Randomised Word Lists and their Relationship with False Memory

Abstract

Does unconscious awareness have any effect on false memory whatsoever? In this (within-subjects) study, 62 undergraduate psychology students were given a set of  8 lists to remember containing 12 words each. These words were associated with a critical word e.g. sleep. The lists were mixed together before participants learned them in order to test whether false memory is affected unconsciously. Other studies like in Roediger and McDermott’s, showed a high level of false memory as they did not mix up their list making it apparent to participants that the words were related to some sort of topic. The results within this study demonstrated false recall but produced lower false memory than Roediger and McDermott’s study implying their participants had a greater conscious awareness of the critical word thus, recalling the word on more occasions.

Introduction

False memory by simple definition is when memories which did not occur are recalled. This can have catastrophic consequences in certain situations. The idea of false memory can be abused by individuals who plant traumatic memories which did not even take place like rape or sexual abuse, in others (Very Well Mind). These memories can also be planted in adults as well as children despite never taking place (Ost et al).

Within Memory, there are two types of ways memory can be recalled. This is through either Reproductive Memory whereby simple items or numbers can be recalled through rote learning (Barlett,1932) or Reconstructive Memory where information can be falsely altered or added when trying to remember a complex event e.g. information about an earlier life event. The idea is that reproductive stimuli are either recalled or simply forgotten but with reconstructive memory, an individual can incorrectly conjure up a similar memory to try and make up for forgotten information. This has been demonstrated by Loftus and Palmer’s 1974 study where individuals who watched a video clip of a car accident incorrectly recalled broken glass at the site of the crash when asked about it. This provides support for the role of reconstructive memory within false memory demonstrating it is ever present in complex situations and is the cause of added incorrect information being produced.

However, Roediger and McDermott have shown that false memory is not just limited to reconstructive stimuli but is also applicable to reproductive stimuli. They done this by showing their participants 6 lists of words related to a critical word before asking them to recall them in a recognition test which contained the critical word and some unrelated words. Their experiment showed a high level of false memory as participants recalled the critical word 63% of the time in the recognition phase. (Roediger and McDermott). This prompts the question, were participants consciously aware of the critical word due to the grouping of the lists being strongly associated with general topics like sleep, mountain, needle etc.? Roediger and McDermott believed that each word involved in the topics sprouted an unconscious memory trace which eventually, led to the critical word.

In this study, the words in the lists presented to participants in the learning phase were completely mixed together rather than being presented one by one. This prevents the participants from uncovering the general theme or topic the words presented to them. This means if participants falsely recall the critical word in the recognition phase then this will be due to unconscious awareness as the lists are randomised. By doing the study this way, three hypotheses’ can be produced. The first is that the mixing up of these lists will not affect false memory in any way shape or form. The second is that the mixing up of the lists will eradicate false memory. The last hypothesis would be that the mixing up of the lists will influence false memory meaning the critical words will be recognized.

Method

Participants

The experiment consisted of 62 undergraduate psychology students aged 18-47 who took part in order to meet the course requirements for their PS114 Module. Participants were male and female.

Apparatus

Participants used a standard mouse to click on the stimuli presented to them on an Elite One computer with the software Inquist Web player being used.

Materials

The word lists within the study were generated from Roediger and McDermott’s appendix of their study ‘Creating False Memories: Remembering Words Not Presented In Lists’. In the study, 8 words were taken from the appendix and were labelled as ‘critical words.’ Of those 8 words, a list compromising of 12 associated words was created for each word. The lists were then randomly mixed together rather than being shown in sequence to the participants, and these were called the ’96 old words.’ The critical words were not shown in the learning phase. In the recognition phase, the 8 critical words (now labelled as the critical new words) which were not presented in the learning phase were now shown in the recognition phase alongside 40 old words with (five coming from each list) and 48 unrelated, ordinary new words. These words were then randomly mixed as well to produce another set of 96 words.

Design

The study involved a within-subjects design as all the students were exposed to the same conditions. The independent variable within the study was the type of words within the word lists and the dependent variable was the mean recognition score.

Procedure

Participants underwent two phases, with the first being the learning phase. In this phase, a total of 96 ‘old’ words appeared onto the Elite One computer screen one after the other. Each word was shown on screen for a total of two seconds. Participants were instructed to remember as many of the words shown to them as possible without reading them aloud, as they would have to recall some of them in the sequencing phase. After 32 words and 64 words, subjects were allocated a rest for 15 seconds. After this learning phase a filler task was introduced where mental arithmetic problems were shown on screen one after the other. Participants had to use their keyboard to write in the correct answer. The filler task lasted a duration of approximately two minutes and the aim of the task was to prevent the participants from mentally rehearsing the words they learned in the learning phase. Finally, in the recognition phase, participants were shown 40 of the 96 ‘old words’ on the Elite One screen one after the other which appeared in the learning phase. Additionally, 48 unrelated, ordinary new words were also shown one after the other and 8 critical new words which were not shown in the learning phase, were now shown to the participants. Participants then had to select a score on a scale of 1-5 with one being ‘definitely not’, 2 being ‘probably not’, 3 being ‘unsure’, 4 being ‘probably’ and 5 being ‘definitely sure’ that they recognized a word which they saw in the learning phase.

Results

Each participant’s recognition score for each type of word was averaged to give their mean recognition score. For example, this occurs by adding up each participants’ score which they gave on the scale from 1-5 for old words and then dividing by the total number of old words which are 40. The same rule would apply for the 48 unrelated, ordinary new words as well as the 8 critical words. This gives one participant their data. The groups data was collected by totalling all the participants’ individual mean recognition scores for old words, unrelated ordinary new words and critical new words, and dividing by 62 as that is the total number of participants in the study.

Table 1: Indicates the mean recognition scores for the three type of words and the standard deviation and standard error score.

Below, Graph A illustrates the mean recognition scores (displayed on vertical axis) for each type of word.

These results clearly demonstrate the presence of False Memory as participants produced on average a higher recall of the critical new words (3.43) than the unrelated, ordinary new words (2.11). This must mean participants thought they had seen the critical words in the learning phase, but this is due to them actually learning words associated with the critical words.

Discussion

Within this study, the results clearly showed that the old words were recalled the most on average (3.92) and the ordinary, unrelated new words recalled the least (2.11). However, false memory is apparent within this study as the mean number of times the critical word was recalled was 3.43 which is considerably close to the mean number of times the old words were recalled. These results support the hypothesis that the randomisation of the lists will have effect on false memory as the critical words were recognized. In Roediger and McDermott’s study, their participants achieved a mean score of 3.6 for the old words recalled and 3.3 for the critical words. This illustrates that within this study, false memory was found to be lower than what was found in Roediger and McDermott’s study in experiment 1. An explanation of this would be that participants were more consciously aware of the critical word as the lists were presented in sequence whereas in this study, the lists were jumbled up hence making recall of the critical word unconscious. However, in Roediger and McDermott’s experiment 2 they made their lists bigger and added more; making their false memory results were even greater. In fact, the critical words were recalled more or the same amount of times as the old words. Whilst Roediger and McDermott believe this is due to unconscious awareness, it is most likely not. This is because in our study, participants were unconsciously aware yet produced a smaller mean recall for critical words. Roediger and McDermott’s results are most likely due to them providing their participant with sequenced associated word lists which allows the participants to stay consciously aware of the critical word thus, explaining their incredibly high false memory results which are: 0.79 recall for old words and 0.81 recall for the critical words.

The results in this study provides insight into just how substantial false memory is. By using these results, people such as law enforcement officers can be cautious at how they phrase their questions in order to avoid the creation of false memories (Loftus and Palmer). This can help by putting the correct people in jail and keeping the innocent ones out of jail (Shaw).

A critique of this experiment may be that some of the psychology students were far older than others and with age, memory decreases. This may account for perhaps a greater recall for the critical words. This is evident as the oldest participant, aged 47 produced the highest number of recalls for the critical words (4.500) thus, age can be said to be a confounding variable.

References

• Loftus, E. and Palmer, J. (1974). Reconstruction of automobile destruction: An example of the interaction between language and memory. Journal of Verbal Learning and Verbal Behavior, 13(5), pp.585-589.
• Roediger, H., & McDermott, K. (1995). Creating false memories: Remembering words not presented in lists. Journal Of Experimental Psychology: Learning, Memory, And Cognition, 21(4), 803-814.
• Steffens, M., & Mecklenbräuker, S. (2007). False Memories. Zeitschrift Für Psychologie / Journal Of Psychology, 215(1), 12-24.
• False memories lead to false convictions. (2018). Retrieved from http://www.ucl.ac.uk/news/headlines/0917/220917-false-memories
• Bartlett, F. C. (1932). Remembering: A study in experimental and social psychology. Cambridge, England: Cambridge University Press.
• Leading Questions: How Interviewers’ Questioning Influences Eyewitness Testimonies. (2018). Retrieved from https://www.psychologistworld.com/memory/leading-questions-eyewitness-testimony
• What Impact Do False Memories Really Have?. (2018). Retrieved from https://www.verywellmind.com/the-consequences-of-false-memories-2795350
• Ost, J., Foster, S., Costall, A., & Bull, R. (2005). False reports of childhood events in appropriate interviews. Memory, 13(7), 700-710.