Cognitive control of familiarity: Directed forgetting reduces proactive interference in working memory

Proactive interference (PI) occurs when previously learned information interferes with new learning. In a working memory task, PI induces longer response times and more errors to recent negative probes than to new probes, presumably because the recent probe’s familiarity invites a “yes” response. Warnings, longer intertrial intervals, and the increased contextual salience of the probes can reduce but not eliminate PI, suggesting that cognitive control over PI is limited. Here we tested whether control exerted in the form of intentional forgetting performed during working memory can reduce the magnitude of PI. In two experiments, participants performed a working memory task with directed-forgetting instructions and the occasional presentation of recent probes. Surprise long-term memory testing indicated better memory for to-be-remembered than for to-be-forgotten items, documenting the classic directed-forgetting effect. Critically, in working memory, PI was virtually eliminated for recent probes from prior to-be-forgotten lists, as compared to recent probes from prior to-be-remembered lists. Thus cognitive control, when executed via directed forgetting, can reduce the adverse and otherwise persistent interference from familiarity, an effect that we attribute to attenuated memory representations of the to-be-forgotten items.     

Short term forgetting in the absence of proactive interference

It has been claimed that the short-term forgetting shown by the Peterson technique is entirely due to proactive interference from prior experimental items. Two experiments investigated this by studying forgetting when prior items were avoided by testing subjects only once. Both experiments showed significant forgetting, although the degree of forgetting was less than with a multitrial procedure. On the basis of this and other results it is suggested that the Peterson technique comprises two components, a primary memory component which decays within 6 sec, and a more stable secondary memory component. Forgetting with the multitrial procedure is attributed principally to the need to use temporal retrieval cues to avoid confusion between successive items; longer retention intervals are associated with reduced temporal discriminability and hence poorer recall.     

Directed forgetting of faces: the role of response criterion

In two experiments, we examined recognition for faces following item method directed forgetting. During testing, participants reported whether the face was a new face or, if they thought it was a studied face, they identified the instruction paired with the face during study. In both experiments, the proportion of new faces falsely recognized and classified as forget faces exceeded those falsely recognized and classified as remember faces. Despite the use of different response criteria during testing, participants showed greater discrimination accuracy when identifying remember faces than when identifying forget faces. Taken together, these data patterns indicate that participants employed a strength-based criterion when responding. Specifically, participants responding to new faces were more likely to classify those faces as forget faces from study rather than as remember faces from study.     

Implicit proactive interference, age, and automatic versus controlled retrieval strategies

We assessed the extent to which implicit proactive interference results from automatic versus controlled retrieval among younger and older adults. During a study phase, targets (e.g., "ALLERGY") either were or were not preceded by nontarget competitors (e.g., "ANALOGY"). After a filled interval, the participants were asked to complete word fragments, some of which cued studied words (e.g., "A_L_ _GY"). Retrieval strategies were identified by the difference in response speed between a phase containing fragments that cued only new words and a phase that included a mix of fragments cuing old and new words. Previous results were replicated: Proactive interference was found in implicit memory, and the negative effects were greater for older than for younger adults. Novel findings demonstrate two retrieval processes that contribute to interference: an automatic one that is age invariant and a controlled process that can reduce the magnitude of the automatic interference effects. The controlled process, however, is used effectively only by younger adults. This pattern of findings potentially explains age differences in susceptibility to proactive interference.     

Directed forgetting and posthypnotic amnesia: information processing and social contexts

In three experiments we compared posthypnotic amnesia (PHA) with directed forgetting (DF), evaluating subjects' hypnotic susceptibility, hypnotized or not. Experiment 1 suggested that the memory processes in PHA and DF were not the same. Low and high susceptibility subjects responded differently in the two contexts. Experiment 2 demonstrated a context effect for highly susceptible subjects: They showed the usual DF response outside hypnosis but not during hypnosis. Experiment 3 showed that high and low susceptibility subjects responded similarly to DF instructions outside hypnosis, eliminating the susceptibility as an important variable in the DF response. We concluded: (a) The interaction of context and susceptibility was the important determinant of unusual DF responses for high susceptibles, and (b) information processing concepts are too limited to explain PHA and perhaps DF. Theories that include interactions appear necessary for an accurate understanding of hypnotic phenomena and perhaps some phenomena usually focused on by cognitive psychology.     

Working-memory capacity, proactive interference, and divided attention: limits on long-term memory retrieval

Two experiments examined how individual differences in working-memory capacity (WM) relate to proactive interference (PI) susceptibility. We tested high and low WM-span participants in a PI-buildup task under single-task or dual-task ("load") conditions. In Experiment 1, a finger-tapping task was imposed during encoding and retrieval of each list; in Experiment 2, tapping was required during encoding or retrieval. In both experiments, low spans showed greater PI than did high spans under no load, but groups showed equivalent PI under divided attention. Load increased PI only for high spans, suggesting they use attention at encoding and retrieval to combat PI. In Experiment 2, only low spans showed a dual-task cost on List 1 memory, before PI built up. Results indicate a role for attentional processing, perhaps inhibitory in nature, at encoding and retrieval, and are discussed with respect to theories of WM and prefrontal cortex function.     

Personality correlates of retrieval processes in intentional and unintentional forgetting

Early accounts of intentional forgetting relied exclusively on mechanisms that operate at encoding. The results of more recent experiments indicate that memory-retrieval inhibition also is involved, thus linking waking intentional forgetting with hypnotic amnesia. The present Experiment 1 was designed to provide further support for the retrieval-inhibition hypothesis by testing four alternative explanations for some of the recent results. Experiment 2 examined the role of word affect in both intentional forgetting and unintentional forgetting to evaluate theoretical interpretations of the retrieval mechanisms thought to be involved. Unintentional forgetting of negative words was consistent with a repression-like interpretation, whereas intentional forgetting of negative words appeared to be limited by sensitization. Personality correlates of the recall results were identified and were found to be in support of these interpretations.     

Examining a processing tradeoff explanation of proactive interference

Burns (1989) claims that proactive interference effects occur in paired-associate learning because of tradeoffs in relational and response-specific processing. Consistent with this claim, Burns demonstrated that free recall of critical-list responses is better in the interference condition than in the control condition. Burns’s processing trade off explanation predicts that the occurrence of this reverse-interference effect should be positively correlated with the occurrence of traditional interference effects. We present several experiments whose results are inconsistent with-this-prediction. We hypothesize that the reverse-interference effect is a list-length effect. The results of a final experiment, contrasting the predictions of the list-length and processing trade off explanations, support the list-length explanation.     

The role of retrieval practice in directed forgetting

Three experiments tested the possibility that retrieval-induced forgetting is responsible for directed forgetting with the list method. In Experiments 1 and 2, additional List 2 retrieval practice was given to determine whether this would increase directed forgetting. In Experiment 1 all items came from a single category, and in Experiment 2 unrelated words were used. In Experiment 3 additional List 2 study accompanied List 2 retrieval practice. There was no evidence that List 2 retrieval practice, with or without additional List 2 study, affected the magnitude of directed forgetting. It was argued that retrieval-induced forgetting could not account for these results. Accounts with greater viability include retrieval strategy disruption and a modified version of the dissociation hypothesis.     

Working memory span and the role of proactive interference.

The authors investigated the possibility that working memory span tasks are influenced by interference and that interference contributes to the correlation between span and other measures. Younger and older adults received the span task either in the standard format or one designed to reduce the impact of interference with no impact on capacity demands. Participants then read and recalled a short prose passage. Reducing the amount of interference in the span task raised span scores, replicating previous results (C. P. May, L. Hasher, & M. J. Kane, 1999). The same interference-reducing manipulations that raised span substantially altered the relation between span and prose recall. These results suggest that span is influenced by interference, that age differences in span may be due to differences in the ability to overcome interference rather than to differences in capacity, and that interference plays an important role in the relation between span and other tasks.     

The role of proactive interference in mnemonic techniques

The success of many mnemonic techniques, such as the method of loci, is based on the use of specific well-known anchors, which are mentally combined with to-be-learned items and subsequently facilitate their retrieval. In our studies we intended to answer the question of whether the repeated application of the method of loci may result in proactive interference effects, as might be expected due to the applied association of items with the same loci each time the method is used. To this end, we manipulated list similarity in a typical proactive interference design and compared the method of loci with the link method and the rehearsal method, which do not involve the use of a specified set of anchors. Our results replicate those from other studies, which have shown that the use of a mnemonic technique leads to superior recall of list items compared to a simple rehearsal strategy. We were further able to show that the repeated learning of items from different categories results in moderate practice effects over three list-learning trials, whereas this effect is superimposed by an effect of proactive interference if different lists are composed of items from the same category. However, this effect of proactive interference was not increased for the method of loci, and we discuss this finding with regard to its practical implications.     

Predicting one's own forgetting: the role of experience-based and theory-based processes

The authors examined the hypothesis that judgments of learning (JOL), if governed by processing fluency during encoding, should be insensitive to the anticipated retention interval. Indeed, neither item-by-item nor aggregate JOLs exhibited "forgetting" unless participants were asked to estimate recall rates for several different retention intervals, in which case their estimates mimicked closely actual recall rates. These results and others reported suggest that participants can access their knowledge about forgetting but only when theory-based predictions are made, and then only when the notion of forgetting is accentuated either by manipulating retention interval within individuals or by framing recall predictions in terms of forgetting rather than remembering. The authors interpret their findings in terms of the distinction between experience-based and theory-based JOLs.     

The role of horizontal categorization in retroactive and proactive interference

Four paired-associate experiments with a total N of 291 participants investigated the effects of horizontal categorization on retroactive and proactive interference. (Exclusively) horizontal categorization means that unique categorical relationships hold across the A-B and A-C stimulus-response pairs of successive word lists (e.g., fruit--pear, river--Thames, in list 1; and fruit--plum, river--Wolga, in list 2). Experiment 1 found no significant amounts of interference with this type of list organization. However, strong interference arose with the same materials when the categorical structure was destroyed in Experiment 2. A third experiment contrasted two alternative explanations for these results, and Experiment 4 replicated the effect of horizontal categorization (vs. no categorical relationship) in a within-participants design. The results of the four experiments largely fit with a response competition explanation proposed by Bower, Thompson-Schill, and Tulving (1994), adapted to the within-participants designs used here. Overall, the present findings add to a body of evidence demonstrating limits to retroactive and proactive interference.     

Another look at retroactive and proactive interference: A quantitative analysis of conversion processes

Traditionally, the causes of interference phenomena were sought in “real” or “hard” memory processes such as unlearning, response competition, or inhibition, which serve to reduce the accessibility of target items. I propose an alternative approach which does not deny the influence of such processes but highlights a second, equally important, source of interference—the conversion (Tulving, 1983) of accessible memory information into memory performance. Conversion is conceived as a problem-solving-like activity in which the rememberer tries to find solutions to a memory task. Conversion-based interference effects are traced to different conversion processes in the experimental and control conditions of interference designs. I present a simple theoretical model that quantitatively predicts the resulting amount of interference. In two paired-associate learning experiments using two different types of memory tests, these predictions were corroborated. Relations of the present approach to traditional accounts of interference phenomena and implications for eyewitness testimony are discussed.     

Another look at retroactive and proactive interference: a quantitative analysis of conversion processes

Traditionally, the causes of interference phenomena were sought in "real" or "hard" memory processes such as unlearning, response competition, or inhibition, which serve to reduce the accessibility of target items. I propose an alternative approach which does not deny the influence of such processes but highlights a second, equally important, source of interference-the conversion (Tulving, 1983) of accessible memory information into memory performance. Conversion is conceived as a problem-solving-like activity in which the rememberer tries to find solutions to a memory task. Conversion-based interference effects are traced to different conversion processes in the experimental and control conditions of interference designs. I present a simple theoretical model that quantitatively predicts the resulting amount of interference. In two paired-associate learning experiments using two different types of memory tests, these predictions were corroborated. Relations of the present approach to traditional accounts of interference phenomena and implications for eyewitness testimony are discussed.     

Local proactive interference in delayed matching to sample: the role of reinforcement

Discriminability in delayed matching to sample was lower when the samples on consecutive trials differed compared with when samples on consecutive trials were the same. This local proactive interference occurred when correct choices on the previous trial were reinforced but not when correct choices on the previous trial were not reinforced. When the choice on the previous trial was incorrect, discriminability was higher on different consecutive trials than on same trials. These effects were amplified by varying the ratio of reinforcers for correct choices, as predicted by a model that attributes local proactive interference to an interaction between control by the sample on the current trial and the influence of reinforcers for correct choices on previous trials.     

Directed forgetting in pigeons resulting from the reallocation of memory-maintaining processes on forget-cue trials

When procedural artifacts are controlled, it has been difficult to demonstrate directed forgetting in pigeons. However, previous research with pigeons has not allowed for the reallocation of working memory (from forget items to remember items) on forget-cued trials as is possible in human directed forgetting experiments. In the present experiment, directed forgetting was found while controlling for procedural artifacts and allowing the pigeons to reallocate memory resources on forget trials. The results indicate that under these conditions, pigeons have active control over memory processing.     

Directed forgetting in direct and indirect tests of memory: Seeking evidence of retrieval inhibition using electrophysiological measures

We investigated whether directed forgetting as elicited by the item-cueing method results solely from differential rehearsal of to-be-remembered vs. to-be-forgotten words or, additionally, from inhibitory processes that actively impair retrieval of to-be-forgotten words. During study, participants (N = 24) were instructed to remember half of a series of presented words (TBR) and to forget the other half (TBF), as indicated by an instruction cue shown shortly after each word. During test, accuracy and reaction time measures from lexical decisions (indirect memory test) followed by recognition-memory judgements (direct memory test) were supplemented with event-related potential (ERP) recordings. Results from the behavioural measures revealed directed forgetting in the recognition-memory test but not the lexical-decision test. ERPs obtained during recognition indicated that TBR words elicited a larger parietal old/new effect than TBF words overall, suggesting that remember/forget instructions impaired conscious recollection processes more severely than familiarity processes. Moreover, TBF words that were successfully forgotten elicited less parietal activity than correctly rejected new words (the reversed old/new effect; Nowicka, A., Jednorórog, K., Wypych, M., & Marchewka, A. (2009). Reversed old/new effect for intentionally forgotten words: An ERP study of directed forgetting. International Journal of Psychophysiology, 71, 97–102). This was taken to implicate that inhibitory processes likely affected these items. Enhanced negativities for successfully forgotten TBF words relative to new words were observed in the lexical-decision task at early (150–250 ms) and late (800–1000 ms) time windows, suggesting that inhibitory influences disrupt more than just conscious recollection when memory retrieval is tested indirectly.     

Inhibitory processes in memory are impaired in schizophrenia: Evidence from retrieval induced forgetting

Schizophrenic patients are known to exhibit inhibitory impairments in response suppression and selective attention. However, the impairment of inhibitory control in memory retrieval has not clearly been documented. In two experiments, we investigate inhibition in memory retrieval by using the retrieval practice procedure. In Expt 1, a cued recall final test was used. Consistent with previous research, we found similar retrieval‐induced forgetting (RIF) effects in schizophrenic patients and in controls. However, these effects could be the result of interference/blocking or the results of inhibition. In order to reduce the influence of blocking in Expt 2, we used a recognition test. We found that RIF was reduced in patients, compared to healthy controls. The elimination of RIF effect in patients, when the influence of blocking is reduced, indicates that inhibitory processes in memory are altered in schizophrenia. Result suggest that schizophrenic patients suffer from critical impairments in inhibitory processes involved in memory retrieval, similar to the inhibitory deficits found in other cognitive domains.     

Directed forgetting in obsessive-compulsive disorder: replication and extension

Previous research indicates that patients with obsessive-compulsive disorder (OCD) show impaired ability to forget negative material (Behav Res Ther 34 (1996) 633). The present study was intended to replicate and extend these findings by separating emotional valence from personal relevance, using idiographic stimulus selection to control for heterogeneity of OCD concerns, and the addition of an anxious control group. We used a directed forgetting paradigm, in which participants were instructed to either remember or forget a series of words. No differences in impairment were found among the groups during a free recall test, in which they were asked to recall both "remember" and "forget" words. However, during a recognition test, OCD patients showed greater impaired forgetting for OCD-relevant words than did nonanxious and anxious controls. Valence of words did not emerge as a contributing factor to impaired forgetting.