50 years of research sparked by Atkinson and Shiffrin (1968).

In this article we review the framework proposed in 1968 by Atkinson and Shiffrin. We discuss the prior context that led to its production, including the advent of cognitive and mathematical modeling, its principal concepts, the subsequent refinements and elaborations that followed, and the way that the framework influenced other researchers to test the ideas and, in some cases, propose alternatives. The article illustrates the large amount of research and the large number of memory models that were directly influenced by this chapter over the past 50 years.

On testing the strength independence assumption in retrieval-induced forgetting.

Strength independence refers to the assumption that in a retrieval-induced forgetting paradigm, the increase in performance for the practiced items (RP+) is independent of the decrease for the related and supposedly inhibited items (RP-). One way in which this assumption has been tested is by examining the correlation over subjects between these two measures. The finding that there is no such correlation has been taken as evidence for the inhibition account and against noninhibitory accounts of retrieval induced forgetting. We report several, large-scale simulation studies using a simplified version of the SAM model (Raaijmakers & Shiffrin, Psychological Review, 88, 93-134, 1981). The results clearly show that such a noninhibitory model is not likely to predict a significant correlation, despite the fact that on the level of the predicted probabilities such a correlation is clearly present. Additional simulations show that this is a very general result and not specifically related to the SAM model that was used. We conclude that such correlations do not provide a good test for the strength independence assumption and will not be able to distinguish between inhibitory and noninhibitory explanations of retrieval-induced forgetting.

Retrieval-induced forgetting without competition: testing the retrieval specificity assumption of the inhibition theory.

According to the inhibition theory of forgetting (Anderson, Journal of Memory and Language 49:415-445, 2003; Anderson, Bjork, & Bjork, Psychonomic Bulletin & Review 7:522-530, 2000), retrieval practice on a subset of target items leads to forgetting for the other, nontarget items, due to the fact that these other items interfere during the retrieval process and have to be inhibited in order to resolve the interference. In this account, retrieval-induced forgetting occurs only when competition takes place between target and nontarget items during target item practice, since only in such a case is inhibition of the nontarget items necessary. Strengthening of the target item without active retrieval should not lead to such an impairment. In two experiments, we investigated this assumption by using noncompetitive retrieval during the practice phase. We strengthened the cue-target item association during practice by recall of the category name instead of the target item, and thus eliminated competition between the different item types (as in Anderson et al., Psychonomic Bulletin & Review 7:522-530 2000). In contrast to the expectations of the inhibition theory, retrieval-induced forgetting occurred even without competition, and thus the present study does not support the retrieval specificity assumption.

Priming effects from young-old to very old age on a word-stem completion task: minimizing explicit contamination.

We investigated the size of repetition priming effects from young-old to very old age using a newly developed Word-Stem Completion (WSC) task. Retrospectively, we examined the role of explicit, intentional retrieval strategies in priming. We constructed our task by taking factors into account that were known to complicate the measurement of significant and valid priming effects. Within our sample of 170 cognitively healthy elderly persons of 55-94 years old, we found no effects on priming of age, gender, education, intelligence, cognitive status, memory complaints, or depressive symptoms. Participants that subsequent to task administration reported awareness of the study-test relationship obtained higher priming scores. However, analysis of stem-completion times showed that explicit contamination during the task was unlikely. The results suggest that WSC priming is age-invariant up to very old age. This task with increased validity might contribute to the differentiation with Alzheimer's disease by improving specificity of assessment.

How to quantify support for and against the null hypothesis: a flexible WinBUGS implementation of a default Bayesian t test.

We propose a sampling-based Bayesian t test that allows researchers to quantify the statistical evidence in favor of the null hypothesis. This Savage-Dickey (SD) t test is inspired by the Jeffreys-Zellner-Siow (JZS) t test recently proposed by Rouder, Speckman, Sun, Morey, and Iverson (2009). The SD test retains the key concepts of the JZS test but is applicable to a wider range of statistical problems. The SD test allows researchers to test order restrictions and applies to two-sample situations in which the different groups do not share the same variance.

The role of item strength in retrieval-induced forgetting.

In 3 experiments, the role of item strength in the retrieval-induced forgetting paradigm was tested. According to the inhibition theory of forgetting proposed by M. C. Anderson, R. A. Bjork, and E. L. Bjork (1994), retrieval-induced forgetting should be larger for items that are more strongly associated with the category cue. In the present experiments, the authors varied item strength on the study list by manipulating the position of an item within its category (Experiments 1 and 2) and by the number of presentations in the study phase (Experiment 3). Contrary to the predictions from inhibition theory, in all 3 experiments, stronger items did not show more retrieval-induced forgetting than weaker items.

Long-term priming of neighbours biases the word recognition process: evidence from a lexical decision task.

The role of orthographically similar words (i.e., neighbours) in the word recognition process has been studied extensively using short-term priming paradigms (e.g., Colombo, 1986). Here we demonstrate that long-term effects of neighbour priming can also be obtained. Experiment 1 showed that prior study of a neighbour (e.g., TANGO) increased later lexical decision performance for similar words (e.g., MANGO), but decreased performance for similar pseudowords (e.g., LANGO). Experiment 2 replicated this bias effect and showed that the increase in lexical decision performance due to neighbour priming is selectively due to words from a relatively sparse neighbourhood. Explanations of the bias effect in terms of lexical activation and episodic memory retrieval are discussed.

Early assessment of dementia: the contribution of different memory components.

A broad memory test battery (reflecting explicit and implicit memory functioning) was administered to a heterogeneous sample of initially nondemented, community-dwelling elderly subjects. To examine the profile of preclinical dementia, subjects were tested twice: At baseline, all subjects were nondemented according to Diagnostic and Statistical Manual of Mental Disorders (4th ed.; American Psychiatric Association, 1994) criteria; 2 years later, a subgroup had developed dementia. Performance of the preclinically demented subjects was best characterized, relative to that of cognitively impaired subjects who did not develop dementia 2 years later, by an inability to benefit at recall from semantic relations and by absent repetition priming effects. The authors conclude that in addition to testing episodic memory functioning, it is important to be aware of semantic and implicit memory deficits in the early assessment of dementia.

Aging affects both perceptual and lexical/semantic components of word stem priming: An event-related fMRI study.

In this event-related fMRI study, brain activity patterns were compared in extensive groups of young (N=25) and older (N=38) adults, while they were performing a word stem completion priming task. Based on behavioral findings, we tested the hypothesis that aging affects only the lexical/semantic, but not the perceptual component of word stem priming. To this end, we distinguished between priming-related activity reductions in posterior regions involved in visual processing, and regions associated with lexical/semantic retrieval processes, i.e., left lateral temporal and left prefrontal regions. Both groups revealed significant priming-related response time reductions. However, in accordance with earlier findings, a larger priming effect was found in the group of young participants. In line with previous imaging studies, the groups showed common priming-related activity reductions in the anterior cingulate, and the left inferior prefrontal cortex extending into the anterior portion of the left superior temporal gyrus, and at lower thresholds also in the right occipital lobe. However, when directly comparing the groups, greater priming-related reductions were found for the young group in the left anterior superior temporal gyrus and the right posterior occipital lobe. These findings suggest that, converse to current psychological views, aging affects both perceptual and lexical/semantic components of repetition priming.

A model for evidence accumulation in the lexical decision task.

We present a new model for lexical decision, REM-LD, that is based on REM theory (e.g., ). REM-LD uses a principled (i.e., Bayes' rule) decision process that simultaneously considers the diagnosticity of the evidence for the 'WORD' response and the 'NONWORD' response. The model calculates the odds ratio that the presented stimulus is a word or a nonword by averaging likelihood ratios for lexical entries from a small neighborhood of similar words. We report two experiments that used a signal-to-respond paradigm to obtain information about the time course of lexical processing. Experiment 1 verified the prediction of the model that the frequency of the word stimuli affects performance for nonword stimuli. Experiment 2 was done to study the effects of nonword lexicality, word frequency, and repetition priming and to demonstrate how REM-LD can account for the observed results. We discuss how REM-LD could be extended to account for effects of phonology such as the pseudohomophone effect, and how REM-LD can predict response times in the traditional 'respond-when-ready' paradigm.

Semantic context effects and priming in word association.

Two experiments investigated priming in word association, an implicit memory task. In the study phase of Experiment 1, semantically ambiguous target words were presented in sentences that biased their interpretation. The appropriate interpretation of the target was either congruent or incongruent with the cue presented in a subsequent word association task. Priming (i.e., a higher proportion of target responses relative to a nonstudied baseline) was obtained for the congruent condition, but not for the incongruent condition. In Experiment 2, study sentences emphasized particular meaning aspects of nonambiguous targets. The word association task showed a higher proportion of target responses for targets studied in the more congruent sentence context than for targets studied in the less congruent sentence context. These results indicate that priming in word association depends largely on the storage of information relating the cue and target.

A further look at the "language-as-fixed-effect fallacy".

The proper analysis of experiments using language materials has been a source of controversy and debate among researchers. We summarize the main issues and discuss the solutions that have been presented. Even though the major issues have been dealt with extensively in the literature, there still exists quite a bit of confusion about how to analyze the data from such experiments. We discuss a number of the most frequently voiced objections. In particular, we discuss the issue of what happens if in a counterbalanced design only some of the items show the treatment effect. Finally, a possible solution is discussed for the case where only partial matching of items between conditions is possible.

Similar network activated by young and old adults during the acquisition of a motor sequence.

In this functional MRI (fMRI) study, we investigated ageing effects on motor skill learning. We applied an adapted version of the serial reaction time (SRT) task to extensive groups of young (N=26) and elderly (N=40) subjects. Since indications have been provided for age-related shrinkage of brain regions assumed to be critical to motor skill learning, we tested the hypothesis that age effects on implicit sequence learning are larger on a neurofunctional level than on a behavioural level. The SRT task consisted of two identical scan sessions, in which subjects had to manually trail an asterisk appearing serially in one of four spatial positions by means of button-pressing. Reliable response time reductions were already found in the first session for both the young and the elderly groups, when comparing a fixed sequence condition to a random sequence, but the learning effect was greater for the young subjects. In the second session, though, both groups showed a similar degree of learning. This indicates that implicit sequence learning is still intact in elderly adults, but that the rate of learning is somewhat slower. Reliable learning-related changes in brain activity were also observed. A similar network of brain regions was recruited by both groups during the fixed compared to the random sequence, involving several regions that have been previously associated with implicit sequence learning, including bilateral parietal, and frontal regions, the supplementary motor area (SMA), cerebellum and the basal ganglia. The direct group comparison did not reveal any differences in brain activity. In addition, we did not observe any significant differences in activity when comparing the different sessions either, neither for the young nor for the elderly subjects. Hence, we did not find indications for an age-related functional reorganisation of neural networks involved in motor sequence learning. In view of earlier reports of pronounced ageing effects on the performance on declarative memory tasks, our finding of age-related sparing of processes that sustain motor skill learning, provides further support for the proposition of different memory systems relying on different brain substrates.

Deep processing activates the medial temporal lobe in young but not in old adults.

Age-related impairments in episodic memory have been related to a deficiency in semantic processing, based on the finding that elderly adults typically benefit less than young adults from deep, semantic as opposed to shallow, nonsemantic processing of study items. In the present study, we tested the hypothesis that elderly adults are not able to perform certain cognitive operations under deep processing conditions. We further hypothesised that this inability does not involve regions commonly associated with lexical/semantic retrieval processes, but rather involves a dysfunction of the medial temporal lobe (MTL) memory system. To this end, we used functional MRI on rather extensive groups of young and elderly adults to compare brain activity patterns obtained during a deep (living/nonliving) and a shallow (uppercase/lowercase) classification task. Common activity in relation to semantic classification was observed in regions that have been previously related to semantic retrieval, including mainly left-lateralised activity in the inferior prefrontal, middle temporal, and middle frontal/anterior cingulate gyrus. Although the young adults showed more activity in some of these areas, the finding of mainly overlapping activation patterns during semantic classification supports the idea that lexical/semantic retrieval processes are still intact in elderly adults. This received further support by the finding that both groups showed similar behavioural performances as well on the deep and shallow classification tasks. Importantly, though, the young revealed significantly more activity than the elderly adults in the left anterior hippocampus during deep relative to shallow classification. This finding is in line with the idea that age-related impairments in episodic encoding are, at least partly, due to an under-recruitment of the medial temporal lobe memory system.

Can false memories be created through nonconscious processes?

Presentation times of study words presented in the Deese/Roediger and McDermott (DRM) paradigm varied from 20 to 2000 ms per word in an attempt to replicate the false memory effect following extremely short presentations reported by. Both in a within-subjects design (Experiment 1) and in a between-subjects design (Experiment 2) subjects showed memory for studied words as well as a false memory effect for related critical lures in the 2000-ms condition. However, in the conditions with shorter presentation times (20 ms in Experiment 1; 20 and 40 ms in Experiment 2) no memory for studied words, nor a false memory effect was found. We argue that there is at present no strong evidence supporting the claim for a nonconscious basis of the false memory effect.

Alzheimer's disease versus normal ageing: a review of the efficiency of clinical and experimental memory measures.

This paper reviews research findings concerning memory performance in Alzheimer's disease (AD) and normal ageing. Studies using clinical (i.e., episodic) memory tests are compared with studies using various experimental memory paradigms (semantic memory, implicit memory, working memory), in order to determine their efficiency to differentiate between AD and normal ageing. In addition, attention is focused on early and preclinical AD. It is argued that traditional clinical memory tests alone are not best able at detecting AD at an early stage. More specifically, tasks calling upon semantic knowledge may aid to an earlier and more efficient assessment of AD.

Associative priming in a masked perceptual identification task: evidence for automatic processes.

Two experiments investigated the influence of automatic and strategic processes on associative priming effects in a perceptual identification task in which prime-target pairs are briefly presented and masked. In this paradigm, priming is defined as a higher percentage of correctly identified targets for related pairs than for unrelated pairs. In Experiment 1, priming was obtained for mediated word pairs. This mediated priming effect was affected neither by the presence of direct associations nor by the presentation time of the primes, indicating that automatic priming effects play a role in perceptual identification. Experiment 2 showed that the priming effect was not affected by the proportion (.90 vs. .10) of related pairs if primes were presented briefly to prevent their identification. However, a large proportion effect was found when primes were presented for 1000 ms so that they were clearly visible. These results indicate that priming in a masked perceptual identification task is the result of automatic processes and is not affected by strategies. The present paradigm provides a valuable alternative to more commonly used tasks such as lexical decision.

Medial temporal lobe activity during semantic classification using a flexible fMRI design.

In this study, we evaluated the use of a self-paced fMRI design, to allow a flexible speed of responding with only four alternating stimulus blocks to minimize the influence of task switching on a group of young subjects. This was done in view of our intention to use such a design on groups of elderly and demented subjects in the near future. In addition, the hypothesis was tested that the medial temporal lobe is involved in semantic memory similar to episodic memory using a semantic retrieval task. In line with previous imaging studies that compared a semantic (living/nonliving) to a perceptual (alphabetically ascending/descending) classification condition, activity was seen in lateral temporal and inferior frontal regions, indicating the applicability of our design. Additional activity was seen in the right, and, at a slightly lower threshold, also in the left MTL, providing support for the involvement of the MTL in retrieval from semantic memory.

Priming in implicit memory tasks: prior study causes enhanced discriminability, not only bias.

R. Ratcliff and G. McKoon (1995, 1996, 1997; R. Ratcliff, D. Allbritton, & G. McKoon, 1997) have argued that repetition priming effects are solely due to bias. They showed that prior study of the target resulted in a benefit in a later implicit memory task. However, prior study of a stimulus similar to the target resulted in a cost. The present study, using a 2-alternative forced-choice procedure, investigated the effect of prior study in an unbiased condition: Both alternatives were studied prior to their presentation in an implicit memory task. Contrary to a pure bias interpretation of priming, consistent evidence was obtained in 3 implicit memory tasks (word fragment completion, auditory word identification, and picture identification) that performance was better when both alternatives were studied than when neither alternative was studied. These results show that prior study results in enhanced discriminability, not only bias.