Output loss or rehearsal loop? Output-time versus pronunciation-time limits in immediate recall for forgetting-matched materials

Modeling working memory: a computational implementation of the Time-Based Resource-Sharing theory

Working memory is a core concept in cognition, predicting about 50% of the variance in IQ and reasoning tasks. A popular test of working memory is the complex span task, in which encoding of memoranda alternates with processing of distractors. A recent model of complex span performance, the Time-Based-Resource-Sharing (TBRS) model of Barrouillet and colleagues, has seemingly accounted for several crucial findings, in particular the intricate trade-off between deterioration and restoration of memory in the complex span task. According to the TBRS, memory traces decay during processing of the distractors, and they are restored by attentional refreshing during brief pauses in between processing steps. However, to date, the theory has been formulated only at a verbal level, which renders it difficult to test and to be certain of its intuited predictions. We present a computational instantiation of the TBRS and show that it can handle most of the findings on which the verbal model was based. We also show that there are potential challenges to the model that await future resolution. This instantiated model, TBRS*, is the first comprehensive computational model of performance in the complex span paradigm. The Matlab model code is available as a supplementary material of this article.

Short-term memory stages in sign vs. speech: the source of the serial span discrepancy

Speakers generally outperform signers when asked to recall a list of unrelated verbal items. This phenomenon is well established, but its source has remained unclear. In this study, we evaluate the relative contribution of the three main processing stages of short-term memory--perception, encoding, and recall--in this effect. The present study factorially manipulates whether American Sign Language (ASL) or English is used for perception, memory encoding, and recall in hearing ASL-English bilinguals. Results indicate that using ASL during both perception and encoding contributes to the serial span discrepancy. Interestingly, performing recall in ASL slightly increased span, ruling out the view that signing is in general a poor choice for short-term memory. These results suggest that despite the general equivalence of sign and speech in other memory domains, speech-based representations are better suited for the specific task of perception and memory encoding of a series of unrelated verbal items in serial order through the phonological loop. This work suggests that interpretation of performance on serial recall tasks in English may not translate straightforwardly to serial tasks in sign language.

The recall of information from working memory. Insights from behavioural and chronometric perspectives

We describe and evaluate a recall reconstruction hypothesis for working memory (WM), according to which items can be recovered from multiple memory representations. Across four experiments, participants recalled memoranda that were either integrated with or independent of the sentence content. We found consistently longer pauses accompanying the correct recall of integrated compared with independent words, supporting the argument that sentence memory could scaffold the access of target items. Integrated words were also more likely to be recalled correctly, dependent on the details of the task. Experiment 1 investigated the chronometry of spoken recall for word span and reading span, with participants completing an unfinished sentence in the latter case. Experiments 2 and 3 confirm recall time differences without using word generation requirements, while Experiment 4 used an item and order response choice paradigm with nonspoken responses. Data emphasise the value of recall timing in constraining theories of WM functioning.

The word-length effect provides no evidence for decay in short-term memory

Many researchers regard the word-length effect (WLE) as one of the strongest pieces of evidence for time-based decay in short-term memory. We argue that the WLE is, in fact, undiagnostic for the decay hypothesis for two reasons. First, the WLE represents a correlation across words between articulation duration and memory performance, and articulation duration is inevitably confounded with other word characteristics. Recent research has confirmed that such confounds are responsible for much, maybe all, of the WLE. Second, there is strong evidence for an attentional mechanism of refreshing memory traces that can operate concurrently with articulation. Any viable decay-based model must include such a mechanism, but such a model no longer necessarily predicts a WLE, because longer spoken duration does not imply longer postponement of refreshing. We conclude that the WLE is not diagnostic for decay in short-term memory.

Rethinking speed theories of cognitive development. Increasing the rate of recall without affecting accuracy

Researchers have suggested that developmental improvements in immediate recall stem from increases in the speed of mental processes. However, that inference has depended on evidence from correlation, regression, and structural equation modeling. We provide counter-examples in two experiments in which the speed of spoken recall was manipulated. In one experiment, second-grade children and adults recalled lists of digits more quickly than usual when the lists were presented at a rapid rate of two items per second. In a second experiment, children received lists at a rate of one item per second; half the children were trained (successfully) to speak their responses more quickly than usual, at a rate similar to adults' usual rate. Recall accuracy was completely unaffected by either of these response-speed manipulations. Thus, although response rate is a strong marker of an individual's maturational level, it does not appear to determine the accuracy of immediate recall. These results have important methodological and theoretical implications for human development.

Time does not cause forgetting in short-term serial recall

Time-based theories expect memory performance to decline as the delay between study and recall of an item increases. The assumption of time-based forgetting, central to many models of serial recall, underpins their key behaviors. Here we compare the predictions of time-based and event-based models by simulation and test them in two experiments using a novel manipulation of the delay between study and retrieval. Participants were trained, via corrective feedback, to recall at different speeds, thus varying total recall time from 6 to 10 sec. In the first experiment, participants used the keyboard to enter their responses but had to repeat a word (called the suppressor) aloud during recall to prevent rehearsal. In the second experiment, articulation was again required, but recall was verbal and was paced by the number of repetitions of the suppressor in between retrieval of items. In both experiments, serial position curves for all retrieval speeds overlapped, and output time had little or no effect. Comparative evaluation of a time-based and an event-based model confirmed that these results present a particular challenge to time-based approaches. We conclude that output interference, rather than output time, is critical in serial recall.