Critical Nonpresented Words Research Articles

The impact of fabrication on recognition memory: An experimental study

We examined whether fabrication affects memory using a new paradigm combining the Deese/Roediger-McDermott paradigm (DRM) and the Amsterdam Short Term Memory task (ASTM). Participants were assigned to either a forced fabrication or honest condition, and encoded emotionally-negative and neutral wordlists by reading words out loud. The wordlists contained words that were associatively related to each other (e.g., symphony, sound, piano, radio, sing, orchestra), and converged to a non-presented critical word (e.g., music). Next, all participants had to indicate two words that were previously read out loud. However, participants in the forced fabrication condition had to fabricate that they read two additional words out loud, which were associated to the same DRM list. Participants in the forced fabrication condition formed false memories for their self-generated fabrications. Although fabrication did not increase spontaneous false memory rates for non-presented critical words, our results did highlight the perils of forcing people to fabricate.

Novel Approaches and Cognitive Neuroscience Perspectives on False Memory and Deception.

The DRM (Deese–Roediger–McDermott) paradigm produces robust false memories of non-presented critical words. After studying a thematic word list (e.g., bed, rest, and pillow) participants falsely remember the critical item “sleep.” We report two false memory experiments. Study One introduces a novel use of the lexical decision task (LDT) to prime critical words. Participants see two letter-strings and make timed responses indicating whether they are both words. The word pairs Night-Bed and Dream-Thweeb both prime “sleep” but only one pair contains two words. Our primary purpose is to introduce this new methodology via two pilot experiments. The results, considered preliminary, are promising as they indicate that participants were as likely to recognize critical words (false memories) and presented words (true memories) just as when studying thematic lists. Study Two actually employs the standard DRM lists so that semantic priming is in play there as well. The second study, however, uses functional near-infrared spectroscopy (fNIRS) to measure activity in the prefrontal cortex during a DRM task which includes a deception phase where participants intentionally lie about critical lures. False and true memories occurred at high levels and activated many of the same brain regions but, compared to true memories, cortical activity was higher for false memories and lies. Accuracy findings are accompanied by confidence and reaction time results. Both investigations suggest that it is difficult to distinguish accurate from inaccurate memories. We explain results in terms of activation-monitoring theory and Fuzzy Trace Theory. We provide real world implications and suggest extending the present research to varying age groups and special populations. A nagging question has not been satisfactorily answered: Could neural pathways exist that signal the presence of false memories and lies? Answering this question will require imaging experiments that focus on regions of distinction such as the anterior prefrontal cortex.

Motivated forgetting reduces veridical memories but slightly increases false memories in both young and healthy older people.

The aim of the current study is to examine the effects of motivated forgetting and aging on true and false memory. Sixty young and 54 healthy older adults were instructed to study two lists of 18 words each. Each list was composed of three sets of six words associated with three non-presented critical words. After studying list 1, half of the participants received the instruction to forget List 1, whereas the other half received the instruction to remember List 1. Next, all the subjects studied list 2; finally, they were asked to remember the words studied in both lists. The results showed that when participants intended to forget the studied List 1, they were less likely to recall the studied words, but more likely to intrude the critical words. That is, we can intentionally forget something but this can also entail the intrusion of some related false memories.

Effects of Stimuli Repetition and Age in False Recognition.

The aim of the current study is to examine the effects of stimuli repetition and age in false recognition using the Deese-Roediger-McDermott experimental paradigm. Two matched samples of 32 young adults and 32 healthy older adults studied 10 lists of six words associated with three non-presented critical words. On half of the lists, the words were presented once, and on the other five lists, the words were presented three times, always following a same sequential order. After each study list, participants performed a self-paced recognition test containing 12 words: the 6 studied words and 6 other non-studied words (the 3 critical words and 3 distractors). The results show that false recognition increases with age and declines in both samples with repetitions (although more in the young adults than in the older people). Results are discussed in relation to the dual-process theories of (false) memory.

The role of backward associative strength in false recognition of DRM lists with multiple critical words.

Memory is a reconstruction of the past and is prone to errors. One of the most widely-used paradigms to examine false memory is the Deese/Roediger-McDermott (DRM) paradigm. In this paradigm, participants studied words associatively related to a non-presented critical word. In a subsequent memory test critical words are often falsely recalled and/or recognized. In the present study, we examined the influence of backward associative strength (BAS) on false recognition using DRM lists with multiple critical words. In forty-eight English DRM lists, we manipulated BAS while controlling forward associative strength (FAS). Lists included four words (e.g., prison, convict, suspect, fugitive) simultaneously associated with two critical words (e.g., CRIMINAL, JAIL). The results indicated that true recognition was similar in high-BAS and low-BAS lists, while false recognition was greater in high-BAS lists than in low-BAS lists. Furthermore, there was a positive correlation between false recognition and the probability of a resonant connection between the studied words and their associates. These findings suggest that BAS and resonant connections influence false recognition, and extend prior research using DRM lists associated with a single critical word to studies of DRM lists associated with multiple critical words.

The processing of semantic relatedness in the brain: Evidence from associative and categorical false recognition effects following transcranial direct current stimulation of the left anterior temporal lobe

A dominant view of the role of the anterior temporal lobe (ATL) in semantic memory is that it serves as an integration hub, specialized in the processing of semantic relatedness by way of mechanisms that bind together information from different brain areas to form coherent amodal representations of concepts. Two recent experiments, using brain stimulation techniques along with the Deese–Roediger–McDermott (DRM) paradigm, have found a consistent false memory reduction effect following stimulation of the ATL, pointing to the importance of the ATL in semantic/conceptual processing. To more precisely identify the specific process being involved, we conducted a DRM experiment in which transcranial direct current stimulation (anode/cathode/sham) was applied over the participants' left ATL during the study of lists of words that were associatively related to their non-presented critical words (e.g., rotten, worm, red, tree, liqueur, unripe, cake, food, eden, peel, for the critical item apple) or categorically related (e.g., pear, banana, peach, orange, cantaloupe, watermelon, strawberry, cherry, kiwi, plum, for the same critical item apple). The results showed that correct recognition was not affected by stimulation. However, an interaction between stimulation condition and type of relation for false memories was found, explained by a significant false recognition reduction effect in the anodal condition for associative lists that was not observed for categorical lists. Results are congruent with previous findings and, more importantly, they help to clarify the nature and locus of false memory reduction effects, suggesting a differential role of the left ATL, and providing critical evidence for understanding the creation of semantic relatedness-based memory illusions.

The costs and benefits of testing and guessing on recognition memory.

We examined whether 2 types of interpolated tasks (i.e., retrieval-practice via free recall or guessing a missing critical item) improved final recognition for related and unrelated word lists relative to restudying or completing a filler task. Both retrieval-practice and guessing tasks improved correct recognition relative to restudy and filler tasks, particularly when study lists were semantically related. However, both retrieval practice and guessing also generally inflated false recognition for the nonpresented critical words. These patterns were found when final recognition was completed during a short delay within the same experimental session (Experiment 1) and after a 24-hr delay (Experiment 2). In Experiment 3, task instructions were presented randomly after each list to determine whether retrieval-practice and guessing effects were influenced by task-expectancy processes. In contrast to Experiments 1 and 2, final recognition after retrieval practice and guessing was equivalent to restudy, suggesting that the observed retrieval-practice and guessing advantages were in part because of preparatory task-based processing during study. (PsycINFO Database Record

Normative study of theme identifiability: Instructions with and without explanation of the false memory effect.

False-memory illusions have been widely studied using the Deese/Roediger-McDermott paradigm (DRM). In this paradigm, words semantically related to a single nonpresented critical word are studied. In a later memory test, critical words are often falsely recalled and recognized. The present normative study was conducted to measure the theme identifiability of 60 associative word lists in Spanish that include six words (e.g., stove, coat, blanket, scarf, chill, and bonnet) that are simultaneously associated with three critical words (e.g., HEAT, COLD, and WINTER; Beato & Díez, Psicothema, 26, 457-463, 2011). Different levels of backward associative strength were used in the construction of the DRM lists. In addition, we used two types of instructions to obtain theme identifiability. In the without-explanation condition, traditional instructions were used, requesting participants to write the theme list. In the with-explanation condition, the false-memory effect and how the lists were built were explained, and an example of a DRM list and critical words was shown. Participants then had to discover the critical words. The results showed that all lists produced theme identifiability. Moreover, some lists had a higher theme identifiability rate (e.g., 61% for the critical words LOVE, BOYFRIEND, COUPLE) than others (e.g., 24% for CITY, PLACE, VILLAGE). After comparing the theme identifiabilities in the different conditions, the results indicated higher theme identifiability when the false-memory effect was explained than without such an explanation. Overall, these new normative data provide a useful tool for those experiments that, for example, aim to analyze the wide differences observed in false memory with DRM lists and the role of theme identifiability.

False memories in Lewy-body disease.

Recently, de Boysson, Belleville, Phillips et al. (2011) found that patients with Lewy-body disease (LBD) showed significantly lower rates of false memories than healthy controls, using the Deese-Roediger-McDermott (DRM) experimental procedure. Given that this result could be explained by the practically null rate of true recognition in the LBD group (0.09), we decided to replicate the study by de Boysson et al. (2011), but including a new condition that would maximize the true recognition rate (and analyze its effect on the rate of false memories). Specifically, in a DRM experiment, we manipulated (within subjects) two study and recognition conditions: in the "immediate" condition, both the LBD patients and the control group of healthy older people received a different recognition test after each study list (containing twelve words associated with a non-presented critical word), while in the "delayed" condition (similar to the one in de Boysson et al., 2011), the participants received the entire series of study lists and then took only one recognition test. The results showed that, in both samples, the "immediate" condition produced higher corrected rates of both true and false recognition than the "delayed" condition, although they were both lower in the LBD patients, which shows that these patients are capable of encoding and recognizing the general similitude underlying information (gist memory) in the right conditions.

Correction of false memory for associated word lists by collaborating groups.

Collaborative inhibition is often observed for both correct and false memories. However, research examining the mechanisms by which collaborative inhibition occurs, such as retrieval disruption, reality monitoring, or group filtering, is lacking. In addition, the creation of the nominal groups (i.e., groups artificially developed by combining individuals' recall) necessary for examining collaborative inhibition do not use statistical best practices. Using the Deese-Roediger-McDermott paradigm, we examined percentages of correct and false memories in individuals, collaborative interactive groups, and correctly created nominal groups, as well as the processes that the collaborative interactive groups used to determine which memories to report. Results showed evidence of the collaborative inhibition effect. In addition, analyses of the collaborative interactive groups' discussions found that these groups wrote down almost all presented words but less than half of nonpresented critical words, after discussing them, with nonpresented critical words being stated to the group with lower confidence and rejected by other group members more often. Overall, our findings indicated support for the group filtering hypothesis.

False recognition production indexes in forward associative strength (FAS) lists with three critical words.

False memory illusions have been widely studied using the Deese/Roediger-McDermott paradigm (DRM). In this paradigm, participants study words semantically related to a single nonpresented critical word. In a memory test critical words are often falsely recalled and recognized. The present study was conducted to measure the levels of false recognition for seventy-five Spanish DRM word lists that have multiple critical words per list. Lists included three critical words (e.g., HELL, LUCEFER, and SATAN) simultaneously associated with six studied words (e.g., devil, demon, fire, red, bad, and evil). Different levels of forward associative strength (FAS) between the critical words and their studied associates were used in the construction of the lists. Specifically, we selected lists with the highest FAS values possible and FAS was continuously decreased in order to obtain the 75 lists. Six words per list, simultaneously associated with three critical words, were sufficient to produce false recognition. Furthermore, there was wide variability in rates of false recognition (e.g., 53% for DUNGEON, PRISON, and GRATES; 1% for BRACKETS, GARMENT, and CLOTHING). Finally, there was no correlation between false recognition and associative strength. False recognition variability could not be attributed to differences in the forward associative strength.

Recall of false memories in individuals scoring high in schizotypy: Memory distortions are scale specific

Background and objectivesPrevious research has indicated abnormal semantic activation in individuals scoring higher in schizotypy. In the current experiment, semantic activation was examined by using the Deese-Roediger–McDermott paradigm of false memories. MethodsParticipants were assessed for schizotypy using the Oxford-Liverpool Inventory of Feelings (OLIFE). Participants studied lists of semantically related words in which a critical and highly associated word was absent. Participants then recalled the list. ResultsParticipants high in Unusual Experiences and Cognitive Disorganization recalled more critical non-presented words, weakly related studied words, and fewer studied words than participants who scored low on these measures. LimitationsPrevious research using the cognitive-perceptual factor of the Schizotypy Personality Questionnaire found reduced false memories, while the Unusual Experiences subscale of the OLIFE was associated with more false memories. Both scales cover similar unusual perceptual experiences and it is unclear why they led to divergent results. ConclusionsThe findings suggest that subtypes of schizotypy are associated with abnormal semantic activation.

False recall in the Deese-Roediger-McDermott paradigm: The roles of gist and associative strength.

Theories of false memories, particularly in the Deese-Roediger-McDermott (DRM) paradigm, focus on word association strength and gist. Backward associative strength (BAS) is a strong predictor of false recall in this paradigm. However, other than being defined as a measure of association between studied list words and falsely recalled nonpresented critical words, there is little understanding of this variable. In Experiment 1, we used a knowledge-type taxonomy to classify the semantic relations in DRM stimuli. These knowledge types predicted false-recall probability, as well as BAS itself, with the most important being situation features, synonyms, and taxonomic relations. In three subsequent experiments, we demonstrated that lists composed solely of situation features can elicit a gist and produce false memories, particularly when monitoring processes are made more difficult. Our results identify the semantic factors that underlie BAS and suggest how considering semantic relations leads to a better understanding of gist formation.

True and false memories in the DRM paradigm on a forced choice test

Participants studied lists of semantic associates that converged on a non-presented critical word (e.g., sleep; Deese, 1959; Roediger & McDermott, 1995) and took a two-alternative forced choice test. At test, each critical non-presented word was paired with a studied word from the same list. The test was administered either immediately or 7 days after the study phase. Accuracy in distinguishing between the non-presented critical word and the studied list word was above chance at immediate testing. After a 7-day retention interval, however, accuracy did not differ from chance performance: participants were as likely to choose the non-presented critical word as the studied list word.

Comparison of activation level between true and false items in the DRM paradigm

The aim of the present study was to compare the activation levels of true and false memories in the Deese-Roediger-McDermott (DRM) paradigm. For this purpose, we used a lexical decision task (LDT) that can be considered a relative pure measure of activation. Participants had to study a list of words that were semantically associated to a critical non-presented word (CI), and afterwards had to classify the actually studied words, the CI and new words in the LDT. Results indicated that the classification latency of the CI was the same as actually studied words and shorter than new words. The results might be interpreted as evidence that the false and true memory items have the same activation level and that the false memory effect can be based on the indirect activation of the CI at the encoding.

False memories are not surprising: The subjective experience of an associative memory illusion

Four experiments examined subjective experience during retrieval in the DRM false memory paradigm [Deese, J. (1959). On the prediction of occurrence of particular verbal intrusions in immediate recall. Journal of Experimental Psychology, 58, 17–22; Roediger, H. L., & McDermott, K. B. (1995). Creating false memories: Remembering words not presented in lists. Journal of Experimental Psychology: Learning, Memory, and Cognition, 21, 803–814]. Subjects studied lists of related words that were associated with critical non-presented words and then took a recognition test in which they made judgments about their experience of each test item. We tested the prediction from [Whittlesea, B. W. A. (2002). False memory and the discrepancy–attribution hypothesis: The prototype-familiarity illusion. Journal of Experimental Psychology: General, 131, 96–115] discrepancy–attribution hypothesis that subjects experience critical lures as surprising, and that the experience of surprise leads them to call the lures old. We found that subjects were not surprised when they encountered critical lures on a recognition test and, in fact, they reported that they expected to see critical lures more than they expected to see words that they had actually studied. When subjects did experience words as surprising, they called the words new, not old. The results support the idea that false memories in the DRM paradigm occur when critical lures are activated in memory and fluently processed on a test, leading subjects to experience critical lures in much the same way that they experience words they actually studied. The results do not support the idea that false memories are surprising, as stated by the discrepancy–attribution hypothesis.

Transfer of learning in avoiding false memory: The roles of warning, immediate feedback, and incentive

Participants learned semantic associates and were tested in the Deese-Roediger-McDermott (DRM) paradigm. A multiple-trial learning procedure with warning, feedback, and monetary incentive was employed to reduce false memory. Results showed that there was a progressive reduction of false memory over the trials as well as a generalization of learning to new lists of words in avoiding the critical nonpresented words attributable to explicit warning, feedback, and incentive, respectively. Both the feedback and the monetary incentive had an effect beyond what an explicit warning could obtain. In addition, participants were found to achieve false-memory reduction by enhancing the activation of the critical words and the monitoring process, rather than by strengthening the verbatim processing of the list words. It was concluded that false memory may not be as impervious to correction as was believed insofar as an effective training method is applied.

Does collaborative remembering reduce false memories?

Collaborative remembering refers to recall by groups rather than by an individual. Three experiments investigated whether, relative to individual remembering, collaborative remembering decreased correct recall and false recall using the Deese-Roediger-McDermott paradigm. Participants were first asked to study and recall five lists of 15 words that were each semantically associated with a critical non-presented word. Half the participants recalled the words by themselves, while the remaining half were assigned to pairs and collaboratively recalled the words. In Experiment 1, pairs produced the same number of false or correct words as individuals who were tested alone. In Experiment 2, the interpersonal closeness of the groups was also manipulated: friends and pairs who were not friends were assigned to the collaborative groups. Both friends and non-friends produced fewer false or correct words than individuals. Experiment 3, in which the performance of the individuals and non-friend pairs were compared using a recall test of the same 75 words as the previous experiments, replicated the results of Experiment 2. These results are discussed in terms of the retrieval-strategy disruption.

虚再認における指示忘却の効果―活性化‐モニタリング仮説の検討―

This study examined the influence of directed forgetting on false recognition. Participants studied one of two types of lists consisting of words related to a critical non-presented (CN) word: high or low semantic-associative-strength lists. Thirty-one participants were instructed to forget the Ist list before studying the 2nd list (forget group), another group of 30 participants were instructed to remember both lists (remember group). This was followed by a recognition test and Remember/Know judgments. For CN words, the forget group showed more false recognition than the remember group only for high semantic-associative-strength lists. Moreover, higher proportions of Remember judgments were observed than Know judgments in false recognition responses. These findings are discussed in terms of the activation-monitoring hypothesis of false memories.

False Memory in a Short-Term Memory Task

The Deese/Roediger-McDermott (DRM; Roediger & McDermott, 1995) paradigm reliably elicits false memories for critical nonpresented words in recognition tasks. The present studies used a Sternberg (1966) task with DRM lists to determine whether false memories occur in short-term memory tasks and to assess the contribution of latency data in the measurement of false memories. Subjects studied three, five, or seven items from DRM lists and responded to a single probe (studied or nonstudied). In both experiments, critical lures were falsely recognized more often than nonpresented weak associates. Latency data indicated that correct rejections of critical lures were slower than correct rejections of weakly related items at all set sizes. False alarms to critical lures were slower than hits to list items. Latency data can distinguish veridical and false memories in a short-term memory task. Results are discussed in terms of activation-monitoring models of false memory.