Sex differences in eyewitness memory

Evidence

• In 1978, a study was conducted to test the hypothesis that witnessing an event with high emotionality would be associated with poorer recall of the event. Subjects were shown one of two black-and-white videotapes: one showed a violent incident while the other showed a similar, albeit nonviolent, incident. Both males and females displayed poorer recall of the violent condition than the nonviolent condition. Females also showed poorer recall than males for the violent condition.
• In another study, participants were shown a film in which a man was shot and robbed in an otherwise peaceful park. Participants’ accuracy and quantity of recall were assessed when questioned either immediately after viewing the film or one week later. Results showed that females were significantly more accurate than males, although there was no gender difference in quantity of information recalled.
• In a series of two experiments, researchers assessed gender differences both in recall accuracy and susceptibility to incorrect suggestions about witnessed events. In the first of the two experiments, subjects were shown a sequence of 24 slides in which a man stole a wallet from a woman’s purse. Participants were then given a questionnaire to assess accuracy, given a short break during which they were asked to complete a simple color-naming task, and then asked to read a paragraph describing the incident. Half of the participants read paragraphs in which four critical points were incorrect. They then took a final accuracy test. Significant gender differences in accuracy were found concerning the type of information requested. Females were more accurate in reporting details of the woman’s clothing and actions, while males were more accurate in reporting details of the man’s appearance and general surroundings of the event. It was also found that females were more influenced by the incorrect information than males; however, it was later noted that of the four critical items falsified, three were male-oriented details more concerned with the surroundings and less concerned with descriptions of the victim or thief. In order to further assess these gender differences and re-examine suggestibility without this potential confound, a second experiment was devised.
• In the second experiment, subjects were shown a sequence of slides that differed from those used in the first experiment; in this set of slides, a man and woman witnessed a fight, and the man intervened while the woman called for help. The same procedure was used, with the exception that the misleading paragraphs were read one day after viewing the slide sequence. Results confirmed the original finding that females were more accurate for female-oriented items and males were more accurate for male-oriented items. Females were more susceptible to incorrect suggestions concerning male-oriented items and the opposite was true for males. Overall, male and females differed in accuracy of eyewitness memory depending upon the type of information requested.
• Another study was conducted on undergraduate students in an introductory psychology course. During class, one male and one female confederate were brought into the classroom to make an announcement of a study that would be conducted in class the following day. The male made the initial announcement while the female student stood next to him and added a small reminder at the end. The next day, students in the class answered a questionnaire concerning both the male and female seen the day before, as well as objects and actions associated with the event. Overall, females were more reliable than males concerning questions about the female confederate. There was no gender difference found in recall accuracy for details concerning the male or objects and actions related to the event.
• In another study, subjects were shown a two-minute film in which a woman was approached by a man who first asked her for money and upon her refusal assaulted her, stealing her purse and running away. One week later, participants filled out a checklist concerning various aspects of the event. Results showed that females accurately recalled more details describing the victim than males did. Because the victim was female, this finding shows support for other studies that found females are more accurate in recall when asked about female-oriented details. The study did not assess gender differences for the attacker. Results also showed that females were better than males in setting descriptions. Such gender differences in eyewitness performance have also been reported in Indian settings.

  Counter-evidence

• A 1997 study tested subjects on their interrogative recall and narrative recall regarding a young white woman with whom they had interacted around two minutes prior. Overall, men and women performed similarly in recalling most characteristics; however, men were more confident in their responses than women, and women overestimated the length of time that they interacted with the young woman. Both men and women were relatively accurate estimating duration when they used imagery rehearsal prior to giving estimates.
• Another study examined eyewitness behavior by presenting two dissimilar stimuli: a colored picture of a farmhouse kitchen and a watercolor-and-ink picture of a busy street crossing. After viewing each picture for 60 seconds, subjects answered 10 questions pertaining to each picture. Factual questions referred to information about true details in the picture, whereas leading questions referred to information about unconventional details in the picture. No significant gender differences were found in accuracy of recall or in resistance to misleading information.
• A third study showed students in an introductory psychology class a videotape of a simulated robbery under one of four experimental conditions. Afterwards, students completed the Personality Adjective Inventory and responded to a questionnaire regarding details of the crime. Results showed that the accuracy was not significantly influenced by the victim's or the subject's gender.