Eyewitnesses and the Weapon Focus Effect in children Adults and the Elderly
Maria Urtado
Florida International University

ABSTRACT The “weapon focus effect” (WFE), states that a witness of a crime remembers fewer details about the perpetrator’s appearance when the perpetrator has a gun. To date, there are no studies of older adults and the WFE. The present study measures the WFE in older adults (60-70) compared to young adults (20-40) and children (8-12). The study consisted of 180 participants, randomly selected to watch one of three videos. The three videos were the same, with the exception of an object being manipulated. Conclusions were based on how much detail participants were able to recall about the target. Older adults remembered fewer details about the target’s appearance than children and young adults. The results suggest that older adults would take longer focusing on an object that is out of context, therefore, not focusing on the perpetrator, resulting in fewer details remembered.

Eyewitness and the Weapon Focus Effect in children compared to the elderly Many people have been convicted of crimes that they did not commit. Eyewitness testimony is one of the major causes for these wrong convictions. At least 75% of wrongful convictions are the result of inaccurate eyewitness identification (Innocent Project, 2009). Before the public became aware of the large amount of wrongful convictions, psychology researchers were already aware of the problem of memory and eyewitness testimony (Douglas, & Steblay, 2006). DNA testing has helped innocent people that have been convicted, proved their innocence and at the same time it has helped proved what researchers have claim for at least 30, years (Innocence Project, 2009). Eyewitnesses play a very important role in the criminal justice process. There are two important aspects that need to be taken into consideration with regard to eyewitness testimony, these are credibility and accuracy (Bornstein, 1995). Research in these two areas has demonstrated that attorneys prefer to work with adults and young adults over children and elderly witnesses (Yarmey, & Kent, 1980). But even if attorneys prefer to work with adults and young adults, the fact is that children and older adults very often witness or are the victims of crimes. The majority of the research done in eyewitness identification has involved young adults and in recent years there has been an increasing amount of research with children; however there has been very little research done involving older adults and eyewitness identification (Bartlett & Memon, 2007). According to Bartlett & Memon (2007), one of the most important reasons why it is important to conduct studies with older adults and eyewitness identification is that it will help improve theories. In addition, by understanding the ways older adults perform, researchers can develop better predictions about the way they behave and react when witnessing a crime (Bartlett & Memon, 2007). These findings will help improve the ways older eyewitnesses are questioned, and how much weight should be put in their testimonies.
Children and Elderly as Eyewitnesses The elderly as well as children are potential eyewitnesses. Researchers have evaluated how memory processes develop and change throughout the lifespan. Because some children have been subject to physical and sexual abuse there have been more studies about how their memory functions and how they behave as witnesses (Bornstein, 1995). As stated before, there is, however, less research on how the memory of the elderly functions and regarding their role as eyewitness (Bartlett & Memon, 2007). Even though elderly eyewitnesses have been criticized, it has been demonstrated that older witnesses can be as accurate and reliable as young adults, when they are allowed to remember events without pressure and demands (Bornstein, 1995). Also, there is little evidence that suggests that the abilities of eyewitness deteriorate substantially with age (Bornstein, 1995). Research has found, however, that older adults are more vulnerable than young adults about information presented after the event, when the information is misleading; this is because older adults have more difficulty sorting out the real source of events (Bornstein, 1995). There is no research that indicates that the techniques used with younger adults to minimize this problem will work the same way with older adults (Bornstein, 1995). In addition, older adults are vulnerable to the manner in which the questions are being presented, and also to the type of information that they need to remember. In general, it is very important to follow good techniques when questioning eyewitnesses. But in most cases the techniques used with young adults do not have the same effect with older adults. When working with older adults it is important to start with free recall questions, before asking specific questions and to finish with closed ended questions (yes and no) (Bornstein, 1995). In conclusion, eyewitnesses will remember more if proper questioning techniques are used. Research suggests that in the case of older adults, if recognition is emphasized, as opposed to recall or retrieval cues, they will remember more information (Bornstein, 1995). With the development of better techniques elderly eyewitnesses will be better contributors to criminal investigations and trials (Bornstein, 1995).
The Weapon Focus Effect The “weapon focus effect” was best demonstrated by Loftus, Loftus, and Messo (1987). They performed several experiments in which participants were asked to watch a slide sequence. The slides were about customers in a fast food restaurant as they moved through the order line. There were several conditions in the experiment, in one of the conditions; the target person took money from the cashier by either holding a gun or a personal check. When the participants were asked to identify the target, the participants who saw the gun were less accurate than those who saw the check. In addition, the researchers used a corneal reflection device to demonstrate, how the witnesses will focus their eyes longer and more frequently on the gun (Loftus, Loftus, & Messo, 1987). When we think about the reason why a person will focus his or her attention on a weapon it seems plausible to believe that it is because weapons are threatening. There are several hypotheses that attempt to explain the cause of the “weapon focus effect”. According to Loftus, Loftus, & Messo, (1987) the “weapon focus effect” in a real life situation will be a combination of the crime itself and the presence of a weapon. The arousal caused by the high stress of the event, which is intensified by the weapon results in the witness focusing in a smaller area near the weapon (Easterbrook, 1959). Because the witness focuses his/her attention in a smaller area near the weapon, it is more likely that the witness will remember the weapon and will be unable to remember the perpetrator (Loftus, Loftus, & Messo, 1987). In other studies it was found that it is more the fear of the weapon alone what causes witnesses to focus their attention on the weapon. Because of this fear the witness does not spend enough time looking at the perpetrator, with the result of being unable to identify the person responsible for the crime (Shaw & Skolnick, 1999). Recent studies however, have demonstrated that participants in a situation that produces low anxiety also display the “weapon focus effect” (Davis, Smith & Blincoe, 2007). The salience hypothesis state that a weapon causes the “weapon focus effect” in a particular situation when the weapon is not expected, it is argued that this occurs because it is then considered a novel item (Davis, Smith & Blincoe, 2007). In a study by Loftus & Mackworth, (1978, as cited in Davis, Smith & Blincoe, 2007) participants were exposed to a farm scene in which an unexpected object was seen, for example an octopus. In the case of the unexpected object the participants’ eyes focusing on the object occurred earlier and for a longer time than in the scenes where there were no unexpected objects (Loftus & Mackworth, 1978). Many studies have demonstrated that the “weapon focus effect” can occur when unexpected, interesting or novel items appear in certain circumstances (Davies, Smith & Blincoe, 2007). In one study, that was conducted to observe the effects of a threatening object compared to the effects of a novel object. Participants were asked to watch different videos portraying people in a business environment, handling a novel object, a threatening object or no object. After participants watch the video, they were asked to complete a questionnaire. In the questionnaire they were asked to describe the people that they saw in the video. Researchers found that participants gave less accurate descriptions of those people in the video that had a novel item, compared to the neutral item or no item (Pickel, 1998). In another study conducted a year later, the participants watch two videos. The first video showed a man with a gun walking towards a woman in a baseball field. When the man was close to the woman the woman gave him some money. The second video showed the same woman and the same man walking towards the woman and again the woman gave money to the man, like in the previous video. The difference in the second video was that the scene took place in a shooting range. Researchers found that participants gave better descriptions of the man carrying the gun in the video that was filmed in the shooting range, compared to the video that was filmed at the baseball field. Researcher concluded that this was because the gun was out of context in the baseball field but was expected in the shooting range (Pickel, 1999). According to Pickel (1999), these results propose that memory can be influenced not only by the fear of an object but by the object being out of context. Davis, Smith & Blincoe (2007), suggest that because of these findings when conducting studies on the “weapon focus effect” researchers should take into consideration both factors. The study of the “weapon focus effect” in children is difficult because of ethical concerns. Even though experiments have been conducted using frightening objects, effort is made so that the experiments are conducted in a setting that is familiar to the children. In addition the children are always spectators (Davies, Smith, & Blincoe, 2007). According to Huston et al., (1992, as cited in Davies, Smith, & Blincoe, 2007), using children in experiments as spectators even when using frightening objects is justified, because in western societies children very often watch violence in television. Davies, Smith, & Blincoe, (2007), conducted an experiment with children 7 to 10 years of age. In the experiment a table was set with 14 items. The independent variables in the study were the age of the children and the type of item in the table. In the table that was set up for the control group one of the items was a fountain pen. The novelty group had a cellular phone and the experimental group had a syringe filled with red food dye. The dependent variables were the number of items that the children were able to remember and how much the children could remember about the appearance of the experimenter. The results showed, that children that were exposed to the syringe recalled fewer items, than the children exposed to the cellular phone or the fountain pen. In addition, they recalled less features from the experimenter. There was not a big difference between the children exposed to the cellular phone and the children exposed to the fountain pen. The authors interpret the results to be consistent with the arousal hypothesis. However, the novel effect could not be proved because the cellular phone was not necessarily out of context. The authors recommended that further studies were needed in which the objects used are clearly out of context and not frightening (Davies, Smith, & Blincoe, 2007). Previous studies have found that children develop schemas at an early age. In addition after observing relevant cues these schemas can be activated (Pickel, Narter, Jameson, & Lenhardt, 2007). The development of schemas is important, because this is what will trigger the effect of unusual or novel in a particular situation. Therefore, if children have developed these schemas as adults do, it will be easier to develop a study to observe the “weapon focus effect” in children. Pickel, Narter, Jameson, & Lenhardt (2007), conducted an experiment with groups of children between 4 and 8 years of age and a group of adults. The participants watched a videotape of a mother preparing a birthday party for her daughter. The video also shows a man that comes to the house and steals money from the mother. The man in the video that comes and takes money from the mother is the independent variable. There are several versions of the video. In one version the man is dressed as a chef and carries a chef’s knife. A chef’s knife in this condition is consistent with the schema. In another video, the man is dressed as a mail carrier and carries letters. In other versions of the video the man is dressed as a mail carrier but is carrying a chef’s knife, which is inconsistent with the schema. The authors found that across all ages, participants were able to remember less accurate information about the man when he was holding an object that was out of context. In addition, the author stated that “the results provide additional support for the hypothesis that the weapon focus effect occurs because weapons are unusual within many contexts” (p. 68). Even though the results of this study demonstrated that the “weapon focus effect” occurs in preschool and elementary children the, authors believe that the same effects should be found in children of all ages. The visual scene processing model described by Henderson et al. (1999, as cited in Pickel, Narter, Jameson, & Lenhardt, 2007), states that a person viewing a scene will focus on the objects of the scene in order to find an overall schema. Therefore, when the eye sees an object that is out of context, it is natural that it will stay longer on that object to try to understand what is it doing in that particular scene. When the objects seem to belong to the scene, the viewer will not spend much time on them. Another important point is that the participants were questioned right after the tape was viewed. In real life, however, witnesses are questioned after some time, and they are also questioned several times. These two variables can be incorporated in a new research. These studies are conducted to better understand the causes of wrongful identification. Although these causes can not be manipulated by the criminal justice system, when investigators know the effects of a weapon on a witness, they can better choose a particular way of questioning, and also they will be better able to interpret the responses (Pickel, Narter, Jameson, & Lenhardt, 2008). Even though there has been some research done on the “weapon focus effect” in children, there are still variables that have not been manipulated that can give us more inside. Children and elderly often witness crimes; therefore, it will be valuable to observe their responses on the “weapon focus effect”. Since studies have demonstrated that the novel objects can show the “weapon focus effect” it is easier to develop experiments with children and the elderly because ethical concerns are easier to manage with novel items. We have determined that there have been more studies of the “weapon focus effect” in young adults and children than in older adults. We also have determined that older adults very often witness or are victims of crimes. In addition we also determined that adults can be good witnesses when they are questioned properly. For these reasons we believe it is important to study how the elderly respond to the “weapon focus effect”. In the current study we wanted to investigate the effects of a novel item in the elderly compared to the effects in children and young adults. We used three versions of a video in which the only difference was an item being manipulated. In one version a gun was used to produce the WFE, the gun is out of context in the scene and at the same time can be viewed as threatened. In a second video, we used an object that is out of context but that would not be seen as threatened. Finally, in the third video the neutral object was showed, in this video the object was not threatened or out of context. There are limitations in our present study, because the participants are watching a video and not experiencing personally the effects of the weapon or the unusualness of the object. But as we sawed in previous studies the WFE can be obtained using video tape scenarios as opposed to real situations. In addition because of ethical concerns the used of a video was more appropriate. After watching the videos the participants were asked to remember details about the target person, in each video. Based on the items recalled conclusions were drawn. We hypothesized that in the novel or gun condition, the elderly (ages 60-70) would recall less details about a target compared to young adults (ages 20-40), followed by the children (ages 8-12) compared to young adults (ages 20-40). In the control condition we predicted that even thought there was a difference in items recalled, between the age groups, the difference was not as significant as in the other conditions. We predicted that the older adults recalled fewer details than children and young adults; because we believed the elderly would take more time focusing on the item that is out of context, in an effort to make sense out of it.
Method
Design Overview Our research was designed to measure the impact of the “weapon focus effect” in children, adults and the elderly. The WFE states that when a gun is present during a crime, the witness or the victim of the crime will pay less attention to the perpetrator holding the gun, because the witness is focusing his/her attention on the weapon. Previous studies have shown that the WFE can be reproduced by using a weapon, or a novelty item. Previous research has also demonstrated that novelty produces the WFE, because when an object is present in a situation in which the object is not expected the witness will focus his/her attention in the novel item, trying to make sense of the object, therefore creating the WFE. Previous studies have also shown that the WFE can also be reproduced using a video as opposed to real scenes this is important, because, a weapon in a real situation will have high ethical concerns, especially when children are involved. We wanted to compare the “weapon focus effect” in the elderly with adults and children, because we did not find any research involving the elderly and the “weapon focus effect”. The elderly are often witnesses and/or victims of crimes; therefore understanding how the elderly react to the WFE will be beneficial when investigating a crime. In addition it will help improve the way the witnesses are questioned in order to get more accurate information. Having more accurate information and knowing how much weight to put in the witness testimony will also help reduce wrongful convictions. We believe, that the “weapon focus effect” will have more impact in the amount of correct items recalled about the perpetrator’s physical appearance , in the elderly (ages 60- 70), compared to young adults (ages 20- 40), but it will have less impact on memory in children (ages 8-12), compared to young adults (ages 20-40). In this study, we randomly assigned participants to watch one of three videos. The three videos were the same, with the exception of an object being manipulated, to produce the WFE. In one video the object was a gun, in a second video the item was a novel object (novel is defined here as an object out of context), and in the third video the object was neutral in the sense that it was not out of context and was not a threatening object. Conclusions were based on how much accurate detail participants were able to recall about the physical appearance of the target person. The target person is the person that was handling the object being manipulated, that is the weapon, the novel item or the neutral item. Results from each video, were compared across each age group and item manipulated.
Participants
Our study consisted of 180 participants. There were 60 children (ages 8-12), 60 young adults (ages 20-40) and 60 older adults (ages 60-70). Male and female participants were divided equally in each category. We used Florida International University (FIU) as a source of participants. The children were recruited from summer camp. The young adults were recruited from the student population and the older adults were recruited from the staff member population. We are aware that the population used to select the participants is not a true representation of the complete population that is exposed to the WFE, because population from FIU will be more educated and perhaps more aware of what the WFE is. In addition FIU population may be mostly middle class. We used this population because we believed that the results will still be valuable in understanding the effects of the WFE. There were 40% Hispanic, 20% White Caucasian, 30% African American and 10% Asian. No compensation of any kind was given to the participants; therefore participation was voluntary. The participant’s demographic information was kept separate from their answers about the video, in an effort to keep participation anonymous.
Materials
Control Condition. Video 1 (2 min): The video showed a couple with two children, a boy and a girl, on a picnic blanket on the grass at the park. The video started by showing a ball on one side next to the boy, a couple of books next to the father, and a doll next to the girl. Then the camera focused on some crayons that were in the middle of the boy and girl. The next scene showed the boy and girl each coloring with the crayons and the father reading. These scenes were done to establish the context of a happy family having a good time in the park. The mother (target) was not showed in the previous scenes. The last scene, which only lasts 20 seconds, showed the mother (target) opening a picnic basket, taking out a sandwich first, and then a loaf of bread from the basket. Gun Condition. Video 2 (2 min): The same video1 was used for this condition. The only change was that the last scene (when the mother takes out a loaf of bread) was edited and instead of taking out a loaf of bread the target takes out a gun.
A pilot study was conducted in which the target is shown taking out different objects that were considered novel (out of context). Participants were asked to rate the objects using a 9 point scale. They were asked to rate how unusual the object that was taken out of the picnic basket was. Based on these results we concluded that a wig was considered to be out of context in the picnic scenario, and therefore considered a novel object. Novel Condition. Video 3 (2 min): The same video 1 was used for this condition. The only change was that the last scene (when the mother takes out a loaf of bread) was edited and instead of taking out a loaf of bread the target takes out a blond long hair wig.
Procedure
The study was conducted in a classroom environment. Participants were asked to fill out a consent form which explained their rights and possible harm. In the case of children, the parents were contacted when they came to register the children, for summer camp. A researcher explained the experiment to the parents and asked the parents if the children could participate. After the parents verbal consent, they were asked to fill out a consent form. The children were also asked to fill out a consent form. Participants were randomly divided within their own age category in groups of 20. Participants were also assigned randomly to one of the 3 videos. Each participant watched the video on a separate computer and separate from the other participants. In total there were 9 groups each consisting of 20 participants. There were 9 sessions in total. Three groups of 20 participants were exposed to a video with a novel object. The other three groups of 20 participants were exposed to a video with a gun. The last three groups of 20 participants were exposed to a video with an object that was neutral. The dependent variable was how much information the participants were able to recall about the mother (target). Participants received a questionnaire after the video session to evaluate how much information they were able to remember about the mother (target). The questionnaire included questions about the physical characteristics of the mother, (eye color, hair color, hair length, skin color, weight, and clothing) and her clothing (blouse, skirt, pants, shorts, eyeglasses and jewelry). There were many open ended questions like what was the mother (target) doing? Other questions were partially open-ended questions like from the choices below circle the one that best describes the color of the woman’s hair. Participants were also asked if they remembered the object that was taken out of the basket by the mother (target), if participants answered “yes” they were asked to describe it. The questionnaire consisted of 15 questions. The last three questions asked participants for their demographical information. Once the questionnaire was completed participants were asked to identify the mother (target) in a target-present photo line-up using 5 photos, the same procedure used by Pickel, 1998. The females used in the pictures had similar height, weight, hair color and eye color and physical features in general. Participants were told that the target might not be present in the set of photos presented. The photos were arranged randomly in a row and presented on a computer. Participants were instructed to either choose one of the photos or specify that it was not present, if they did not think that the target was present in one of the photos. After the participants made the selection they were asked to rate how confident they were about their selection using a 7-point scale to rate their confidence. At the end of the session the participants were thanked and debriefed.
Results
We predicted that in the novel and gun condition, the elderly (ages 60-70) recalled less details about the target (mother) compared to adults (ages 20-40), followed by the children (ages 8-12) compared to adults (ages 20-40). In the control condition we predicted that even though there were differences in the items recalled between the three groups, in general, the three groups recalled more details that in the gun and novel condition. We expected that all the groups, regardless of age, that watched the video in which the mother (target) took out the gun, would remember fewer details about the target than the groups that watch the video with the neutral object. We were not sure, however, about the results of the video in which a wig is taken out of the food basket. We predicted that because it is unexpected to take a wig out of a food basket, that the amount of details remembered would be similar to the gun condition. The questionnaires were evaluated by two separate researchers. The number of items recalled about the target (mother), and the object being taken out of the basket, was calculated by age and type of object. In addition, details recalled about other peripheral items in the video were evaluated. The questionnaires were divided by age and object (novel, gun, and neutral) and used to determine how many items were accurately recalled. These results allowed us to determine the effect of the object in each age group. Our study is a cross-sectional design because we are comparing three age groups. It is also Quasi-Experimental because one of the variables, age can not be manipulated. We used a Two-Factor 3 by 3, Between-Subjects ANOVA. Two-Factor because we had two variables: Age, [Children (8-12), Adults (20-40) and Elderly (60-70)], and the object taken out of the basket (novel object, gun, neutral object). Between Subjects, because we randomly assigned the participants from each age group and each group was exposed to a different condition. Even though the within-subjects design increases internal validity, we chose the between-subjects design because the effects of carryover were too strong; because if we had exposed each group to all conditions, the participants would have had more opportunities to see the target and therefore not been distracted by the object being manipulated. Error variance was controlled by holding all the extraneous factors as constant as possible. This was accomplished by using the same video in all three conditions and editing just the item being manipulated. Also, participants were randomly assigned to the levels of the independent variable. Finally, we made sure that the weapon was seen, and perceived by the participants to be a weapon. Showing the weapon in a video is a weaker manipulation than showing the weapon in a real life scene. But because there were children (ages 8-12) in our study, in order to address ethical concerns we chose to show the gun in a video, also because of ethical concerns the gun was not being pointed at the participants.

References

Bartlett, J. C., & Memon, A. (2007). Eyewitness memory in young and older eyewitnesses. Mahwah, J, US: Lawrence Erlbaum Associates Publishers.
Davies, G. M., Smith, S., & Blincoe, C. (2008). A “weapon focus” effect in children. Psychology, Crime & Law, 14(1), 19-28.
Douglas, A. B., & Steblay, N. (2006). Memory distortion in eyewitnesses: A meta-analysis of the post-identification feedback effect. Applied Cognitive Psychology, 20(7), 859-869.
Bornstein, B. H. (1995). Memory processes in elderly eyewitnesses: What we know and what we don’t know. Behavioral Sciences and the Law, 13(3), 349-363.
Innocence Project (2009). Eyewitness Misidentification. Retrieve from http://www.innocenceproject.org/understand/Eyewitness-Misidentification.php
Loftus, E. F., Loftus, G. R., & Messo, J. (1987). Some facts about ‘weapon focus’. Law and Human Behavior, 11(1), 55-62.
Pickel, K. L. (1999). The influence of context on the ‘weapon focus’ effect. Law and Human Behavior, 23(3), 299-311.
Pickel, K. L. (1998). Unusualness and threat as possible causes of ‘weapon’ focus. Memory, 6(3), 277-295.
Pickel, K. L., Narter, D. B., Jameson, M. M., & Lenhardt, T., T. (2008). The weapon focus effect in child eyewitness. Crime & Law, 14(1), 61-72.
Pozzulo, J. D., & Lindsay, R. C. L. (1998). Identification accuracy of children versus adults: A meta-analysis. Law and Human Behavior, 22(5), 549-570 Yarmey, A. D., & Kent, J. (1980). Eyewitness identification by elderly and young adults. Law and Human Behavior, 4(4), 359-371.