False Memories - Case Study Example

Investigation on a factor that can contribute to false memories JULY (2014) Background The fragile nature of the memory has drawn the interest of many researchers in cognitive psychology for many years with a number of reports indicating that under certain conditions memories are not reliable. A good example is the evidence available indicating the susceptibility of human memory, where by suggesting to somebody that they have seen something, there may be a tendency of them remembering seeing the event despite the non-occurrence of the event. The human memory fallibility may be prone to problem because of it be often imperative to being able to make an accurate differentiation of events that actually took place and those that did not. It is a worthwhile course to venture into understanding the factors behind distortion of bearing in mind the how important eyewitness accounts are considered to be in judicial processes. There has been a lot of debate on mechanism of misinformation due to its robustness. According to Loftus et al.(1978), the original memory is believed to undergo permanent alteration or is “over-written” by subsequent information which is erroneous on the other hand Gardiner (1993) there is confusion of participants memory sources, where there is a tendency of remembering witnessing of an event without remembering where the event was seen, and as a result this leads to memory errors. There is a possibility that any of the given explanation may be correct, the accounting of the data may also be given through procedural data. The acceptance of any of the two given accounts difficult because of the post-event information being typically presented in a written form and then there being a recognition test which is written forced choice type. This makes it a possibility that contextual orthographic information which its presentation takes a narrative form is likely to make participants subsequently choosing the same response and the event of presenting the cues again. The possibility is likely with the previous reports indicating that there can be enhancement when there is a match between contextual cues at encoding stage and the cues at retrieval stage. A good example involves a study by Zaragoza (1994) where one of the two series of slides was shown to the subjects where there was depiction of college student with different objects of the same type being changed in some of the slides. In one of the slides a screwdriver would be shown with the other slide showing a wrench and in the audio narrative the objects were only referred to a “tool”. The second phase would involve subjects reading description of events in the slides, but this time specific tools being named with halt the time the tool being mentioned being incorrect. The third phase involved the subjects listing five examples of specific type of objects, with the condition that the listed examples were to exclude those seen in the slides. The subjects who had read the narrative which was incorrect had a lower chance of including the written objects (that they had not seen in the slides) as opposed to the control subjects with figures of 28% vs. 43%), and were also more likely to list incorrectly items which they had seen in the slides (33% vs. 26%). Misinformation is also manifested in the effects of leading questions where there is revelation of several ways in which there can be an influence on eyewitness reports. An example being where participants had watched a film of automobile accident. The participants were then asked about the speed of the cars at the point of smashing into each other. This elicited estimates of speed which were higher and also false claims of seeing broken glass on a later test when compared to questions where verbs such as hit or bumped were used as opposed to smashed (Loftus & Palmer, 1974). Subsequent studies also revealed that post event questions which were misleading had the potential of causing other distortions in the reports given by eyewitness. An example being study of Loftus (1977) that involved participants viewing an accidents that involved a green car and the participant then was subjected to questions which were misleading indicating that the car was actually was blue in colour. When later the participants were asked to make a selection of the colour of the car from a colour wheel, there was a marked tendency in the participants to make a shift in the direction of misinformation where there is selection of blue-green colour, and this tendency was not seen in the control participants. Thus the colour selection of the misled participants chose a colour that was a blend of the original and the colour given in postevent information. Main objective In this project there is investigation on a factor that can contribute to false memories. Hypothesis Hypothesis 1 H0:In Auditory testing main effect for information score is NOT significant H1In Auditory testing main effect for information score is significant Hypothesis 2 H0 The Narrative Modality X Information Type interaction is NOT significant H1 The Narrative Modality X Information Type interaction is significance Hypothesis 3 H0 In the text testing condition a main effect for Information Type is NOT significant H1 In the text testing condition a main effect for Information Type is significant Hypothesis 4 H0 Narrative Modality X Information Type interaction is NOT significant H1Narrative Modality X Information Type interaction is significant Materials A 7-minute video of a simulated office theft was made. Sound was not recorded. The target video was shot from a variety of camera angles and included scenes with both close-up and wide-shot footage. The video image was displayed on a 2 x 3m screen using a Prolite video projector. Ten target items/events were chosen from the video and these included both central and peripheral details. Two narratives were composed describing items and events that occurred in the video in chronological order. Each narrative was approximately 500 words in length. In each narrative, the 10 target items/events were randomly divided into two sets, and each item served in the misled and neutral conditions. For example, 5 of the 10 items were presented in the narrative such that they gave misleading information (e.g. reference was made to a brown jacket when it was in fact black), whereas the remaining 5 were neutral giving no descriptive information about the item or event (e.g. reference was made to a jacket but no information on its colour was given). The assignment of target items/events was counterbalanced across the two narratives such that each item served in the misleading and neutral conditions. That is, each critical item served as a neutral item for half the participants and as a misled item for the other half. Reference was also made in the narrative to other non-critical details presented in the video in order to reconstruct the entire series of events as they occurred, and so that participants did not focus only on the target items/events. Auditory narratives corresponding to the three text narratives were recorded using a male narrator. Each narrative was approximately three and a half minutes in duration and were played back using a compact disk. A forced-choice recognition test consisting of 10 questions relating to the target items was employed. A two-item forced choice recognition test was used. For example using the jacket critical item, the test question was ‘There was a jacket hanging over the back of the Doctor’s chair. What colour was it?’ with the forced choice response being either “black” or “brown”. The ordering of the questions was randomised to eliminate chronological memory effects, and two tests were used to control for response bias with the difference being the order of response options. For example, on the question relating to the colour of the jacket one test would give the option “black” or “brown” whereas on the other test it would be “brown Design Participants were allocated to either intervention group (misleading) or a control group (neutral). The between subjects variable was post event Narrative Modality (Auditory and Text) and the test type (Auditory Test and Text Test). The design is as summarized below. IV1 = Information Type (within-subjects) i. Neutral ii. Misleading IV2 = Post-event Narrative Modality (between-subjects) i. Auditory ii. Text IV3 = Test Type (between-subjects) i. Auditory Test ii. Text Test DV = Test items correctly recalled Procedure Participants were tested in groups of between 3 and 10 individuals in each session, and prior to testing they provided voluntary informed consent. Participants were randomly assigned to one of the narrative modality conditions (auditory vs. text) and test modality conditions (auditory vs. text) based on their order of arrival at the laboratory. Prior to experimental sessions, there was no warning that a memory test would follow, and no warning that any of the narratives contained misleading information. After viewing the video of the original event, participants completed an unrelated 10-minute visual filler task. At the conclusion of the filler task, the participants received one of the two counterbalanced versions of the narrative with half receiving the written narrative and half the auditory narrative. Depending on which testing session they were in, half of the text group received Narrative 1and half received Narrative 2. The participants in the text condition were asked to read the vignette once only at their own pace, whereas those in the auditory condition heard the story via the pre-recorded compact disk. The participants then completed a second 10-minute unrelated filler task after which they were given one of the counterbalanced recognition tests in either the text or auditory modality. The questions were presented in random order, such that they did not follow the chronological sequence of events shown in the original presentation. At the conclusion of the testing session participants were thanked, debriefed and released. Results In the auditory testing condition we observed a main effect for Information the scores for the control groups being high (M=4.17 SD=1.15) than for Misled group (intervention) (M=3.61; SD=1.20) for the auditory testing F (1, 34) =5.312 (p=0.027). This shows that the HO for hypothesis 1 is not true and should not be accepted but the alternative hypothesis H1 will be accepted. The Narrative Modality X Information Type interaction did not reach significance F (1, 34) =0.090 (p=0.766) and thus the HO for hypothesis 2 should be accepted. Table 1: Tests of Within-Subjects Effects for Auditory Test Source SS df MS F p Partial η2 InfoType 7.347 1 7.347 5.312 0.027 0.135 Modality x InfoType 0.125 1 0.125 0.090 0.766 0.003 Error 47.028 34 01.383 In the text testing condition a main effect for Information Type was observed, with Auditory being high (M =4.056 SD = 0.201) and text being lower (M=2.722; SD=0.297): F (1, 34) =16.010 (p=0.013). This shows that the HO for hypothesis 3 is not true and should not be accepted but the alternative hypothesis H1 will be accepted. And a Narrative Modality X Information Type interaction significant F (1, 34) =6.872 (p=0.013) this is indication that HO for hypothesis 4 is not true but the alternative hypothesis H1 will be accepted. Table 2 : Tests of Within-Subjects Effects for InfoType Source SS df MS F p Partial η2 InfoType 11.681 1 11.681 16.010 0.000 0.320 Modality x InfoType 5.014 1 5.014 6.872 0.013 0.168 Error 24.806 34 0.730 Discussion This study aimed at investigating the factors that can contribute to false memories. There were four hypotheses that guided the study. The first hypothesis was that in Auditory testing main effect for information score is significant and the second was that in Auditory testing main effect for information score is significant. The test result were in agreement with the first hypothesis while for the second hypothesis the test result was not in agreement since the null hypothesis was found to be true. The other hypothesis that were used were that In the text testing condition a main effect for Information Type is significant and Narrative Modality X Information Type interaction is significant where both were found to be true. 2. In the auditory testing condition we observed a main effect for Information Type, however the Narrative Modality X Information Type interaction did not reach significance. This results is in agreement with Zaragoza (1994) study where one of the two series of slides was shown to the subjects where there was depiction of college student with different objects of the same type being changed in some of the slides. The subjects who had read the narrative which was incorrect had a lower chance of including the written objects (that they had not seen in the slides) as opposed to the control subjects with figures of 28% vs. 43%), and were also more likely to list incorrectly items which they had seen in the slides (33% vs. 26%). This echoes with earlier suggestion that it a possibility that contextual orthographic information which its presentation takes a narrative form is likely to make participants subsequently choosing the same response and the event of presenting the cues again. The possibility is likely with the previous reports indicating that there can be enhancement when there is a match between contextual cues at encoding stage and the cues at retrieval stage. 3. In the text testing condition we observed a main effect for Information Type, and a Narrative Modality X Information Type interaction. The findings concurs with Loftus (1977) that involved participants viewing an accidents that involved a green car and the participant then was subjected to questions which were misleading indicating that the car was actually was blue in colour. When later the participants were asked to make a selection of the colour of the car from a colour wheel, there was a marked tendency in the participants to make a shift in the direction of misinformation where there is selection of blue-green colour, and this tendency was not seen in the control participants. Thus the colour selection of the misled participants chose a colour that was a blend of the original and the colour given in postevent information. 4. The results obtained in this study clearly points out the shortcomings that text testing presents in daily life situations. As a person encodes what is written in the test the response is likely to be affected by the ability to encode the text and relate it to the true scenario. 5. One of the limitations of this study is on the number of participants involved. In order to improve on the results, the number of participants is to be higher. References Drivdahl, S. B. (2001). The role of emotion and self-reference in the creation of false memories for suggested events. Dissertation Abstracts, 61, 6156. Drivdahl, S. B., & Zaragoza, M. S. (2001). The role of perceptual elaboration and individual differences in the creation of false memories for suggested events. Applied Cognitive Psychology, 15, 265–281. Gardiner, J. M., & Java, R. I. (1993). Recognition memory and awareness: An experiential approach. European Journal of Cognitive Psychology, 5, 337–346. Loftus, E. F.,&Palmer, J. E. (1974).Reconstruction of automobile destruction: An example of the interaction between language and memory. Journal of Verbal Learning and Verbal Behavior, 13, 585–589. Loftus, E. F.,et al. (1978). Semantic integration of verbal information into a visual memory. Journal of Experimental Psychology: Human Learning and Memory, 4, 19–31. Loftus, E. F. (1977). Shirting human color memory. Memory & Cognition, 5, 696–99.Loftus, E. F. (1979a). Eyewitness testimony. Cambridge, MA: Harvard University Press. Zaragoza, M.S. and Lane, S.M. 1994. Source misattributions and the suggestibility of eyewitness memory. J. Exp. Psychol. Learn. Mem.Cogn. 20: 934–945. Read More

The third phase involved the subjects listing five examples of specific type of objects, with the condition that the listed examples were to exclude those seen in the slides. The subjects who had read the narrative which was incorrect had a lower chance of including the written objects (that they had not seen in the slides) as opposed to the control subjects with figures of 28% vs. 43%), and were also more likely to list incorrectly items which they had seen in the slides (33% vs. 26%). Misinformation is also manifested in the effects of leading questions where there is revelation of several ways in which there can be an influence on eyewitness reports.

An example being where participants had watched a film of automobile accident. The participants were then asked about the speed of the cars at the point of smashing into each other. This elicited estimates of speed which were higher and also false claims of seeing broken glass on a later test when compared to questions where verbs such as hit or bumped were used as opposed to smashed (Loftus & Palmer, 1974). Subsequent studies also revealed that post event questions which were misleading had the potential of causing other distortions in the reports given by eyewitness.

An example being study of Loftus (1977) that involved participants viewing an accidents that involved a green car and the participant then was subjected to questions which were misleading indicating that the car was actually was blue in colour. When later the participants were asked to make a selection of the colour of the car from a colour wheel, there was a marked tendency in the participants to make a shift in the direction of misinformation where there is selection of blue-green colour, and this tendency was not seen in the control participants.

Thus the colour selection of the misled participants chose a colour that was a blend of the original and the colour given in postevent information. Main objective In this project there is investigation on a factor that can contribute to false memories. Hypothesis Hypothesis 1 H0:In Auditory testing main effect for information score is NOT significant H1In Auditory testing main effect for information score is significant Hypothesis 2 H0 The Narrative Modality X Information Type interaction is NOT significant H1 The Narrative Modality X Information Type interaction is significance Hypothesis 3 H0 In the text testing condition a main effect for Information Type is NOT significant H1 In the text testing condition a main effect for Information Type is significant Hypothesis 4 H0 Narrative Modality X Information Type interaction is NOT significant H1Narrative Modality X Information Type interaction is significant Materials A 7-minute video of a simulated office theft was made.

Sound was not recorded. The target video was shot from a variety of camera angles and included scenes with both close-up and wide-shot footage. The video image was displayed on a 2 x 3m screen using a Prolite video projector. Ten target items/events were chosen from the video and these included both central and peripheral details. Two narratives were composed describing items and events that occurred in the video in chronological order. Each narrative was approximately 500 words in length. In each narrative, the 10 target items/events were randomly divided into two sets, and each item served in the misled and neutral conditions.

For example, 5 of the 10 items were presented in the narrative such that they gave misleading information (e.g. reference was made to a brown jacket when it was in fact black), whereas the remaining 5 were neutral giving no descriptive information about the item or event (e.g. reference was made to a jacket but no information on its colour was given). The assignment of target items/events was counterbalanced across the two narratives such that each item served in the misleading and neutral conditions.

That is, each critical item served as a neutral item for half the participants and as a misled item for the other half.

Read More