Showing papers on "Interference theory published in 1995"

The neural basis of the central executive system of working memory.

Abstract: WORKING memory refers to a system for temporary storage and manipulation of information in the brain, a function critical for a wide range of cognitive operations. It has been proposed that working memory includes a central executive system (CES) to control attention and information flow to and from verbal and spatial short-term memory buffers1. Although the prefrontal cortex is activated during both verbal and spatial passive working memory tasks2–8, the brain regions involved in the CES component of working memory have not been identified. We have used functional magnetic resonance imaging (fMRI) to examine brain activation during the concurrent performance of two tasks, which is expected to engage the CES. Activation of the prefrontal cortex was observed when both tasks are performed together, but not when they are performed separately. These results support the view that the prefrontal cortex is involved in human working memory.

The Formation of False Memories

Abstract: For most of this century, experimental psychologists have been interested in how and why memory fails. As Greene2 has aptly noted, memories do not exist in a vacuum. Rather, they continually disrupt each other, through a mechanism that we call "interference." Literally thousands of studies have documented how our memories can be disrupted by things that we experienced earlier (proactive interference) or things that we experienced later (retroactive interference).

Functional mapping of sequence learning in normal humans

Abstract: The brain localization of motor sequence learning was studied in normal subjects with positron emission tomography. Subjects performed a serial reaction time (SRT) task by responding to a series of stimuli that occurred at four different spatial positions. The stimulus locations were either determined randomly or according to a 6-element sequence that cycled continuously. The SRT task was performed under two conditions. With attentional interference from a secondary counting task there was no development of awareness of the sequence. Learning-related increases of cerebral blood flow were located in contralateral motor effector areas including motor cortex, supplementary motor area, and putamen, consistent with the hypothesis that nondeclarative motor learning occurs in cerebral areas that control limb movements. Additional cortical sites included the rostral prefrontal cortex and parietal cortex. The SRT learning task was then repeated with a new sequence and no attentional interference. In this condition, 7 of 12 subjects developed awareness of the sequence. Learning-related blood flow increases were present in right dorsolateral prefrontal cortex, right premotor cortex, right ventral putamen, and biparieto-occipital cortex. The right dorsolateral prefrontal and parietal areas have been previously implicated in spatial working memory and right prefrontal cortex is also implicated in retrieval tasks of verbal episodic memory. Awareness of the sequence at the end of learning was associated with greater activity in bilateral parietal, superior temporal, and right premotor cortex. Motor learning can take place in different cerebral areas, contingent on the attentional demands of the task.

Interference and inhibition in cognition: An historical perspective

Abstract: Part 1 Historical perspective: interference and inhibition in cognition - an historical perspective, F.N. Dempster. Part 2 Developmental perspectives: interference effects in memory and reasoning - a fuzzy-trace theory analysis, V.F. Reyna interference or facilitation in infant memory?, C. Rovee-Collier and K. Boller interference processes in memory development - the case of cognitive triage, C.J. Brainerd the evolution of inhibition mechanisms and their role in human cognition and behaviour, D.F. Bjorklund and K.K. Harnishfeger the development of cognitive inhibition - theories, definitions and research evidence, K.K. Harnishfeger. Part 3 Adult perspectives: selective attention and the inhibitory control of cognition, W.T. Neill et al memory interference and misinformation effects, A.L. Titcomb and V.F. Reyna skilled suppression, M.A. Gernsbacher catastrophic interference in neural networks - causes, solutions and data, S. Lewandowsky and S.-C. Li inhibitory processes in cognition and ageing, J.M. McDowd et al.

Impairments on nonspatial self-ordered and externally ordered working memory tasks after lesions of the mid-dorsal part of the lateral frontal cortex in the monkey

Abstract: Monkeys with lesions of the mid-dorsal part of the lateral frontal cortex, which extends above the sulcus principalis as far as the midline (MDL lesions), were shown to exhibit severe and long-lasting impairments on certain nonspatial working memory tasks: the self-ordered and externally ordered tasks (experiments 1, 2, 3, 5, 6, and 8). These tasks, which were modeled on similar ones previously used with patients, measure the capacity to monitor, within working memory, self-generated choices and the occurrence of externally ordered stimuli. Monkeys with lesions of the adjacent posterior dorsolateral frontal cortex, which surrounds the arcuate sulcus (PA lesions), performed as well as the normal control animals on these tasks. Experiments 4 and 5 showed that the critical variable accounting for the impairment on the self-ordered and externally ordered working memory tasks by monkeys with MDL lesions is the size of the set of stimuli that must be monitored. Furthermore, the MDL lesions did not affect basic recognition memory (experiment 6), or primacy and recency mnemonic effects [i.e., the capacity to discriminate between the initial (or final) items and other items in a list of stimuli (experiments 4 and 7), or the capacity to select from a set of stimuli on the basis of a learned fixed sequence (experiment 9)]. Thus, lesions of the mid-dorsal part of the lateral frontal cortex give rise to an impairment in working memory that depends on the size of the set of the stimuli that have to be monitored.

Learning and memory in combat veterans with posttraumatic stress disorder.

Abstract: Objective The authors investigated a broad range of memory functions for stimuli unrelated to trauma to determine whether symptoms such as intrusive memories might reflect an underlying cognitive deficit unrelated to the psychological content of the traumatic memory in patients with posttraumatic stress disorder (PTSD). Method The authors measured the intellectual functioning of 20 male combat veterans with PTSD and 12 normal comparison subjects using the WAIS and evaluated them for performance on memory using the California Verbal Learning Test. Results Veterans with PTSD showed normal abilities in the functions of initial attention, immediate memory, cumulative learning, and active interference from previous learning. However, these veterans showed a circumscribed cognitive deficit, manifested by the presence of substantial retroactive interference and revealed by a significant decrement in retention following exposure to an intervening word list. Conclusions The data suggest that patients with PTSD may have fairly specific deficits in the monitoring and regulation of memory information.

Susceptibility to memory interference effects following frontal lobe damage: Findings from tests of paired-associate learning

Abstract: Patients with frontal lobe lesions were adminstered tests of paired-associate learning in which cue and response words are manipulated to increase interference across two study lists. In one test of paired-associate learning (AB-AC test), cue words used in one list are repeated in a second list but are associated with different response words (e.g., LION-HUNTER, LION-CIRCUS). In another test (AB-ABr test), words used in one list are repeated in a second list but are rearranged to form new pairs. Compared to control subjects, patients with frontal lobe lesions exhibited disproportionate impairment of second-list learning as a result of interference effects. In particular, patients exhibited the poorest performance during the initial trial of the second list, a trial in which interference effects from the first list would be most apparent. These findings suggest that the on-line control of irrelevant or competing memory associations is disrupted following frontal lobe lesions. This disruption may be indicative of an impaired gating or filtering mechanism that affects not only memory function but other cognitive function as well.

Memory and Cognitive Abilities in University Professors: Evidence for Successful Aging

Abstract: Professors from the University of California at Berkeley were administered a 90-min test battery of cognitive performance that included measures of reaction time, paired-associate learning, working memory, and prose recall Age effects among the professors were observed on tests of reaction time, paired-associate memory, and some aspects of working memory Age effects were not observed on measures of proactive interference and prose recall, though age-related declines are generally observed in standard groups of elderly individuals The findings suggest that age-related decrements in certain cognitive functions may be mitigated in intelligent, cognitively active individuals

Generators of electrical and magnetic mismatch responses in humans.

Abstract: Studies bearing on generators of the electric and magnetic mismatch responses in humans to change in a repetitive sound are reviewed. It was concluded that the main contribution of the mismatch negativity (MMN) and of its magnetic equivalent MMNm, elicited even in the absence of attention, originates from auditory cortex on the supratemporal plane. In addition, those responses probably have one or two sources on the right lateral temporal cortex and, at least the MMN, a source also in the right frontal cortex. The activation caused by stimulus change in the sensory-specific cortex is, presumably, a manifestation of preperceptual change detection, whereas the frontal activation might be associated with conscious perception of, attention switch to, stimulus change.

Speed of processing and verbal learning deficits in adults diagnosed with attention deficit disorder.

Abstract: While frontal lobe dysfunction has been hypothesized to account for the cognitive and behavioral symptoms characteristic of Attention Deficit Disorder (ADD), inconsistent neuropsychological findings have been reported in adults with residual ADD. Twenty-five ADD adults were compared to 30 gender-, age-, and education-matched healthy controls on measures of sustained attention, psychomotor speed and integration, executive functioning, and verbal learning. There were no group differences in accuracy on the Continuous Performance Task ; however, ADD patients exhibited slower reaction time to targets. In addition, patients exhibited slowed psychomotor speed relative to controls. No group differences were observed in executive functioning. For memory, patients acquired less information and displayed inconsistent application of a semantic clustering strategy. Cuing memory with semantic prompts improved memory performance to a greater degree for patients than for controls. Patients were susceptible to retroactive interference and item recall inconsistency. The results indicate a selective pattern of deficits, revealing slowed cognitive processing and a significant list-learning deficit. Neurobehavioral deficits are discussed in relation to hypotheses of frontal lobe dysfunction in ADD.

Cortical Metabolic Activation in Humans during a Visual Memory Task

Abstract: A delayed match-to-sample (DMS) task of abstract, visual memory was performed during the uptake period of 18F-fluorodeoxyglucose. The increase in glucose uptake of cortical and subcortical regions ("activation") during the DMS task was compared with that during a control, immediate match-to-sample task using positron emission tomography. Both discriminant analysis and paired t tests supported the observation that the dorsolateral prefrontal area underwent the greatest activation, while a factor analysis revealed the functional correlation matrices of the tasks. Activations in the ventral premotor cortex and supramarginal and angular gyri were highly correlated with the change in the dorsolateral prefrontal cortex. The basal forebrain/ventral pole region showed a smaller but independently significant change. The findings support the role of the dorsal prefrontal region in the nonspatial working memory of humans.

Interference effects in young children's long-term retention.

Abstract: Classical interference effects were examined in preschool and kindergarten children's paired-associate recall (Experiment 1) and free recall (Experiment 2). Children in the control conditions learned a single picture list, whereas children in the experimental conditions learned 2 picture lists in succession. After 24 hr, children recalled items from the one list they had learned (control conditions), items from only List I (retroactive interference conditions), or items from both lists (modified free-recall conditions). Analyses based on the trace-integrity framework indicated that (a) children were susceptible to interference, (b) the locus of interference effects was at storage, (c) both younger (preschool) and older (kindergarten) children experienced similar amounts of interference, and (d) variations in trace strength generally did not modulate the magnitude of interference effects.

Script-Based Inference And Memory Retrieval In Subsymbolic Story Processing

Abstract: DISCERN is an integrated natural language processing system built entirely from distributed neural networks. It reads short narratives about stereotypical event sequences, stores them in episodic memory, generates fully expanded paraphrases of the narratives, and answers questions about them. Processing in DISCERN is based on hierarchically-organized backpropagation modules, communicating through a central lexicon of word representations. The lexicon is a double feature map system that transforms each orthographic word symbol into its semantic representation and vice versa. The episodic memory is a hierarchy of feature maps, where memories are stored “one-shot” at different locations. Several high-level phenomena emerge automatically from the special properties of distributed neural networks in this model. DISCERN learns to infer unmentioned events and unspecified role fillers, generates expectations and defaults, and exhibits plausible lexical access errors and memory interference behavior. Word semantics, memory organization, and appropriate script inferences are automatically extracted from examples. DISCERN shows that high-level natural language processing is feasible through integrated subsymbolic systems. Subsymbolic control of high-level behavior and representing and learning abstractions are the two main challenges in scaling up the approach to more open-ended tasks.

Serial position effects and duration of memory for nonspatial stimuli in rats.

Abstract: Recognition memory for lists of nonspatial items was tested in rats using a nonmatching-to-sample task. The recency effect at short retention intervals disappeared as the interval increased, but primacy did not develop although responding was still above chance up to 2 hr after 10-s sample times. Neither proactive nor retroactive interference was apparent. Rats transferred the nonmatching-to-sample rule to completely novel stimuli. The study failed to replicate the prominent U-shaped serial position curve found in a similar study by P. Reed, T. Chih-Ta, J.P. Aggleton, and J.N.P. Rawlins (1991), for which E.A. Gaffan and D. Gaffan (1992) had found the data less variable than expected. Evidence of primacy in this procedure remains insubstantial.

Interference or facilitation in infant memory

Abstract: Publisher Summary Interference, in general, is determined by the similarity of the interfering material and the material to be retained In adults, the effects of retroactive interference on forgetting are stronger after shorter retention intervals, whereas the effects of proactive interference are stronger after longer ones The interference of new learning with old is an active process Historically, research using visual attention procedures has found little evidence that infants are susceptible to interference of either type during the first year of life This chapter reviews theories of interference and problems associated with measuring interference in infants, as well as prior infant research on interference It also presents work on proactive and retroactive interference Effects of new information depend critically on its timing and new information can facilitate performance on a retention test as well as impair it

Evoked brain potentials and memory: more positivity in response to forgotten items.

Abstract: Brain evoked potentials (EPs) were recorded in human subjects participating in a free recall memory task involving retroactive interference. A learning list was presented five times. The fifth repetition was followed by an interference list. Both lists were composed of either words or abstract figures, and each subject experienced each of four possible combinations. Analysis of items recalled during the learning phase revealed larger N400 and P600 amplitudes for those items that were later forgotten vs remembered following the interference. This contradicts the usual finding that more positivity is associated with better memory. However, both the present as well as the extant findings can be explained in terms of cognitive resource allocation. Specifically, items receiving greater allocations are more likely to be immediately recalled. However, as the number of items in working memory increases, the allocation required to add new items also increases. Thus, items learned on later trials would receive larger allocations (i.e., larger positivities) than items learned earlier, yet would be more likely forgotten following the interference because their presence in memory would not be reinforced during later trials, as is the case with items learned earlier.

Memory decrement and proactive interference

Abstract: We hypothesized that subjective expectations were responsible for the release from proactive interference (PI) rather than the actual change in taxonomic subcategories. The experimental conditions included informing or not informing subjects of the change in taxonomic subcategory, crossed with whether the taxonomic subcategory was actually changed or remained unchanged. The interest in whether a release would be seen when the subjects were informed of a change, yet no change was actually administered, was most specific. Significantly higher recall was found for the condition in which subjects were informed of a change, but no actual change occured (I-NC) than in the control condition in which subjects were neither informed of a change nor did one occure (NI-NC). The overal results indicated that control condtions revealed lower recall than other conditions, F(3, 129) = 2.96. p < .05 on the critical trial. However, even when subjects were not told of a change, but a change occured (NI-C) a release from PI was observed, M=.69. A release from PI may be induced by the subject's self-awareness of the change in taxonomic subcategory. Russ-Eft (1979) wrote that "proactive interference (PI) is a phenomenon characterized by a decreasing probability of recall with an increasing number of to-beremembered items" (p. 422). However, since the late 1970's, little has been done to further understand this concept. Nevertheless, we were interested in the phenomenon because of the importance it plays in diverse applications in society. For example, it may be that if the phenomenon of PI is understood better, we will be able to utilin. that knowledge when developing academic learning curricula, industrial training programs, and other procedural learning situations. It may be possible to make those learning experiences more fruitful by helping learners commit more information to long-term memory. In the PI paradigm, subjects are presented with a list of three (sometimes more) items that are designated the to-beremembered (TBR) items. After a distraction period subjects are prompted to recall these items. The original research varied the amount of time the distraction interval lasted (Brown, 1958; Peterson & Peterson, 1959). They found that the longer retention interval resulted in a greater memory decrement. Brown (1958) postulated that memory loss after repeated trials was a result of decay. The decay theory states that information presented to the memory system starts to deteriorate immediately and automatically from short-term memory. He found a decrement in memory performance, as a function of time (Brown's finding was used to validate the decay theory). He presumed that because memory performance decreased as the retention interval increased, it must be the increased length of time that is responsible for poor memory performance. However, Keppel and Underwood (1962) found that by maintaining a constant retention interval, there was still a decrement of memory from the first to the later trials when utilizing the same type of distractor task as Peterson and Peterson (Wickens, 1970). This finding led Keppel and Underwood to suggest that there was another process other than decay contributing to the loss of memory. They introduced a theory of memory loss being "attributable to PI [Proactive Interference] from previous items" (Baddeley, 1966, p. 302). Wickens (1970) was interested in determining what mechanism might compensate for the effect of PL He began an intensive study of the release from PI phenomenon. In Wickens' research, four trials were conducted in one category (e.g., fish). The study items were then changed to another category (e.g., tools) on the critical trial. He found that on this MODERN PSYCHOLOGICAL STUDIES 57 William J. Thornton and Bennett L. Schwartz on the critical trial. He found that on this crucial trial, when the category was changed, the subjects showed improved memory for the new category items. Thus, after having been given four consecutive trials of four words in each trial, the subjects have learned a total of 16 items from the same category (e.g., fish). If asked to recall the last group of four fish names, the subjects will have the previous 12 names of fish interfering with his or her memory. However, on the fourth trial, if the category was shifted to tools the subjects would have only four tools to remember, with no prior tools impeding upon the memory of those four. Thus, release from PI allows memory performance to improve. Gardiner, Craik, and Birtwhistle (1972) referred to the 1970 study by Wickens in which summarized several prior studies which demonstrated that "rapid accumulation of PI depended upon similarity, across trials, of the TBR [tobe-remembered] items and further that performance could be restored almost to the original level if, on the next trial the TBR items were drawn from a different class" (p. 778). However, despite the general agreement as to the value of the release from PI effect when studying encoding, there is still no clear determination of the underlying causes. Gardiner et al. (1972) summarized three basic hypotheses that could account for the release from PI. The first explanation is Wickens' (1970) attentional hypothesis which states that the subjects are alerted by the change in the TBR material on the critical trial and consequently gives more attention to the new materiaL The second possibility described by Gardiner et al., is Posner's storage hypothesis which suggests that "PI reflects spontaneous interaction during storage between the traces of current items and those of similar items stored from preceding trials. By this view the 'release' items are less vulnerable to intertrial interference" (p. 778-779). The third interpretation offered by Gardiner et al. was the retrieval hypothesis, attributed to Wickens. In this hypothesis, once "the TBR material is changed, the novel items supply a new, and thus more effective retrieval cue" (p. 779). Wicken's research demonstrates that switching to a new category allows memory performance to increase to first trial levels. However, it is not clear from his work as to what defines a category. In fact, Gardiner et al. (1972) found no release when subjects were not informed about a subtle category shift (e.g. garden flowers to wild flowers). It may not be that categories, per se, are responsible for the release phenomenon. Rather, subject's subjective expectations may play a key role in the release phenomenon. As such, we suspected that subjective expectations alone may lead to the release from PI. To test this, we lead people to believe a category-shift had occurred when, in fact, no shift actually occurred (O'Neill, Sutcliffe & Tulving, 1976). Our hypothesis was in partial agreement with each of the three hypotheses summarized by Gardiner et al. (1972). As with the attentional hypothesis, we too suspected that the subjects would be alerted by the change in category. This would allow the subjects to give more attention to the new material because the prior material is not going to be requested at recall. However, we suspected that the same effect would occur if the subjects thought that the category had changed even when the category remained constant. As to the storage hypothesis, we anticipated that the release items were less vulnerable to intertrial interference. However, it was our suggestion that this was a result of the subject's subjective expectation that information from the prior category would not be required at test, rather than the fact that it would not. As to Wickens' retrieval hypothesis, we proposed that it was the belief of category change that supplied the new, and thus more effective retrieval cue, rather than the change itself. We suspected that the mechanism responsible for the release from PI was not the actual change in categories, but rather it was the subjective expectation that allows people to reorganize memory output, and thus permits release whether or not the category had actually changed. 58 MODERN PSYCHOLOGICAL STUDIES

Assessing Higher Order Functions of the Cerebral Cortex Using Functional MRI: The Working Memory Function of Prefrontal Cortex as a Case Study

Abstract: The prefrontal cortex has been implicated in a variety of higher cognitive faculties, including planning, problem solving, and language. However, little is known about the specific cognitive mechanisms that this brain area subserves. The application of functional neuroimaging techniques has recently begun to shed light on this question. This chapter briefly reviews the role of frontal cortex in working memory, and discusses a recent study designed to examine this function of prefrontal cortex using magnetic resonance imaging. The results of this study support the role of prefrontal cortex in working memory, identify specific regions that appear to be involved and, more generally, demonstrate that functional MRI (fMRI) using conventional scanning equipment can be used to study the activation of associative cortex in individual human subjects. These results set the stage for more detailed examinations of the functional organization of prefrontal cortex, as well as other regions of associative cortex.

Buildup of proactive interference in japanese kanji learning

Abstract: "The drscriminative assumption on buildup of proactrve rnterference rn short-term memory predicts that when intertnal srmilarity of items is high. the proactive interference is built up while interlist srmilarity of items is low, the proactive interference is not built up. To test the discriminatrve assumption in Japanese Kanlilearning, intertrial similarity was changed by the acoustic, the radical (as one of the figurative properties), and the radical plus semantic properties in Kany. For the acoustic and the radical plus semantic propertres, the proactrve rnterference was built up in the non-discriminative list with high intertrial srmilarity but not built in the discriminative list with low intertrial similarity, which supported the prediction. The prediction was not supported for the radical property. The findings were drscussed with reference to the drscriminatrve function of these Since Keppel and Underwood (1962) found proactive interference in short-term memory by using the Brown-Peterson paradigm, many experiments have been performed on buildup and release of proactive interference (e.g., Kikuno, 1983; Mori, 1979; Radtke & Grove, 1977; Radtke, et al., 1982; Watkins & Watkins, 1975; wickens, 1970). In the typical research paradigm on buildup of proactive interference, subjects are given successively three or four trials ofthree items in each trial. When the subjects' performances decrease with increasing trials, buildup of proactive interference is documented. Among several factors affecting buildup of proactive interference, intertrial similarity of items is assumed to be one of the most important determinants. Fujita (1985) examined effects of the interlist similarity on buildup of proactive interference by using the items of taxonomic categories. Each list had four trials of three items in each. For the non-discriminative list all trials had three items of the same category (e.g., flower) and for the discriminative list each trial had three items of four different categories (e.g., four-footed animals, fruits, fish, and carpenter's tools). The percent correct recall decreased significantly with increasing trials for the non-discriminative list, whereas the percent did not change with trials for the discriminative list. Since the discriminative list has the items of quite different categories in each trial, the items in each trial are encoded and retrieved distinctively by using the different category names as memory cues. Thus the proactive interference is not built up. Since the non-discriminative list has the items of the same category in all trials, on the other hand, only one category name is used as the