Fig. I. An incomplete conceptual diagram of information flow in immediate memory tasks. No reference whatever to locations or pathways in the nervous system isimplied. PRELIMINARY DESCRIPTION OF THE MODELThe overall schema in which we are operating is given in Fig. I. Our preliminary assumptions are based on a clear distinction between information which has been categorized (i.e., identified or perceived) and information which has not been categorized. Like Tulving (1968) we see no defensible reason for distinguishing between perception and learning, with respect to individual elements. Once categorization has argument is, on the contrary, that PAS bears important qualitative similarities to the comparable precategorical storage system in vision (Sperling, 1963) though the relevant time parameters appear to be of different order of magnitude. While other writers (e.g., Mackworth, 1965;Neisser, 1967) have previously considered the existence of an acoustic-equivalent to the visual sensory store (and have suggested longer duration in the case of audition than in vision) there has been no comprehensive attempt to make explicit the properties of such a store. In considering these properties, our main objective has been to give an explanation for various serial position data which have been reported in immediate memory. Although the system is in this sense ad hoc it leads readily to a number of testable implications, two of which were confirmed in the experiments reported herein. We shall first describe the PAS system and its properties; second, review the evidence in its favor from both our laboratory and others'; and finally, suggest the relation of the model to a general approach to memory theory.

No abstract

No abstract

Two empirical challenges to the traditional "modal model" of short-term memory are that neither the Brown-Peterson distraetor technique nor the recency effect in recall is well accommodated by that position. Additionally, the status of memory stores as such, has declined in response to proceduralist thinking. At the same time, the concept of coding, on which the modal model is silent, is increasingly central to memory theory. People need to remember things in the short term, but a dedicated store does not need to be the agency.The duration of the debate about short-term memory corresponds almost exactly to the time I have so far spent as a psychologist. Progress has been made, but we are still searching for a way to understand the concept of short -term memory; in this paper, I intend to provide the context for that search.Early suggestions that experimental tasks such as the Brown-Peterson distractor task could provide a direct look at short-term memory were soon replaced by the so-called modal model, originally articulated by Waugh and Norman (1965) and soon refined by Atkinson and Shiffrin (1968). I propose four challenges to that model. The first two challenges are empirical, and the last two are theoretical. I. The distractor task. The concept of short-term storage provided a handy explanation for the rapid loss of information over a few seconds of distraction (Peterson & Peterson, 1959). This suggestion was called into question by the data of Keppel and Underwood (1962), which showed that distraction does not cause any appreciable loss of information on the very first trial of an experiment. The modal model must explain this first-trial performance as somehow reflecting long-term memory (Cowan, 1988, p. 169;Waugh & Norman, 1965). Of course, the verbal item was demonstrably very familiar to subjects before they ever walked into the experimental room; therefore, it was "in" long-term memory, so the modal model must maintain that it was the occurrence of that item in that time and place that is somehow registered in long-term memory. But, such an active state of information in long-term memory is what some workers mean by short -term memory to start with.So, we cannot take performance in the distractor task as an uncontaminated signature of short-term memory. But models are more usually overthrown by better, alternative models than by data. And as it happens, an alternative hypothesis based on temporal distinctiveness (Crowder & Neath, 1991;Glenberg & Swanson, 1986;Johnson, 1991) distractor experiments, including certain temporal features of the task that are quite foreign to the modal model.As an example of the latter, Turvey, Brick, and Osborn (1970) showed that performance sometimes does not decline regularly with the duration of distractor activity in the Brown-Peterson paradigm. Such flat forgetting functions occur when the retention interval is varied between subjects, rather than within subjects, as is usually done. We have recently replicated this finding in an informal classroom project. A point of...

Four hypotheses are given for backward masking by a stimulus suffix of auditory information about the end of an immediate memory list. Masking by erasure (displacement), integration (overwriting), and attentional interruption (diversion of a readout process) was rejected by evidence that (a) multiple suffixes produce a smaller masking effect than single suffixes; (b) a suffix that occurs simultaneously with the last memory item on the list has a smaller effect than one that is delayed by a few hundred milliseconds; and (c) the size of the suffix effect depends on its delay, independent of the rate at which the list was presented. These pieces of evidence are consistent with a masking model based on the operation of lateral inhibition among memory representations of auditory sensory events that are classified by time of arrival and by sensory channel characteristics. A functional model for the suffix effect was constructed from three factors believed to depend on the delay between a memory list and a suffix: (a) integration at short list-suffix delays, (b) backward masking through lateral inhibition at longer delays, and (c) readout of target stimulus information at delays of as long as a second or more.