Running Head: FALSE MEMORY and FORGETTING

1

False Memory

Running Head: FALSE MEMORY AND FORGETTING

The Effect of Stressful Stimuli

on the Development of False Memory and Forgetting

[JHK1]Sarah Vogt

HanoverCollege

Abstract

This study addresses the effect a stressful stimulus has on one’s ability to correctly recognize details of an event. During the first part of the studyall participants (N = 13) completed either a difficult or easy dual-task condition and it was found that there was a significant difference between the difficulty level (defined as tracking error) of the two conditions t(11) = 25.7, p <.01. During the second part of the study the participant watched a video and was then given a written recognition-based memory test. From this test, the number of false memories and the number of forgotten details for each participant was recorded. Results indicated no significant difference between the non-stressful and stressful conditions for the number of false memories t(11) = -0.21, p = 0.84, and no significant difference between the two conditions for the number of forgotten detailst(11) = 1.04, p = 0.32. Although there is no significant difference between the two conditions for neither the number of false memories nor the number of forgotten details, the trend indicates that the experience of a stressful stimulus prior to viewing the video may lead to a decrease in the number of false memories.

The Effect of Stressful Stimuli on the Development of False Memory and Forgetting

According to Martin Conway (1997), the reason why humans are imprecise with regard to recalling events is because memories are not literal records of experiences; rather, they are interpretations of experiences. In other words, just as would be the case with individual interpretations of literature, there could be many different versions or accounts of an event. This implies that a memory may not necessarily be inaccurate if two people have different memories of a particular aspect of an event. For example, if one person would happen to remember a perpetrator’s shirt as blue and her hair as dark brown and curly when witnessing a crime, and another person happened to witness the shirt as green and the perpetrator’s hair also as dark brown and curly, this implies that the memory itself is inaccurate only because the initial interpretation of the perpetrator’s appearance was inaccurate. [JHK2]One witness’s experience, in this case, was that he or she preserved what was most important at the time of the event (in this case, the color of the perpetrator’s hair), and what was later more likely to be remembered.

Conway (1997) also talks not only about why memory may often be imprecise, but he also argues that the development of false memory plays a role in how we later recall a past interpretation of an event. Conway (1997) draws the distinction between two types of memory awareness where one in particular is more susceptible to result in false memory. One type of memory awareness—autonoetic consciousness—features the conscious recollection of a previously studied item or experience and involves detailed recall of previous thoughts, feelings, and sensory-perceptual experiences. In some cases, however, when the item or experience is not studied in great detail, it is thought to be a type of noetic consciousness where there may be feelings of familiarity with an event, but there is less of a sensory-perceptual experience and less vivid mental imagery available due to the fact that those details to be recalled were not important to the observer at that particular time. It is during this noetic consciousness that false memories normally occur.

According to Conway (1997), a false [JHK3]memory is a memory of an experience that never occurred. These experiences may range from something as concrete as a memory of a specific object, or even as dramatic as a feeling of anxiety. But why do false memories occur? First, according to Conway (1997), even when a memory is thought to be fully extracted from long term memory, it is never completely accurate, but instead faithfully represents what has been kept in long term memory. False memories, then, occur because of the human mind’s attempt to make sense of fragmentary information or experiences. [JHK4]Even if the memory faithfully represents what has been stored in long term memory, it can still be inaccurate because it may not be an accurate representation of reality based on the initial interpretation.

As well, Conway (1997) argues, false memories can be attributed to a combination of actual feelings at the time of the event and other external experiences that “fill in the gaps,” making it difficult to distinguish between what is real and what is not. When experiences and other cues are inaccurately confused with the event itself, one tends to believe these false experiences because they create a sense of familiarity, and, in turn, helps the individual to organize and make sense of the event when attempting to recall the pertinent information.

Not only is false memory alone important for understanding how we later recall a past interpretation of an event, but past research has shown that stress or anxiety experienced can have an impact on the number of false memories one generates based on that event (Roberts, 2002; Payne, Nadel, Allen, Thomas, Jacobs,2002). Subjects in one such study studied a series of pictures and a series of words and then attempted to recall all the pictures presented, at a low-stress time and then later at a high-stress time. False memories in this study were defined as the number of intrusions from words at the time of recall. It was found that subjects who experienced more vivid imagery were more vulnerable to falsememories when they were stressed (Roberts, 2002). In another study, subjects experienced either an experimentally induced stress or no stress condition and were presented with both related words and non-related “lure” words. Results indicated that in the stress condition, more false memories appeared with related words when asked to recall whether the word had appeared (Payne, Nadel, Allen, Thomas, Jacobs,2002).

There has been conflicting findings, however, in past research on the effect of stress on forgetting. While some research has shown that stress has had either no effects or interfering effects on memory (Stawski, Sliwinski, Smyth, 2006), other research has shown that stressful or anxiety-provoking stimuli does have an impact on the amount of missing information when attempting to recall an event (Goodman, Hirschman, Hepps, Rudy, 1991). In particular, results from this study indicate that an increase in stress may actually lead to higher recall ability.

But why does this all matter? Evidence shows that humans’ inability to recall pertinent information directly implicates that one should be increasingly wary of eyewitness testimony in our justice system (Wells, 2003). Although one’s ability to identify the defendant as the perpetrator is a strong indicator of guilt to a jury, errors in eyewitness identification remain a far too extreme commonality in our justice system today. According to Wells (2003), observers, namely jurors, have a hard time distinguishing between accurate eyewitness testimony and testimony that is inaccurate. After all, it is hard not to believe a witness when the defendant is right in the court room in front of the witness stand. One study that tested judgment of accuracy involved a cross-examination of eyewitnesses to crimes that were staged. Upon asking subject jurors whether they believed the witnesses made accurate or false identifications, results have shown that these subject-jurors have had little to no ability to discriminate between what is a real identification and what is a mistaken one. In fact, these jurors tended to overestimate the accuracy of the eyewitness. If this is true, then it is easy to see how much of an effect eyewitness testimony can have on the jury’s decision-making process, and how much the possibility of erroneous or false memory can be overlooked (Lindsey, et al., 1989; Wells, 2003).

The present study focuses on the effect that experimentally-induced stress has on one’s ability to recall and recognize information. This ability was measured in terms of the number of false memories generated after watching a video clip as well as the number of forgotten details. It is hypothesized that an increase in stress level just prior to a memory task will result in an increase in the average number of both false memories and forgotten details.

[JHK5]Method

Participants

There were 13 total participants in this experiment, all of whom were Caucasian students from a small Midwestern liberal arts college, aged 18-22. Participation was voluntary, although one participant received credit for an introductory psychology class in return for her participation. Participants included three men and ten women, all with normal color vision and either normal or corrected acuity, based on self-report.

Equipment

This experiment was run on Windows XP using a 15-inch diagonal LCD monitor,model number FPD1565. The resolution was set at 1024 x 780. The monitor was part of a Gateway brand computer with an Intel Pentium-4 Processor, model number E4300. The specific program used in this experiment was written in Java, and can be found in Cognition Laboratory Experiments (subcategory Dual-Task Experiment), a computer program designed by Dr. John Krantz[JHK6] for experiments related to dual-tasking. The participants use a mouse to track the dot and left-click the mouse when a particular target appears on the screen.

Stimuli

For the dual task experiment, most of the stimulus variables remained static throughout each condition. For the primary task of tracking the dot, both conditions (difficult/stressful and less difficult/non-stressful) were set to a dot size of 4 pixels with a range angle that varies up to 360 degrees with each update. All secondary task stimulus variables remained at the default setting for both conditions.

About half of the participants experienced the difficult/stressful condition (N = 7), while the other half of the participants experienced the less difficult/non-stressful condition (N = 6). The difficult/stressful condition was set to a smaller box size of 25 pixels and a faster dot speed of 25 pixels per update, while the less difficult/non-stressful condition was set to a larger box size of 50 pixels and a slow dot speed of 2 pixels per update.

As for the visual stimulus, the video clip ran on VLC media player on the same computer as the dual task experiment. It depicted a criminal assault of a woman by a male perpetrator and another female’s observation of the event. The video clip lasted for a total of one minute and four seconds, and the speaker volume was set to a medium, comfortable level for all participants[JHK7].

Procedure

Participants were randomly assigned to either the difficult/stressful condition or the less difficult/non-stressful condition. There were no practice trials for either of the conditions.

Before each of the ten trials, the fixation box appeared, and participants were instructed to track the dot using the mouse, with the goal of keeping the dot inside of the box as much as possible. In addition to this primary task, participants were instructed to left-click the mouse whenever an “X” appeared on the screen. The margin of error when tracking the dot was collected as well as the accuracy of performing the secondary task (correctly identifying the “X” when it appeared within the specified response window). Each trial duration lasted 20 secondsand the “X” appeared 50 percent of the time.

After the dual-task portion of the experiment, all participants viewed the video clip of the assault. Following the video, all participants were given the memory task, which was basically a test based on what had been seen in the video clip. The questionnaire was in a checklist format, broken down into nine sections: what items were seen, what was heard, what weapon was used (if any), where the crime took place, what the crime was, perpetrator’s appearance, victim’s appearance, and a section asking whether the perpetrator was caught in the end of the scene. The first dependent variable, the number of false memories, was measured as the total number of incorrect answers based on what actually happened or appeared in the clip. The second dependent variable, the number of forgotten details, was measured as the number of answer choices that were actually present that were not selected by the participant.

[JHK8]Results

In order to analyze the data, an independent t-test was run to try to find a significant difference between the stressful and non-stressful groups with regard to the number of false memories and forgotten details scored from the questionnaires. First, it was found that stress was, in fact, manipulated between the two conditions, defined as a significant difference between the difficulty level (tracking error) of the two conditions t(11) = 25.7, p <.01. There was no significant difference found between the stressful and non-stressful condition for the mean number of reported false memories t(11) = -0.21, p = 0.84. The mean number of false memories for the stressful condition was 5.71, while the mean number of false memories for the non-stressful condition was 6. There was also no significant difference found for the mean number of forgotten details t(11) = 1.04, p = 0.32. In this case, the mean number of forgotten details for the stressful condition was 12, while the mean for the non-stressful condition was 10.5.

Discussion

This experiment was designed to explore whether a change in stress level just prior to completing a memory task would have an impact on the number of false memories and the number of forgotten details reported by each participant. In particular, it was hypothesized that those in the difficult/stressful condition would have a greater number of false memories as well as a greater number of forgotten details. Although there was no significant difference between the two conditions found for false memories and for the number of forgotten details, there were more false memories reported on average in the non-stressful condition (M =[JHK9] 6), which goes against our hypothesis that stress may lead to an increase in the number of false memories. This result implies that stress does not necessarily have a negative influence on what humans are capable of remembering, and, in fact, may actually have a positive effect on the ability to remember pertinent information. However, participants in the stressful condition reported more forgotten details on average than in the non-stressful group. Therefore, although these results are not significant, those who experience stress may develop false memories less often, and, in fact, may just be remembering less in general due to the fact that they reported more forgotten details on average.

Although our hypotheses were refuted, there are still important implications for everyday life. One such obvious implication is the effect stress in general can have on the human mind, especially with attention processing. One law, the Yerkes-Dodson [JHK10]Law, basically claims that a greater amount of stress is not necessarily a bad thing, but can actually be helpful in moderate amounts. Once the stress level reaches a certain point, however, attention and other cognitive functions begin to decrease. An example of this would be taking a test. If a student pumps him- or herself up on caffeine before a test, a mixture of high-attentiveness and high-stress may cause the individual to perform poorly, whereas if stress had been present in moderate amounts, attentiveness would be at its peak and, therefore, may have led to much better performance on the test (Cahill, 2005).

So what does this imply about the result of the present study? Because the level of stress generated by the dual-task experiment was likely a little to moderate amount, attentiveness may actually have been at its peak, which may have led to better cognitive functioning in terms of the number of reported false memories for those in the difficult/stressful condition compared to those who were less stressed (and, therefore, less attentive) and in the less difficult condition. Had the stress been of a greater amount for the stressful condition, there may have been different results. As well, based on the results of the present study, it may very well be that stress has no effect on one’s ability to recall pertinent information.

There are also implications for the reliability of eyewitness identification. If conditions are particularly stressful at the time of questioning by authority, a witness may have a hard time remembering a perpetrator’s appearance if the stress level is too high or too low. Although stress is not likely to be the only factor affecting one’s ability to recall such important information, these results imply that an attempt to foster a moderately stressful environment may prove to be beneficial when questioning a witness.

There were, however, limitations to the present study. The participant may have experienced stress during the dual-task experiment, but it may be true that the participant was not experiencing any left over stress from that portion of the experiment. A way to remedy this would be to replace the secondary task of identifying the “X” with another secondary task of watching the video clip. That way, stress should have a direct impact on the ability to remember what went on in the video clip. As well, recorded answers may not have been the result of a false memory, but may have instead been a guess reported by the participant. Future research should also look into using an anxiety scale at different intervals during the course of the experiment to track the level of stress or anxiety that the participant is feeling to make sure that it is actually a factor in one’s ability to recall what had happened in the video clip.