شهادت ناظران عینی برای جرم و جنایت شبیه سازی شده نوجوانان توسط مجرمان مرد و زن با ویژگی های نقش جنسیتی سازگار یا ناسازگار

Eyewitness recall by 60 adolescents and 60 young adults in Experiment 1 and by 64 children and 63 preadolescents in Experiment 2 for a simulated theft in which gender-role characteristics and sex of criminal were manipulated (i.e., masculine male, feminine male, feminine female, masculine female) was investigated. Gender-role flexibility impacted on crime details and accuracy for criminal features and children's reports included fewer crime and criminal features and central crime details than did preadolescents. Children's gender-role beliefs differentially affected errors made when describing criminals with inconsistent gender-role characteristics. Adolescents provided fewer crime and criminal details and criminal features than did young adults, although both converted inconsistent into consistent gender-role information for the “feminine male” criminal. Forensic implications of the findings are discussed.

Eyewitnesses serve as a critical part of legal strategy because they can provide police with evidence that leads to the arrest and prosecution of criminals, specifically descriptions of what transpired during the crime and descriptions of the criminal(s). Arrest records show that bystanders are available for police to interview in over half of all misdemeanor crimes (Smith & Visher, 1981) and that their presence encourages victims to report the crime (Greenberg, Wilson, Ruback, & Mills, 1979), resulting in more than 75% of all arrests. Juvenile crimes are oftentimes spurred by peer pressure, making it likely bystanders and/or victims will be the same-age or younger than the criminal (Brown et al., 1986 and Zimring, 1998). Surprisingly, juveniles are an underrepresented group in research on eyewitness behavior, even though they commit 18% of the crimes and represent one-fourth of crime victims in the United States (Finkelhor, 1998, as cited in Office for Victims of Crime Monograph, 1999). Preadolescents and early-to-middle adolescents (ages 10 to 15 years) constituted 32% of the 1999 juvenile arrests (Snyder, 2003) and were responsible for 42% of property crimes (U.S. Department of Justice, 1999), with larceny–theft accounting for 70% of the Property Crime Index from 1997–2001. High as they are, these arrest records are conservative estimates of actual delinquent behavior because only one-third of all crimes are reported to police (Law Enforcement Assistance Administration, 1982). Accordingly, the eyewitness ability of juveniles to provide accurate and reliable legal testimony has been of considerable interest to both legal and psychological professionals. Accuracy in describing a crime and criminal is complicated by the fact that eyewitness recall is incidental. That is, people's attention is usually directed to completing the goals for the activities in which they are engaged (e.g., shopping, waiting for a bus) at the time of a crime rather than on memorizing a criminal's actions and appearance. Consequently, eyewitnesses may fill in gaps in their recall. One way that they may do that is to rely on cognitive schemas (i.e., mental representations that organize and interpret experiences), preconceived notions, or stereotypes about crimes and criminals when reporting crimes they witnessed (Holst and Pezdek, 1992, List, 1986, Smith, 1991 and Tuckey and Brewer, 2003b) and when describing and identifying a criminal (MacLin and Herrera, 2006 and Shoemaker et al., 1973), regardless of whether these beliefs are accurate (Bull, 1979). The goal of the research reported in this article was to address the impact of such schemas, particularly stereotypes about crimes (e.g., bank robberies) and about criminals (e.g., young adult), on eyewitness memory. I chose to focus on and manipulate one salient schema, the gender schema, as an illustration of how beliefs may influence eyewitness memory. Societal expectations about criminals include information in terms of their probable sex, appearance, and behavior. Thus, evoking ideas of a criminal are likely to activate eyewitness' gender schema (i.e., stereotypes of being a male or female in our culture). Information that is congruent with this schema (e.g., criminals are male) may be remembered differently than information that is incongruent (e.g. interpreting a female suspect's behavior of taking a bicycle without permission as borrowing rather than stealing, but interpreting identical behavior by a male suspect as stealing rather than borrowing). Gender schemas themselves may also affect eyewitnesses' understanding of a crime. A given behavior, such as patting victim on head while asking to use the bicycle, judged to be feminine will be considered gender-consistent when performed by a female suspect and may be interpreted as sweet-talking, but will be considered gender-inconsistent when performed by a male suspect and may be interpreted as bullying. These differential interpretations presumably influence police discretionary decisions about consequences, such as to arrest or not arrest the suspect (Hoyt & Scherer, 1998). Role of cognitive schemas in eyewitness memory The current investigation examined how cognitive schemas affect eyewitnesses' incidental memory for a simulated juvenile crime by an adolescent criminal who varied in both sex and gender-role characteristics. Cognitive schemas for crimes and criminals, organized as expectations about what typically occurs in crimes and about criminals' physical attributes, behaviors, motives, and personality (Hansel, 1987 and Stalans, 1993), are likely to guide eyewitnesses' memory processes (Hamilton et al., 1990, Hamilton and Sherman, 1994 and Sherman et al., 2000). Experiment 1 compared immediate recall by adolescents (12 to 15 years) and young adults (18 to 22 years), and Experiment 2 compared immediate recall by children (6 to 7 years) and preadolescents (9 to 11 years). Evidence from the adult and developmental literatures on eyewitness recall has shown that more expectancy-consistent information than expectancy-inconsistent information is recalled (e.g., Srull & Wyer, 1989), but this advantage is affected by the strength of the expectations and the type of information reported (Stangor & McMillan, 1992). For example, in Farrar and Goodman's (1990) schema confirmation-deployment model, there is a three-step process of activating a schema, confirming information consistent with the schema, and then deploying the schema in recall. That is, eyewitnesses will activate (or formulate if none exists) appropriate cognitive schemas to aid perception, comprehension, encoding, and retention of the crime and criminal. During the schema confirmation phase of processing, the eyewitness will devote cognitive resources to verifying expectancy-consistent information in the activated schemas, distinguishing it from expectancy-inconsistent information. Thus, schema activation decreases cognitive effort, making schema-consistent information easier to interpret and more accessible during recall (Sherman et al., 2000, Smith and Studebaker, 1996, Srull and Wyer, 1989 and Tuckey and Brewer, 2003a) and provides a framework for remembering the experience. Schema deployment begins after schema activation and leads to selective attention to expectancy-inconsistent information to establish a distinct memory for the crime and criminal while devoting little attention to processing expectancy-consistent information. The model also suggests that if there are no existing schemas, the eyewitness will generalize from past experiences, but will not differentiate event information presented in the crime as consistent or inconsistent with expectations, and thus will not engage in schema deployment. Although crime and criminal schemas can facilitate recall of expectancy-consistent information, errors of omission, distortion, or commission in testimony often occur when a criminal's actions and/or physical characteristics are expectancy-inconsistent. Under low cognitive demands of familiar events, even children can attend to expectancy-inconsistent information; however, under high cognitive demands of eyewitness situations, people fail to process expectancy-inconsistent information, instead focusing on expectancy-consistent information (Van Knippenberg, Dijksterhuis, & Vermeulen, 1999). This leads to filling in memory gaps with expectancy-consistent, but inaccurate information about the criminal and the crime (Sherman & Bessenoff, 1999). Thus, eyewitnesses may provide incomplete reports by omitting expectancy-inconsistent information or produce inaccurate reports by distorting stored inconsistent information to reflect their biases or adding expectancy-consistent information that was not actually present (Harris et al., 1988, Holst and Pezdek, 1992 and Smith and Studebaker, 1996). For example, Tuckey and Brewer (2003a) found that adult eyewitnesses who were shown a filmed bank robbery converted expectancy-inconsistent information (e.g., robbers were carrying bags) into expectancy-consistent information (e.g., robbers were carrying guns). Eyewitnesses also mistakenly report salient details that are typical of a particular kind of crime, but that were not actually present (Greenberg et al., 1998 and Smith and Studebaker, 1996), perhaps as a way of filling-in-the-gaps for missing, ambiguous, or forgotten information (Brewer & Nakamura, 1984). The production of such “intrusion errors” during a trial would decrease eyewitness reliability (Holst & Pezdek, 1992). Drawing from the adult literature, there is evidence that people have well-developed event schemas for crimes, such as different kinds of robberies (e.g., bank, convenience store, mugging) (García-Bajos and Migueles, 2003, Holst and Pezdek, 1992, List, 1986 and Tuckey and Brewer, 2003b), and have well-developed schemas for criminals including typical physical attributes and personal traits (Bowers and MacLin, 2004, March, MacLin and Herrera, 2006, Madriz, 1997 and Reed and Reed, 1973). Thus, there is general agreement concerning the actions involved in common crimes and a criminal's appearance. For example, stereotypes of adult criminals by college students (Bowers & MacLin; Greenberg et al., 1998; MacLin & Herrera) include that most are male, Caucasian, approximately 24-years-old (ranging from 16 to 35), have brown or black hair, and exhibit typical masculine physical characteristics for their hairstyle (i.e., short), build (i.e., medium), height (i.e., tall, about 6 ft. in height), weight (between 180 and 200 lbs), and personality traits (e.g., aggressive). These stereotypes are derived from a variety of sources (e.g., newspapers, internet, television shows) besides direct experience, although the features commonly reported are not legally correct, relevant, or complete (Smith, 1991). Research has also shown that eyewitnesses differentially recall information about crimes and criminals (e.g., Bull and Green, 1980, Chapman, 1973 and Shoemaker et al., 1973). For example, eyewitnesses recall more central than peripheral aspects of a crime (Cassel and Bjorklund, 1995 and Tuckey and Brewer, 2003a) and more information about the criminal's appearance than about peripheral aspects of the crime (Parker et al., 1986 and Shapiro et al., 2005). Eyewitness testimony research has indicated age-related differences in memory for crimes. In recalling various crimes committed by adults, both children (5 to 7 years) and preadolescents (8 to 10 years) have less complete and accurate reports for petty thefts (compared with early adolescents,11/12 years, or with young adults, 18 to 21 years) (Cohen and Harnick, 1980 and Sutherland and Hayne, 2001); for theft (compared with early adolescents and middle adolescents, 14 to 16 years) (Saywitz, 1987); and for a kidnapping (compared with late adolescents, 16/17 years or with young adults) (Coxon & Valentine, 1997). Similar age-related differences have been found for recall of crimes committed by juvenile offenders. Children provide fewer central features about a bicycle theft than do preadolescents and adults (Shapiro et al., 2005). Moreover, early adolescents (Sutherland & Hayne, 2001) and middle adolescents (Marin, Holmes, Gruth, & Kovak, 1979) report less event information than do college students. There are also age-related differences in eyewitness descriptions of criminals. Children provide less information than adults about an adolescent criminal's appearance (Shapiro et al., 2005) and are less accurate than preadolescents in their descriptions of the physical appearance of an adult (e.g., face, clothing) who gave them a simulated health inspection (Davies, Tarrant, & Flin, 1989). Age-related differences have also been found for specific descriptors, such as age, height, and weight. Goetze (1980, as cited in Davies, Stevenson-Robb, & Flin, 1988) found that children were less accurate than preadolescents and early adolescents at estimating height and weight of an adult purse snatcher, whereas all three age groups did poorly in estimating age. Davies et al. (1988) replicated and extended Goetze's age-related accuracy differences using absolute judgments for height, weight, and age of an adult school visitor, but reported that relative judgments (which used the interrogator as a baseline) were more accurate, particularly for preadolescents and early adolescents. These age-related findings suggest that juveniles have more limited processing ability, possibly due to absent or newly formed schemas containing little knowledge about a crime, than do adults (Farrar and Goodman, 1990 and Luna et al., 2004). Influence of gender schemas in recall of criminal behavior and appearance Societal expectations that thefts are committed by men and that thieves have masculine characteristics suggest that both sex and gender-role are salient aspects of the criminal schema. Gender schemas can be particularly useful when information about a criminal is limited as presentation of one attribute (e.g., appearance) triggers inferences about other attributes (e.g., traits, role behaviors) because the components are interrelated (Deaux and Kite, 1985 and Deaux and Lewis, 1984). Thus, when eyewitnesses are exposed to a theft, gender schemas will enhance recall for a criminal's expectancy-consistent gender-related behavior and appearance, but may distort recall for expectancy-inconsistent information (Bem, 1981, Buckner and Fivush, 1998, Carter and Levy, 1988, Cordua et al., 1979, Deaux and Major, 1987, Koblinsky et al., 1978, Liben and Signorella, 1980, Martin and Ruble, 2004 and Welch-Ross and Schmidt, 1996). The social-cognitive literature indicates phase-like shifts from inflexibility to flexibility in gender-role stereotype beliefs and tolerance of expectancy-inconsistent behaviors and attributes from ages 4 to 18, rather than showing a linear pattern of development (Ullian, 1976). When children are in the initial phase of schema confirmation in which information about male and female roles is being consolidated (Trauter et al., 2005), they demonstrate rigid adherence to their stereotypical scripts for activities and intolerance towards ‘rule-breakers’ although they can draw inferences using gender knowledge about same-sex characters (Martin et al., 2002 and Ruble and Martin, 1998). When children are in the integration phase of schema-confirmation for gender knowledge, typically during preadolescence (Welch-Ross & Schmidt, 1996), they exhibit flexible gender-role attitudes, are capable of making complex inferences about other-sex characters, and can attend to gender-role inconsistent information during schema deployment (Carter and Patterson, 1982, Martin, 1989 and Martin et al., 1990). In contrast to the flexibility shown by preadolescents, attitudes and tolerance (particularly for gender-role violations of physical attributes of male targets, such as hairstyle and clothing), become increasingly more rigid again from early-to-middle adolescence (ages 12–16 years), shortly after the transition to junior high school (Alfieri et al., 1996, Horn, 2007, Hurlock, 1973, Katz, 1979, Lobel, 1994, Lobel et al., 2004 and Stoddart and Turiel, 1985). Adolescent inflexibility may stem from a combination of increased cognitive abilities and onset of sexual maturation that lead youth of this age to be preoccupied with how others view them, to over-identify with stereotypes and gender-typed activities, and to conform their behaviors and physical appearance in exchange for social acceptance by peers (Crockett, 1990, Huston and Alvarez, 1990, Lobel et al., 2004 and Stoddart and Turiel, 1985). As adolescents begin to understand that gender roles are culturally relative social conventions used to regulate societal behavior (Carter & Patterson, 1982), they concomitantly increase discriminatory behavior towards individuals who demonstrate atypical behavior and physical characteristics (Horn, 2007, Hurlock, 1973 and Lobel et al., 2004). In contrast to adolescents, young adults show decreased gender stereotyping and discrimination and increased tolerance for counter-stereotypical behavior (Lobel et al., 2004, Plumb and Cowan, 1984 and Urberg, 1979), perhaps as a result of their ability to integrate knowledge for social conventions with personal choice (Carter and Patterson, 1982 and Nucci, 2001). However, there is some evidence suggesting that the age of the target being evaluated may affect children's and adults' use of gender stereotypes. Children in Urburg's (1982) study chose more stereotyped responses for activities performed by children than by adults. Similarly, Powlishta (2000) reported that adults rated child targets as possessing stereotype-consistent personality traits (i.e., girls were feminine), but showed flexible attitudes in their ratings of adult targets. Consequently, both juvenile and adult eyewitnesses may have difficulty describing juvenile criminals who vary from adult norms in personal appearance, behaviors, and clothing. Albeit for different cognitive and social reasons, both children and adolescents show greater inflexibility and less tolerance than do preadolescents and adults when evaluating gender norm violations, particularly for a boy with feminine physical appearance, judging these violations to be as serious as moral norm violations (Blakemore, 2003, Horn, 2007, Lobel, 1994, Lobel et al., 2004 and Stoddart and Turiel, 1985). Consequently, when faced with a thief who exhibits expectancy-inconsistent (i.e., counter-stereotypical) appearance and behaviors (Devine, 1989), child and adolescent eyewitnesses may be more vulnerable than preadolescent and young adult eyewitnesses to gender-role distortions (Huston and Alvarez, 1990, Koblinsky et al., 1978 and Liben and Signorella, 1993) because they will report information about how the thief should look and what the thief should have been doing rather than what they actually observed. Current research The current research examined how variations in criminal sex and gender-role characteristics influence age-related differences in immediate recall of a juvenile crime and criminal. The to-be-remembered-stimulus was a misdemeanor bicycle theft, as crimes against property are the predominant offenses committed by juveniles (Office of Juvenile Justice and Delinquency Prevention, 2003). Accuracy in eyewitness testimony was examined for adolescents and adults in Experiment 1 and for children and preadolescents in Experiment 2. For both experiments, Cassel and Bjorklund's (1995) film script in which a 14-year-old male criminal steals the bicycle of a 9-year-old female victim was used, with two modifications. First, the videotape viewed by participants did not simply contain the theft. Instead, the theft was embedded within another activity–twin co-eds' outing at the zoo–to simulate real-life situations in which the eyewitnesses continue with a planned activity before being interviewed by the police. Also, the 10-minute zoo outing segment of the videotape served the purpose of an interpolated task. Second, four versions of the bicycle theft were created that crossed criminal sex and criminal gender-role consistent or inconsistent appearance and behavior. Thus, in two versions, a male or a female adolescent criminal demonstrated gender-role consistent physical and behavior characteristics. In the other two versions, a male or female adolescent criminal demonstrated gender-role inconsistent physical and behavior characteristics. Assessments of recall and gender-role beliefs were made through questionnaires for adolescents and young adults (Experiment 1) and through interviews for children and preadolescents (Experiment 2). In order to encourage participants' use of criminal stereotypes to interpret and report expectancy-congruent information (e.g., Rohan & Pettigrew, 1992), they were not informed that they would need to recall the theft. There were a number of hypotheses about eyewitness recall. Eyewitnesses were expected to recall and correctly elaborate more central than peripheral features of the crime (Hypothesis 1); older eyewitnesses, who have more developed cognitive schemas and more flexible gender-role beliefs, were expected to recall and correctly elaborate more crime and criminal features and report fewer errors when describing the criminal than were younger eyewitnesses (Hypothesis 2). Consistent with reliance on a male-as-criminal stereotype, eyewitnesses were expected to recall and elaborate more crime and criminal features for the male than for the female thief (Hypothesis 3). Eyewitnesses were also expected to recall and elaborate correctly more crime and criminal features for the gender-role consistent thief than for the gender-role inconsistent thief as expectancy-consistent information is easier to process in high-load eyewitness situations (Hypothesis 4a). Additionally, younger eyewitnesses and those with weak gender-role flexibility were expected to recall fewer crime and criminal features and to produce more schema-conversion errors (i.e., changing information to be gender-consistent) when describing the feminine male thief than were older eyewitnesses and those with strong gender-role flexibility (Hypothesis 4b). This hypothesis is consistent with the masculine criminal stereotype, but also reflects societal disproval of feminine attributes in boys and tolerance of masculine attributes in girls (Golombok & Fivush, 1994). Moreover, as gender-role beliefs are adhered to rigidly by younger eyewitnesses, processing of expectancy-inconsistent characteristics would be difficult. Finally, the strength of gender-role flexibility will impact on recall and elaboration, such that eyewitnesses who had weak gender-role flexibility would show low recall and elaboration of crime and criminal features and low criminal error in the gender-role inconsistent versions, but would show high recall and elaboration in the gender-role consistent versions (Hypothesis 5).