Deese–Roediger–McDermott paradigm

The Deese-Roediger-McDermott (DRM) paradigm^[1][2] in cognitive psychology is an example of false memory.. The DRM Paradigm refers to the tendency to falsely recall a target word from a set list of words centered around that target word. This paradigm was first studied in 1959^[1] and has been repeated in more recent years to confirm this phenomenon.^[2] More research has also been performed to show that falsely remembered memories, also known in psychology as confabulation,^[3] extends to sentences, sequences, videos, and entire events.^[4][2] The DRM Paradigm has provided much information on false recall and the types of memory errors commonly made.^[5]

Primary Work

Frederic Bartlett (1932) is credited with the first experiment on false memories.^[4] Bartlett had subjects read an Indian folktale and then recall the tale repeatedly. The results of his experiments, although he reported no aggregate data, were that there were distortions in the subjects’ memories over repeated attempts. His data was never repeated and was actually disproved by Wheeler and Roediger (1992) when their repeated tests showed improvement. Wheeler and Roediger (1992) showed that three tests taken in succession after studying pictures drastically improved retention on relative to taking the test once or taking no test.^[6] Although Bartlett’s data was never proven in experiments; there have been multiple experiments that follow his lead. Roediger and McDermott followed the lead of these original studies like one performed by Bransford and Franks (1971,) who discovered false memories were obtained in sentence sequences.^[7]

Deese

James Deese performed many studies on recall and was interested in the effects that associative factors would have on recall. While he was at Johns Hopkins University, Deese performed a study to examine the "occurence of verbal intrusions in immediate recall."^[1]. In this study, Deese used lists consisting of 12 words each and presented them to 50 students at John Hopkins for a test of immediate recall. During the tests, certain words occurred as intrusions varying from as low as 0% to 44%. Deese gathered data on word associates by giving participants a free association test.^[1] Participants were given lists of words and were asked to write next to each given word the first word that came to mind. Based on these word-association frequency tests, he was able to hypothesize that the probability of a particular word to be falsely recalled as an intrusion was determined by the average amount of times the word occurs in association with the larger group of words in each of these 12 lists.^[1]

After gathering data on word associates, he gave different participants lists of 12 words which were highly associated with a critical word. For example, a sample list may have been: bed, slumber, pillow, nap, siesta, rest, tired, dream, awake, yawn, drowsy, and snore.^[8] All of these words are semantically related to the word sleep; however, sleep is not included in the list. Deese found that 44% of participants included sleep, an extra-list intrusion, in their immediate free recall.^[1] This was repeated for many critical words, and a strong positive correlation (R = .873) was found for the relationship of “average percent occurrence of word as associate to words on list” to “percent occurrence of word in recall” (i.e. percent of times sleep was chosen as first word to come to mind for the list words against the percent of times sleep was falsely recalled).^[1]

Roediger and McDermott

Experiment One

Roediger and McDermott repeated Deese’s experiment over 30 years later. For Roediger and McDermott's experiment; the experiment was broken down into two parts.^[9] The first experiment was made to replicate Deese’s (1959) results, which was people having false recall for words that were critically related to a list but not actually presented.^[1] They did this by producing six lists of 12 words from Deese’s experiment. They restricted their study to the six targets that produced the highest level of erroneous recall from Deese’s experiment.

They formed word lists for each target word from the first 12 associates listed in Russell and Jenkin’s (1954)^[10] word association norms. Subjects were tested in groups and were instructed that they would hear the list of words and after each list would have to write down the last few items first (standard instruction for this task). Then they were given two and a half minutes to recall all of the words in order, and to rate their confidence for each word. The subjects then read the next lists (there were 6 total) and then had a 2-3 minute conversation. After the conversation, participants rated their confidence for each word being on the original list. The experimenter read the six target words and the subjects were asked to raise their hands if they thought these six words were on the list. Most subjects raised their hand and were then informed that none of the read aloud words were on the lists.^[2]

The results for experiment one were exactly the same as Deese's study.^[2] People recalled the non-present target words almost as much as words actually presented. Participants rated the target words at the highest confidence 58% of the time, which means that they felt the target words had been studied over half the time.

Experiment Two: Purpose

In the second experiment, the goal was to expand on the results found by Deese and confirmed in their first experiment. The wanted to determine the false-alarm rates for the target words when the related list had not been presented, and to obtain the participants judgments for the experiment while recognizing non-presented words.

Procedure

For this experiment, they expanded the number of target words to 24; each target words had 15 word-associates.^[2] The subjects studied a certain number of lists determined by the experimenter; but all 24 lists were on a later recognition test where subjects were asked to determine if they had studied the list or not. When subjects said it was present, they were asked to judge if they remembered the item from the list or if they simply knew it occurred. The recognition test occurred 5 minutes after this task. Subjects were then told that they would see a long list of words, some of which would be from the beginning part of the experiment, and would have to say if those words were new or from the old lists. If they said the words were old, then they had to say if they remembered it being old or knew it was old. The recognition test was 96 items, 46 old and 46 new. After the recognition test, experimenters asked if the subjects knew what the experiment was about. Most said it was about memorizing lists of words.^[2]

Results

The results from the second experiment were ground breaking as subjects recalled the target words over half of the time, which was higher than Deese’s results.^[11] In fact, people remembered non presented words more than presented ones. The results for recognition show that the act of recall enhanced later recognition. For the remember/know judgments, 72% of the participants were likely to remember the target word (the words not on the list) when the words associated with that target word were presented and correctly recalled by the participant.^[2] When the word associates were presented but not recalled, the participants remembered the target word over half of the time, which is still very high. The recognition results for falsely recalled words closely resembled those for correctly recalled, studied items. The probability of recognizing falsely recalled items was extremely high (.93), and most of these items were judged by participants to be remembered (.73) rather than known (.20).^[2] Words that were falsely recognized were judged to be remembered in 58% of the cases, which is close to the same rate for words that were studied and actually on the list (52%). These results greatly show how false memory can be drastically affected. People remember words that were not shown to them just as much (or more) than words actually presented to them.

Further Work

Roediger and McDermott found, along with other psychologists, that even when the participants are warned about the illusion of false recognition, they still produce the critical word.^[12][13] McDermott performed a separate study where she found that in some cases, participants were more likely to recall the words not on the list than those on the lists.^[14] After these Roediger and McDermott experiments, the above results are now explained as the Deese-Roediger-McDermott (DRM) paradigm. The DRM paradigm and accompanying procedure gave results which can be summarized simply: people remember events that never happened.

The DRM Paradigm has been revealed to apply to images as well as words. Subjects remembered false images as memories where they reported having pictures of items that were presented as words.^[15] Names can also be falsely remembered with specific critical lures towards specific names.^[16] However, there is also evidence that people strategically avoided falsely reporting (not falsely recalling) names without even being told about the DRM Paradigm and its effects on false recall.^[16]

A study by Otgaar et al. (2011) showed that the participants attention is divided during encoding, effects are seen only in the false memory, not in the true memory of the participants.^[17] This study also showed that dividing attention reduces the susceptibility of children to the DRM Paradigm while it increases the susceptibility of adults.^[17] Other studies show that individual differences appear in the tendency to falsely recall target words.^[18]

In a study done by Van Damme et. al.^[19] (2010) the same procedure was followed as Roediger-McDermott. However, The experimenters wanted to see what affect timing of the test had on the DRM Paradigm. The test for recall of words either occurred immediately or seven days after the study part of the experiment. The experiment measured how accurate the subjects were during the test and found no difference in the Deese-Roediger-McDermott Paradigm for the two differently timed tests. This experiment showed that regardless of the timing of the test, results of the Deese-Roediger-McDermott Paradigm will stay the same.

Fabiani et. al (2000) found that although people often cannot distinguish between true or false memories in terms of conscious decisions, like a recognition judgment, brain functions reveal activity in separate parts of the brain for true and false memories. ^[20]

The question of why the DRM paradigm still remains. Semantic memory networks cause priming, and this priming summates. In the above experiments, as each word is presented, the associated critical word is activated, and increasingly so with each successive associated word.^[21] The above explanation is known as activation monitoring theory. Semantic information can override more tangible information, such as auditory or visual information which was how the words were presented.^[22]. There are still experiments being done on the DRM Paradigm. All the above experiments, and many others, attempt to expand and explain even further the paradigm that Deese discovered and Roediger-McDermott elaborated upon so well.

References

1. ^ ^{a b c d e f g h} Deese, J. (1959). "On the prediction of occurrence of particular verbal intrusions in immediate recall". Journal of Experimental Psychology, 58, 17–22
2. ^ ^{a b c d e f g h i} Roediger, H. L., III, & McDermott, K. B. (1995). "Creating false memories: Remembering words not presented in lists". Journal of Experimental Psychology: Learning, Memory, & Cognition, 21, 803–814
3. ^ Berrios, German E. (1 December 1998). "Confabulations: A Conceptual History". Journal of the History of the Neurosciences 7 (3): 225–241. doi:10.1076/jhin.7.3.225.1855. 
4. ^ ^{a b} Balota, D. A., & Marsh, E. J. (2004). Key readings in cognition: Cognitive psychology. New York, NY: Psychology Press.
5. ^ Koriat, A., Pansky, A., & Goldsmith, M. (2011). An output-bound perspective on false memories: The case of the Deese-Roediger-McDermott (DRM) paradigm. In A. S. Benjamin, A. S. Benjamin (Eds.) , Successful remembering and successful forgetting: A festschrift in honor of Robert A. Bjork (pp. 297-328). New York, NY US: Psychology Press.
6. ^ Roediger III, H. L., & Karpicke, J. D. (2006). Test-enhanced learning: Taking memory tests improves long-term retention. Association for Psychological Science, 17(3).
7. ^ Brainerd, C. J., & Reyna, V. F. (2005). The science of false memory. New York, NY: Oxford University Press.
8. ^ Rosanoff, A. (Ed.). (1929). Free association test. (Kent-Rosanoff.) (Repr. from Manual of Psychiatry). Oxford England: Wiley. Retrieved from EBSCOhost.
9. ^ Koriat, A., Pansky, A., & Goldsmith, M. (2011). An output-bound perspective on false memories: The case of the Deese-Roediger-McDermott (DRM) paradigm. In A. S. Benjamin, A. S. Benjamin (Eds.) , Successful remembering and successful forgetting: A festschrift in honor of Robert A. Bjork (pp. 297-328). New York, NY US: Psychology Press.
10. ^ Saltz, E. (1961). The effect of induced stress on free associations. The Journal of Abnormal and Social Psychology, 61(1), 161-164. doi: 10.1037/h0043312
11. ^ Miller, M. B., Guerin, S. A., & Wolford, G. L. (2011). The strategic nature of false recognition in the DRM paradigm. Journal Of Experimental Psychology: Learning, Memory, And Cognition, 37(5), 1228-1235. doi:10.1037/a0024539
12. ^ Gallo, D. A., Roberts, M. J., & Seamon, J. G. (1997). Remembering words not presented in lists: Can we avoid creating false memories? Psychonomic Bulletin & Review, 4, 271-276.
13. ^ McDermott, K. B., & Roediger, H. L., III (1998). False recognition of associates can be resistant to an explicit warning to subjects and an im- mediate recognition probe. Journal of Memory & Language, 39,508- 520.
14. ^ McDermott, K. B. (1996). The persistence of false memories in list re- call. Journal of Memory & Language, 35,212-230.
15. ^ Foley, M., & Foy, J. (2008). Pictorial encoding effects and memory confusions in the Deese–Roediger–McDermott paradigm: Evidence for the activation of spontaneous imagery. Memory, 16(7), 712-727. doi:10.1080/09658210802220054
16. ^ ^{a b} Mukai, A. (2005). Awareness of the false memory manipulation and false recall for people's names as critical lures in the Deese-Roediger-McDermott paradigm. Perceptual And Motor Skills, 101(2), 546-560. doi:10.2466/PMS.101.6.546-560
17. ^ ^{a b} Otgaar, H., Peters, M., & Howe, M. L. (2011). Dividing attention lowers children's but increases adults' false memories. Journal Of Experimental Psychology: Learning, Memory, And Cognition, doi:10.1037/a0025160
18. ^ Watson, J. M., Bunting, M. F., Poole, B. J., & Conway, A. A. (2005). Individual Differences in Susceptibility to False Memory in the Deese-Roediger-McDermott Paradigm. Journal Of Experimental Psychology: Learning, Memory, And Cognition, 31(1), 76-85. doi:10.1037/0278-7393.31.1.76
19. ^ Van Damme, I., Menten, J., & d'Ydewalle, G. (2010). The effect of articulatory suppression on implicit and explicit false memory in the DRM paradigm. Memory, 18(8), 822-830. doi:10.1080/09658211.2010.509733
20. ^ Fabiani, M., Stadler, M. A., & Wessels, P. M. (2000). True but not false memories produce a sensory signature in human lateralized brain potentials. Journal of Cognitive Neuroscience, 12(6), 941-949. doi:10.1162/08989290051137486
21. ^ Roediger, H.L., Balota, D. A., & Watson, J.M. (2001). Spreading activation and arousal of false memories. In H. L. Roediger, J. S. Nairne, I. Neath, & A. M. Surprenant (Eds.), The nature of remembering: Essays in honor of Robert G. Crowder (p. 95-115). Washington, DC: American Psychological Association
22. ^ Mather, M., Henkel, L. A., & Johnson, M. K. (1997). Evaluating characteristics of false memories: Remember/know judgments and memory characteristics questionnaire compared. Memory & Cognition, 25(6), 826-837.