in these people's viewing habits and feelings about the program after they have actually participated in a face to face interaction with the VJs. We want to see how involving people in local programming influences their feelings about it and the likelihood of their watching it. Pause Now that l have told you something of the purpose of the project, let me explain your task today. Currrently, we are at the point of selecting the people who will be trained to play the roles of these television personalities. You are going to be seeing two of the 16 candidates who auditioned last year for the roles. We will be asking you for your opinions about them. The pe0ple we eventually select to play the roles of show hosts will be trained to act as a pair of video jockeys on a simulated music video program. What we would like you to do is help us in this selection process. Pause Because the part that these two people play is critical to the research, we want to choose them carefully and this requires a rather complicated procedure for making the choice. The first thing we did was advertise for volunteers to make audition tapes for the role of "guest video jockey” in a simulated MTV-program. They were told that they would be paid $10 for making the tape. From the list of names of the people who phoned in answer to our ad, we scheduled one woman and one man to audition at the same time. None of the people knew their partners before they arrived. When they arrived at the studio, and met one another for the first time, they were given some time to get to know one another. After this time to get to know each other and to practice, they made the audition tape. Pause You will see an audition tape made by two of the volunteers. But first you will see something else. Although they thought the studio cameras and microphones were turned off during the getting acquainted session, we actually had the cameras and mikes on so that what they did and said was recorded. Remember that these people will be required to interact with the audience at social gatherings in the community during the actual research, so we used this "candid camera” technique to get a sense of how these people acted offcamera. Pause So first you are going to see them in this getting acquainted period when they think they can't be seen or heard. During this part of the tape, you will see the two VJs as well as what they are watching-the music video they would be commenting on when making their audition tape. Then, you will see the actual audition tape that these people made and hear their comments on the video. We would like you to help us by doing two things. First, just form a general impression of the two people. Don't worry too much about how well they perform as video jockeys this first time out. Remember they will be trained for the parts. It's more important that you just get a sense of how much you like them or not and how much you think the 17 to 25 year olds in the final project would like to watch them on the simulated television program. The second thing we want you to help us with is to establish the pace or speed at which these two candidates act. In order to do this, we want you to "chop up" the actions and speech of the candidates during both the "getting acquainted” and "audition" segments into pieces or units. Each time you chop the tape you will be indicating that something has been completed and something new is beginning. Let me explain this. Pause As you are watching and listening to the candidates, what we would like you to do is press the button in front of you whenever you see what to you is a meaningful unit of 158 action. For example, if you see a person turn their head, look at someone, raise their arm, wave, bring their arm back down, and turn their head back, you can say that you saw six meaningful actions (head turn, eye movement, raised arm, waving, moving arm, head turn) or one meaningful action (waving at someone). Or suppose a person is talking. You might consider several sentences in a row to be a meaningful action, or each sentence, or even each separate thought. Any number of meaningful actions you decide on is the answer we want. There are no right or wrong answers. Just press the button whenever you feel that you have seen a meaningful unit of action. Remember that you will be pushing the button in front of you every time you see the candidates perform a meaningful unit of behavior. I would like you to get the feel of it, so imagine that you just observed the candidates performing a meaningful behavior. Push the button firmly--and release. The light should go on while the button is pressed and go off when you take your finger off the button. Do not begin pushing the button to measure what you feel are meaningful units of action until the tape begins in which the two candidates are getting acquainted. Pause Ok, we are just about ready to roll the tape. To make the presentation more like what the viewers will eventually see on their cable channel at home and to familiarize you with the MTV-type format, just in case you don't know it, we begin the tape with three music videos. After the videos, you will see the “candid camera" conversation that these two candidates had when they were getting acquainted and didn't know they were being recorded and then you will see the actual audition tape with their commentaries. Remember to pay attention to the music video the candidates are watching during the practice session because it is the one they will be seen commenting on during the audition. Pause OK, we are ready to begin the presentation. Relax but pay attention. It will take about one half hour. At the end your experimenter will return and give you the questionnaire soliciting your impressions. Remember-begin pushing the button to measure what you feel are meaningful units of action when we get to the point in the tape in which the two candidates are getting acquainted. 159 APPENDIX I Consent form PSYCHOLOGICAL RESEARCH CONSENT FOFWI MUSIC VIDEO EXPERIMENT l UNDERSTAND THAT IN THIS EXPERIMENT: (1) I will watch a videotape and will be asked for my opinions and impressions. (2) I am free to discontinue my participation in the study at anytime without penalty. (3) The results of the study will be treated in strict confidence, and that I will remain anonymous. Within these restrictions, results of the study will be made available to me at my request. (4) My participation in the study does not guarantee any beneficial results to me. (5) At my request, I can receive additional explanation of the study after my participation is completed. Signed: Date: Course for which credit is to be given: Instructor: Time/Days: C. Hansen: Vlfinter 1987 160 APPENDIX J Partial Debriefing Sheet INFORMATION ABOUT THE MUSIC VIDEO EXPERIMENT The study in which you have participated, obviously, is concerned with MTV (Music Television). MTV is a 24-hour rock video programming service available as a channel on most cable services. The audience to which it seems to be targeted can be characterized as younger children through adolencents to older teenagers (roughly age 9 to age 20). Although solid research data is not available, viewing of MTV in this age group seems fairly pervasive. It has become a major marketing avenue for record albums, artists, and various products and services related to the rock music industry. Interestingly, while viewing of MTV is widespread, relatively little research has been conducted by Psychologists or Sociologists on the impact of this material on viewers. Indeed, with a few exceptions, research has not been conducted on the content of MTV. That is, researchers have not even investigated what it is that viewers are being exposed to. The experiment in which you participated is one of the initial stages of a larger investigation into MTV and its impact on the viewer. We will eventually be using the data you have provided to explore the impact of MTV on such things as attitudes, values, buying habits, and musical appreciation. In the not too distant future, we will have the data from this experiment analyzed. At that point we will be sending a summary of some of the results to you along with a more detailed explanation of the experiment. Thank you again for your participation. Your reactions will be valuable to us in this investigation of MTV. Christine H. Hansen Dept. of Psychology 161 APPENDIX K.1 Scoring and Coding Minimally Structured Recall Subjects have been asked to recall what they remember about the interaction between the male and the female. In both types of interactions a lot of things happen as you know. Recall that subjects have seen of the the following two interactions: (Conversation 1) He comes on to her. She reciprocates his sexual overtures. He is nice to (praises) her. OR (Conversation 2) He comes on to her. She does not reciprocate his overtures. He is nasty to (derogates) her. Described below are the kinds of things subjects will recall. The actual coding system comes later, but these are the kinds of answers you can expect. (1) His behaviors or conversation connoting his coming on to her that were actually available in the tape of the conversation that they saw. (2) His behaviors or conversation connoting his coming on to her that were actually not available in the tape of the conversation that they saw. This, of course, would be an error in recall--they are remembering things which are consistent with what they saw, but these things did not actually happen in the interaction. (3) Her behaviors or conversation connoting her reciprocation of his sexual overtures that were actually available in the tape of Conversation 1, above. (4) Her behaviors or conversation connoting her reciprocation of his sexual advances that were not actually available in the tape of Conversation 1, above. This would be an error; they are remembering something consistent with her reciprocation in Conversation 1 that did not actually occur. (5) Her behaviors or conversation connoting her nonreciprocation of his advances that actually occurred in Conversation 2, above. (6) Her behaviors or conversation connoting her nonreciprocation of his sexual overtures that were not actually present in the tape of Conversation 2, above. This would be an error; they are falsely remembering something consistent with her nonreciprocation. Note: The most likely recall pattern for the previous six categories would be for subjects who saw conversation one to recall things in Category 1 and 3 and for subjects who saw Conversation two to recall things in Category 1 and 5 because these were the behaviors that were actually available on the tapes they saw. But, people can make many different kinds of errors, like recalling consistent things that were not really on the tapes or even things that coincidentally were present on the other tape, so all these categories are needed-just in case. (7) His behaviors or conversation connoting his praise of her that were actually available in Conversation 1, above. (8) His behaviors or conversation connoting his praise of her that were not present in Conversation 1, above. (9) His behaviors or conversation connoting his derogation of her that were present. (10) His behaviors or conversation connoting derogation that were not present. 162 (11) The balance of events on the tapes which actually did occur. Subjects might recall an event having nothing to do with (a) his coming on to her, (b) her reciprocation or nonreciprocation, or (c) his praise or derogation. (For example: he talked about spending a lot of money on concert tickets.) (12) Or, subjects could incorrectly recall an event which did not connote either a, b, or c (above) that actually did not occur at all in the tapes. (For example: he ate a pizza.) Coding Category A: Code the statement an A if it connotes his coming on to her sexually. If the event described actually occurred in either of the conversations, code the statement as an A1. If the event really did not occur in either tape, code the statement as an A2 (error). Category B: Code a statement as a B if connotes her reciprocation of his overtures. If the behavior actually occurred in one of the interactions, code it as a B1. If the behavior recalled never occurred (errors), code it as a B2. Category C: Code a statement as a C if it connotes her nonreciprocation of his overtures. Again, code it as a Cl if the behavior occurred during the interaction and as a C2 if it did not occur. Category D: Code statements as a D if they connote his praise of her. If the behavior actually occurred, code it as a D1; if it did not occur in the interaction, code it a 02. Category E: Code the response as an E if it connotes his derogation of her. If the behavior was really present in the tapes, code it an E1; if it was not, code it an E2. Category F: This is a residual category. Only code a statement an F if it absolutely does not fit into any of the categories above. Don't think of this as a ”garbage can" category for all the statements you have trouble with. Statements placed here will probably have to be discarded eventually, so place statements into this category with the same care as those placed into other categories. Your codings will go on the coding sheets. All categories of statements have been assigned a code number from 1 to 12. After you have judged each response, total the number of responses receiving each code number and enter that figure in the summary box at the top of the coding sheet. If no statements received a particular code, enter a "0" in that category. Finally, put the total number of male responses and the total number of female responses in the summary box. 163 female SUBJECT NUMBER JUDGE STATEMENT "UMBER lg 59% ESE};] Iglg (TEE E E (5“; E «ISLE 6» E RE GEE-PE (BEE €51 63E “E <é-PE ”BE (35%] (8‘- 3,,E N E "‘ E tIIIC—fiésE °E8§I QEI EJE QUE (13.? E .QEI (‘6: E QJE élel SE éE‘EIE —- E JE @‘aEI "iii lfigE {FEE 813E 8 E @211 EEE GEE ér—‘E E 63E E a a5 E E A N ; re [7 r- , 2 E “El E E E E E («LEI “MEI E E E — E -' E E E E E @5- E E .- E - E E @QEI E E E E E b E c,- E E E E ”DIE ”ODE—I _ IE “QB _ E Q EI E —E menE -—E ’IEI 6GB E E ,E E E E EIE EB DE EIB EE, EB 164 NT MU AT ION BHEET: CO ENTER BUMMARY 0" FIRST SHEET. SUBJECT UMBER MOE STATEMENT NUMBER E E E E E E El E N E N..,E E E %%E ENE 92,? E E E ' E N..- «OI S E E E E; :- ("GEE E <~ El NOE «LE ~QE E E @E— E E @- E, E — E “EEI _ E “mEl ,DE E E —E’JGE83E—E E—E E E E E E E 3 to o O N ‘- t; 53 L: 2 a g: E E E E E E E3 Ea EL:I Ell; Elg Elg %E E E E E E [3 C EI IEI IE IE] El E , E E E E E E E E 803on E E E E — E E E E E E ®E ®E E c E , E E eGE 63E E ’E eE E E E E E E E‘E EEI E’E EEI E'E EIE 165 APPENDIX L.1 Scoring and Coding Structured Recall In this measure, two types of coding are needed. The first coding rates how accurately subjects recall the female's response to a statement from the male (in the getting acquainted segment) and the male's response to a statement from the female (in the audition segment). The second codes the strength of each response. Accuracy: In judging accuracy of recall, we are looking for accuracy of what was conveyed by the statement in the interaction--not necessarily word-for-word recall. We have given a great deal of thought to the range of possible responses we expect and have come up with a set of rules we would like you to follow in coding them. Ultimately, though, you will have to reach agreement with us and with each other on how to score them; this is not an easy task--but one that can be done (at least that is what Robbie Stewart tells us). In practicing between ourselves, we have found that the first type of coding (accuracy) is much harder than the second (strength), so the instructions will be much more extensive for the accuracy task. ACCURACY (DOM-7:: You will need to know the subject‘s experimental condition to code the statements. You need to know whether she/he received the Reciprocation/Praise interaction script or the Nonreciprocation/Derogation script. There are 10 statements you will code in all: 4 from the getting acquainted segment (either Reciprocation or Nonreciprocation by the female) and 6 from the audition segment (either Praise or Derogation by the male). Here are the basic rules for coding each of the 10 individual responses: Reciprocation Condition Here is the statement from the script and the perfect response for the first item: #1 Male: "That's a nice blouse you're wearing.” Female: “Well, I'm glad you like it.“ The best responses will contain the word you . What is critical, here is that she is pleased that the male likes the blouse. Any response that does not contain this component cannot be scored a +5. Statements missing this component can not be rated above a +3, but they could be less. Here is a rank ordering of possible responses and their appropriate ratings: +5 Well, I'm so happy you like it. (synonyms are all right) I'm happy you think it's nice. (again, the meanings are synonymous) +3 You like it? (has the key point, but is hesitant) +2 I like it too. (gets some points for the liking component) Thanks, i like it too. 0 Thanks. (not at all like the real response) Thank you very much. 166 #2 Male: “Nice material. Really soft and sexy." Female: 'It feels nice to wear it.“ In the second response, feels nice (while being worn) is the critical component. +5 It feels so nice when I wear it. +4 It feels nice. (missing the part about being worn) +3 I feel sexy when I wear it. (talks about wearing but misses feels nice) It feels soft when I wear it. I like wearing it. +2 It is nice, isn't it? (some points for nice ) 0 Thanks 0 I'm glad you like it. #3 Male: ”Well, if I knew where your, ah, the thermostat was, I'd turn it up for you.” Female: ”I'd like that.” Here, I and like (or synonyms) are most important. What is really meant by this response is a warm and positive acceptance of his offer and the sense that it would really please her. Positive statements indicating a warm acceptance will be given some points. +5 I'd love that. I'd like that. M like that a lot. +3 That would be nice. (less emphatic) +1 Okay (acceptance, but no sense of her being pleased) Thanks 0 Why not? (flippant remark) #4 Male: ”I'm really excited about getting this job. If we do get it, we'd have to spend a lot of time together, and we'd probably get to go to some concerts." Female: 'I'd love it; it'd be great. I feel like I know you better than some of the guys I've dated.“ Treat this response as two components. I'd love it or It'd be great is the first component, and know you better... dated is the second. Both components are needed for a perfect rating, and if either component is missing, the rating goes down to +3 (or less). +5 That would be great. I feel like I know you better than guys I've dated for a long time. +4 I'd love it. I feel like we're friends already. (close to meaning, but less emphatic) +3 I'd love it. That would be great. Great. I'd really like that. (synonymous with I'd love it) I feel like I know you better than some of the guys I've dated. +2 Great (does not have the sense that she would really like it, but points for great). 167 I feel like we're friends already (some of the right words but misses the better than ...connotation) 0 Awesome (definitely not in the script) Nonreciprocation Condition #1 Male: "That's a nice blouse you're wearing.” Female: "Thank you.” Here we want the whole response for a perfect (-5) response. She is responding in a polite, but noncommital way. -5 Thank you -4 Thanks 0 I like it I'm glad you like it #2 Male: ”Nice material. Really soft and kind of sexy.” Female: "It gets wrinkled when you even touch it sometimes ." The important words are wrinkled (or a synonym) and you even touch it. Actually, she is telling him in a nice way not to touch her. -5 When you even touch it, it gets wrinkled. (order isn't critical) It gets messed up If you even touch it. (synonym) -4 It gets wrinkled if you touch it, sometimes. (missing even ) -3 It gets wrinkled easily if anyone touches it. (missing you ) -2 It gets wrinkled. (points for wrinkled) -1 Please don't touch it. (a point for touch but missing you ) 0 Get your hands off me. (not there) #3 Male: "If I knew where your, ah, the thermostat was, I'd turn it up for you.” Female: ”That's OK. l'llbefine" Here, we want two components for a perfect answer: That's okay (or a synonym) and I'll be fine (or a synonym). If either component is missing, do not score higher than -3. -5 That's all right. I'll be fine. (synonym for one of the two components) -4 That's fine. I'll be okay. (transpose) -3 That's OK. I'llbefine. 0 No thanks. Don't get smart. 168 #4 Male: ”I'm really excited about getting this job. If we get it, we'd have to spend a lot of time together, and we'd probably get to go to some concerts.” Female: "I suppose so, but I don't feel like I know you that well yet.” This response has two components also. I suppose so (which expresses hesitancy, not acceptance) and I don't know you that well yet. If either is missing, do not score above -3. If one is present, give it a -3. -5 I guess so, but I really don't feel like I know you that well. (hesitancy and same meaning) -3 I suppose so. I don't know you that well yet. -1 I don't know; maybe after we know each other better. (a point for know , but not accurate) 0 No way, buddy. (not there at all) Praise Condition #1 Female: Without a doubt, Bruce is one of America's hottest performers. Male: You're right, looking back on his 84-85 concert tour, it's true; he really appeals to a wide range of people. Here, we are looking for You're right and it's true (or synonyms), the 84-85 concert tour and also the he really appeals component for a perfect score. Because there is some semantic overlap between you're right and it's true, however, if one of these is missing, deduct one point only. If both are missing, though, only score it a +2. The 84-85 concert tour component or the he really appeals to a wide range of people component are worth one point (see examples) each. +5 You're absolutely correct. You can see from his 84-85 tour that lots of people like him. (same meaning) +4 Looking back on his 84-85 concert tour, it's true; he really appeals to a wide range of people. (missing you're right) You're right, looking back on his 84-85 concert tour, he really appeals to a wide range of people. (missing it's true) +3 As you said, he is really popular. (has synonyms for most of the meaning) He had a really successful tour, as you accurately point out. (ditto) +2 Looking back on his 84-85 tour, he really appeals to a wide range of people. (missing you're right and it's true) +1 He really appeals to a wide range of people. He is really popular. He had a great 84-85 concert tour. 0 That's a great point. (not there) 169 #2 Female: For a live stage performance, this is definitely one of his better produced videos. Male: That‘s an excellent point. Some people felt that the bad lip sync was distracting, but, small technicalities aside, it was one of the great performances. lf missing excellent point , do not score above +3. Bad lip sync and one of the great performances are each worth one point. +5 Excellent point. Despite the bad lip sync, it was a great video. (same meaning) +4 Excellent point. It was one of the great performances. (missing lip sync) Great point. You don't really notice the bad lip sync. (missing great performance) +3 Excellent point. That's a really great point. (synonym) +2 Some people felt that the bad lip sync was distracting, but, small technicalities aside, it was one of the great performances. (missing excellent point) +1 It was one of the great performances. (or synonym) The bad lip sync was distracting. (or synonym) 0 You said it. (not there) #3 Female: Right, it's what you hear that's important, not what you see. Male: Boy, I love the way you go to the heart of it. I love and go to the heart of it (or a synonym) are the critical components. If one is missing, score it as a +3. +5 I love the way you see deeply into things. (synonymous meaning) +4 I love your analysis. (misses positive connotation of go to the heart of it just slightly) +3 You really go to the heart of it. I love the kinds of comments you make. (has love but misses in last part) +1 That's the heart of it. (no you go to the heart... ) I love the way you talk. (a point for I love but wrong thing is loved) 0 You're so bright. (not there) #4 Female: The video portrays hard-working, middle-class America as Bruce sees it. A powerful point-of-view direct from his upbringing. And he sure anticipated the new mood of patriotism. Male: Well, that's an interesting observation. I hadn't thought of that. You have a lot of good insight. Each of the phrases is a component. If either is missing, score a +3. If two are missing, score a +1. +5 That's a good point. I never thought of it. You really have insight. (synonyms okay) 170 +3 I didn't think of that. That's an interesting point. +1 That's an interesting observation. You have a lot of good insight. That takes insight to see that. 0 He sure did. (not there) #5 Female: Bruce reveals a lot about himself in this video. The Vietnam theme surfaced because he had a brother who was missing in action there. Male: Yeah, they publicized that aspect quite heavily when they released the video, but you're right to remind the audience of that because it makes the video that much more personal. Here, we are looking for the you're right to remind the audience component, especially. Publicized heavily and more personal are worth points, too. +5 Even though that was well publicized, it makes the video more personal when you remind the audience of it. (same meaning) +4 That was well publicized, but you're right to remind the audience. +2 Yeah, they publicized that aspect quite heavily when they released the video, and it makes the video that much more personal. (missing you're right) +1 They publicized that aspect quite heavily when they released the video. That makes the video that much more personal. 0 I heard that too. (not there) #6 Female: Well, there are lots of different things to like about the Boss in this video. Male: That‘s for sure. And, like you, he puts it all together. The critical component to get a perfect (+5) score is the like you component (comparing her to Bruce). That's for sure and he puts it all together (or synonyms) are worth one point. +5 That's certainly true. And, like you, he puts it all together. That's for sure. And he puts it all together--like you. That's for sure. And, like you, he has really got his act together (same meaning) +4 Like you, he puts it all together. Like you, he has really got his act together (same meaning) +2 That's for sure. He puts it all together. (missing like you ) +1 That's for sure. He puts it all together. 0 He's great. (not there) 171 Derogation Condition #1 Female: Without a doubt, Bruce is one of America's hottest performers. Male: Well, looking back on his 84-85 concert tour, it's obvious that he really appeals to a wide range of people. The critical component is it's obvious (or a synonym). 84-85 concert tour and he really appeals to a wide range of people are worth one point each. -5 From his 84-85 concert tour, it's obvious that he really appeals to many people. -3 That's obvious. Anyone can see that. -2 He had a great tour in 84-85. He appeals to a lot of people. (missing obvious ) -1 It was a great tour in 84-85. He really appeals to a wide range of people. 0 He sure is. (not there) #2 Female: For a live stage performance, this is definitely one of his better produced videos. Male: Well I don't know about that. A lot of people felt that the bad lip sync was distracting, but I guess you'd have to look at it more closely than you did to really evaluate it. But, technicalities aside, that was one of the great performances. Here, the hesitancy of I don't know about that and you'd have to look at it more closely than you did are the important components. If either is missing, don't score above -3. The lip sync part is worth one point, as is one of the great performances, but only one of these is needed if the first components are there. -5 Well I'm not sure about that. Some people felt that the bad lip sync was distracting, but I guess you'd have to look at it more closely than you did to really evaluate it. -3 You'd have to look at it more closely than you did to really evaluate it. Well I don't know about that. I don't necessarily agree with that. (same meaning) -2 Even with the bad lip sync, it was one of the great performances. -1 The lip sync was bad. It was one of the great performances. 0 It had a few flaws. (not there) #3 Female: Right, it's what you hear that's important, not what you see. Male: Sure, and I suppose when people go to concerts they keep their eyes closed. We are looking for a sarcastic comment here about keep their eyes closed. Without sarcasm, their response can not be worth more than -3. -4 Are you saying people are blind at concerts? (sarcasm, but just misses eyes closed ) 172 And people don't watch it, I suppose? (ditto) People don't keep their eyes closed at concerts. (not sarcastic) Oh sure. (one point for sure ). What a dumb comment. (not there) Female: The video portrays hard-working, middle-class America as Bruce sees it. A powerful point-of-view direct from his upbringing. And he sure anticipated the new mood of patriotism. Male: Well, that goes without saying. I'm sure that won't be news to our audience. Both that goes without saying and I'm sure that won't be news to our audience (or synonyms) are important components. If either is missing, score a -3. -5 -3 -2 -1 0 #5 Well, that's obvious. I'm sure everyone knows that. (same meaning) That goes without saying. I'm sure that won't be news to our audience. That's not news. (misses audience) Our audience knows that. (a point for audience ) Yes, he did. (not there) Female: He sure reveals a lot about himself in this video. He had a brother who was missing in action in Vietnam. Male: Yeah, they publicized that aspect heavily when they released the video but it really only makes a small contribution compared to the tremendous impact of his energy in the video. The important point of this response is a demeaning only makes a small contribution . Publicized and tremendous impact of his energy are worth one point each. -5 -4 -1 #6 Yeah, they publicized that aspect heavily when the video came out, but that's not very important compared to the tremendous impact of his energy in the video. That was publicized heavily, but it is not the most important thing about the video. (missing energy) That's not very important compared to the tremendous impact of his energy in the video. (missing publicized ) That isn't an important contribution. There is a lot of energy in the video. That was heavily publicized. I remember that. (not there) Female: Well, there are lots of different things to like about the Boss in this video. Male: That's for sure. But, the point is that he puts it all together. Here, he is telling her that she missed the point. Both But and the point is he puts it all together are important for a perfect (-5) score. If either is missing, give it a -3. 173 That's for sure is worth one point. -5 Right, but the point is that he puts it all together. (same meaning) That's for sure, but you missed the point. (ditto) -4 But, the point is that he puts it all together. (missing That's for sure ) -3 The point is that he puts it all together. -2 You missed the point. ( some points for point ) -1 That's for sure. 0 Yes, there are. (not there) 174 STRENGTH OF INTENT ACCURACY OF RESPONSE RECALL c333... .8 :3. 9.9.5 99.38 coach? .5 =2. 53.39.. .8 :3. x3) :30. .8 33¢ ...-53: 20>... ..o :26 it... 99.58 22¢ .3 Ed 322 .3 .35 9.9.6 93:» .2 053?. 35an.: 9...qu SE. «52...... 982 232...... 25. 9:3: to: 23...“. 3582.: 3:82.”. o: .8 23:8 oz 239:. 3559.; Etta :2» 9:52 to: 052.2 5.5 Santa {oz 8.2.: 3529?; 9.2:. =4 GETTING ACQUAINTED: Recip +2 *3 0000000 -2 ~10 +1 Norecip -3 Reciprocation +2 +3 +4 +5 -3 -'2 -10 +1 Nonreciprocation ~5-4 STATEMENT 1 00000000000 '2-10+1+2+3 0000000 -3 +2 +3 +4 +5 -’2 -10 +1 -3 00000000000 -s-4 STATEMENT 2 -2—10+1+2+3 OOOOOOO ~25 -2 -1 0+1 +2 +3 +4 +5 -3 00000000000 .4 -5 STATEMENT 3 -2-10+t+2+3 0000000 -3 STATEMENT 4 AUDITIW: Praise +2 +3 0+! OOOOOOO Derogation Praise +2 +3 +4 +5 Derogation -2 -1 0+1 '3 OOO -2-l OOOOOOOO -s—4 —4 -3 : “NJ STATEMENT 1 -2-IO+I+2+3 0000000 -3 +2+3+4+S 00000000000 -2 -10 +1 ~35 STATEMENT 2 +2 +3 +4 +5 -3-2-IO+I OOOOOOOOOOO -s-4 '2-I0+I*2+3 0000000 -3 STATEMENT 3 -2-l0+l+2+3 OOOOOOO ~25 STATEMENT 4 000 0 +2 +3 +4 +5 STATEMENT 5 -2-IO+I+2+3 0000000 -3 .....O 4. ...s 0 STATEMENT 6 SUBJECT NUMBER JUDGE 175 APPENDIX M.1 Coded Unitizing Segments: Reciprocation-Praise (C a ”coming on" behavior, R = "her reciprocation", P a ”his praise”) Reciprocation (M) Hello. (F) Hi. (M) [C1 Are you here for the VJ job too?] (F) [R1Yeh. I guess we're going to be partners]. (M) Great. (M) My name's Marty. (F) Hi, my name is Jeanine. (M) Nice to meet you Jeanine. How ya doin'? (F) Good. (F) Looks like Bruce Springsteen. (M) It is. Born in the USA. It's a good video. (F) 80 have you ever DJ'd before? (M) No, but I had a class in public speaking. How about you? (F) l DJ'd at a campus radio station in Milwaukee. That was really fun. I'm glad to be back in Michigan, though. (M) Where do you live? (F) [R2 I just moved from Ypsilanti, and I got an apartment over on Walton.] (M) Oh. I hit a muskrat driving down there a couple of days ago. Real traumatic experience. (F) Oh my God. I would hate to have that happen. I'm glad that didn't happen to me. (M) [C2 Jeanine, do you...ah... live alone?] (F) [R3 I do right now, how about you?] (M) I live with a guy who just got divorced, but he's out so much it's almost as if I live alone. (F) How do you like your roommate? (M) Well, like I said he's out a lot, but when he's in he drives me crazy. If there's one thing I do not like, it's someone really cheap or who watch every penny. For instance, we went to W. Virginia over the weekend, and when we got back he added up how much everyone contributed to the general fund, and it turned out that he thought he paid too much. So he wanted us to reimburse him. (F) Sounds cheap. (M) Very. Talk about cheap. You should see his wardrobe. (F) laughs (M) Plaid levi shirts and velour shirts. (F) Oh my God. Sounds like you're not very compatible. (M) [C3 No. I've been thinking about trying to move out, and I'm gonna have to find someone to live with.] (F) Um hmm. (M) [C4 That's a nice blouse you're wearing] (F) [R4 Oh, I'm glad you like it.] (M) [05 Nice material. Really soft and sexy. ] (F) [R5 It feels nice to wear it.] (F) It's kinda cold in here, I wish they'd turn up the heat. (M) Really. Do you get cold easily? (F) [R6 Well. Not too easily.) (F) [CG If I knew where your thermostat was l'd turn it up for you.] (F) [R7 Oh. I'd like that.] (M) Jeanine, did you see him in concert. (F) Yeh. I didn't like paying the $20 for the ticket tho. (M) $20. Oh, that isn't too bad. I heard some people paid as high as $400 for these tickets. (F) You're kidding. (M) No, but it was front row seats, of course. (F) I still wouldn't do it. [C7 he glances down her upper body.) 176 (M) [CB You know, I'm really excited about getting this job. If we do get it, we'd have to spend a lot of time together, and we'd probably get to go to some concerts] (F) [R8 I'd love it. It'd be great. I feel like I know you better than some of the guys I've dated.] (M) [CS You'll get to know me better.] Praise (F) Without a doubt. Bruce is one of America's hottest performers. (M) [P1 You're right, looking back on his 84-85 concert tour, it's true; he really appeals to a wide range of people. (F)For a live stage performance, this is definitely one of his better produced videos] (M) [P2 That's an excellent point. Some people felt that the bad lip sync was distracting, but, small technicalities aside, it was one of the great performances. Super video.] (F) Right, It's what you hear that's important, not what you see. (M) [P3 Boy, I love the way you go to the heart of it.] (F) Well, he sure feels the pulse of the country. This video portrays hard-working, middle-class America as Bruce sees it. A powerful point-of-view direct from his upbringing. And he sure anticipated the new mood of patriotism. (M) [P4 Well, that's an interesting observation. I hadn't thought of that. You have a lot of good insight] (F) Well, Bruce reveals a lot about himself in this video. The Vietnam theme surfaced because he had a brother who was missing in action there. (M) [P5 Yeah, they publicized that aspect quite heavily when they released the video, but you're right to remind the audience of that because it makes the video that much more...personal.] (F) Well, there are lots of different things to like about the Boss in this video. (M) P6 That's for sure. And, like you, he puts it all together] (F) OK. We'll be back after this next video with some concert news of local interest. 177 APPENDIX M2 Unitizing Coded Segments: Nonreciprocation-Derogation (C = ”coming on“ behavior, NR = ”her nonreciprocation", D = ”his derogation") Nonreciprocation (M) Hello. (F) Hi. (M) [C1 Are you here for the VJ job too?] (F) [NR1Yeh. I guess we're going to be partners] (M) Great. (M) My name's Marty. (F) Hi, my name is Jeanine. (M) Nice to meet you Jeanine. How ya doin'? (F) OK. I'm kind of nervous about this interview. (M) Oh don't be. Just be yourself and act natural. (F) Looks like Springsteen. (M) Born in the USA. (F) Have you ever DJ'd before? (M) No, but I had aclass in public speaking. How about you? (F) Well, I had a job working at a campus radio station as a DJ. And I met a lot of nice people and everything. I'm glad to be back in Michigan, now. (mumbles about parents in Marquette) (M) (mumbles Oh boy) Where do you live? (F) [NR2 Well, I just moved from Ypsilanti over to an apartment on Walton.] (M) Oh, Walton. I hit a muskrat there a couple of days ago. (F) You're kidding. (M) No, real traumatic experience. (F) I wouldn't want that to happen to me. (M) [02 Jeanine, do you live alone?] (F) [NR3 I do right now; do you?] (M) I live with a guy who just got divorced, but he's out so much it's like I live alone. (F) How do you like your roommate? (M) Um, lfi