Search

Short-term verbal memory is improved when words can be chunked into larger units. Miller (1956) suggested that the capacity of verbal short-term memory is determined by the number of chunks that can be stored in memory, rather than by the number of items or the amount of information. But how does the improvement due to chunking come about, and is memory really determined by the number of chunks? One possibility is that chunking is a form of data compression. It allows more information to be stored in the available capacity. An alternative is that chunking operates primarily by redintegration. Chunks exist only in long-term memory, and enable the corresponding items in short-term memory to be reconstructed more reliably from a degraded trace. We review the data favoring each of these views and discuss the implications of treating chunking as data compression. Contrary to Miller, we suggest that memory capacity is primarily determined both by the amount of information that can be stored but also by the underlying representational vocabulary of the memory system. Given the limitations on the representations that can be stored in verbal short-term memory, chunking can sometimes allow the information capacity of short-term memory to be exploited more efficiently. (202 words).

Many complex tasks require people to bind individual events into a sequence that can be held in short term memory (STM). For this purpose information about the order of the individual events in the sequence needs to be maintained in an active and accessible form in STM over a period of few seconds. Here we investigated how the temporal order information is shared between the presentation and response phases of an STM task. We trained a classification algorithm on the fMRI activity patterns from the presentation phase of the STM task to predict the order of the items during the subsequent recognition phase. While voxels in a number of brain regions represented positional information during either presentation and recognition phases, only voxels in the lateral prefrontal cortex (PFC) and the anterior temporal lobe (ATL) represented position consistently across task phases. A shared positional code in the ATL might reflect verbal recoding of visual sequences to facilitate the maintenance of order information over several seconds., Highlights • Representing sequences in a short-term memory (STM) task requires a shared positional code between encoding and recall. • We present first multivariate analysis of imaging data from a STM task of sequence position representation. • Most of sensory and association cortices represented positional code within a single task phase. • Only lateral prefrontal cortex and the anterior temporal lobe represented position consistently across task phases.

We contrast two computational models of sequence learning. The associative learner posits that learning proceeds by strengthening existing association weights. Alternatively, recoding posits that learning creates new and more efficient representations of the learned sequences. Importantly, both models propose that humans act as optimal learners but capture different statistics of the stimuli in their internal model. Furthermore, these models make dissociable predictions as to how learning changes the neural representation of sequences. We tested these predictions by using fMRI to extract neural activity patterns from the dorsal visual processing stream during a sequence recall task. We observed that only the recoding account can explain the similarity of neural activity patterns, suggesting that participants recode the learned sequences using chunks. We show that associative learning can theoretically store only very limited number of overlapping sequences, such as common in ecological working memory tasks, and hence an efficient learner should recode initial sequence representations., Author summary The ability to remember multiple individual events as a sequence is necessary for most complex human tasks. There is clear evidence that human sequence learning is accompanied by change in the way sequences are represented in the brain but the exact nature of the change remains unclear. In this study we use brain imaging to ask what is the neural mechanism underpinning sequence learning: we contrast two computational models of learning—associative and recoding—and test their predictions with neural activity data from the dorsal visual processing stream. We provide evidence that, instead of strengthening the initial cortical representations of sequences, learning proceeds by recoding the initial stimuli using a different set of codes. Furthermore, we show that associative learning without recoding is not theoretically capable of supporting long-term memory of short ecological sequences present in every day tasks such as reading, speaking, or navigating.

Neural mechanisms which bind items into sequences have been investigated in a large body of research in animal neurophysiology and human neuroimaging. However, a major problem in interpreting this data arises from a fact that several unrelated processes, such as memory load, sensory adaptation, and reward expectation, also change in a consistent manner as the sequence unfolds. In this paper we use computational simulations and data from two fMRI experiments to show that a host of unrelated neural processes can masquerade as sequence representations. We show that dissociating such unrelated processes from a dedicated sequence representation is an especially difficult problem for fMRI data, which is almost exclusively the modality used in human experiments. We suggest that such fMRI results must be treated with caution and in many cases the assumed neural representation might actually reflect unrelated processes.

The question of whether overt recall of to-be-remembered material accelerates learning is important in a wide range of real-world learning settings. In the case of verbal sequence learning, previous research has proposed that recall either is necessary for verbal sequence learning (Cohen & Johansson Journal of Verbal Learning and Verbal Behavior, 6, 139-143, 1967; Cunningham, Healy, & Williams Journal of Experimental Psychology: Learning, Memory, and Cognition, 10, 575-597, 1984), or at least contributes significantly to it (Glass, Krejci, & Goldman Journal of Memory and Language, 28, 189-199, 1989; Oberauer & Meyer Memory, 17, 774-781, 2009). In contrast, here we show that the amount of previous spoken recall does not predict learning and is not necessary for it. We suggest that previous research may have underestimated participants' learning by using suboptimal performance measures, or by using manual or written recall. However, we show that the amount of spoken recall predicted how much interference from other to-be-remembered sequences would be observed. In fact, spoken recall mediated most of the error learning observed in the task. Our data support the view that the learning of overlapping auditory-verbal sequences is driven by learning the phonological representations and not the articulatory motor responses. However, spoken recall seems to reinforce already learned representations, whether they are correct or incorrect, thus contributing to a participant identifying a specific stimulus as either "learned" or "new" during the presentation phase.

Human bias towards more recent events is a common and well-studied phenomenon. However, recent studies in a wide range of sensory domains have shown that this recency bias persists even when past events contain no information about the future. Reasons for this suboptimal behaviour are not well understood and the internal model that leads people to exhibit recency bias is unknown. Here we use a well-known orientation estimation task to frame the human recency bias in terms of incremental Bayesian inference. We show that the only Bayesian model capable of explaining the recency bias relies on a weighted mixture of past states. Furthermore, we suggest that this mixture model is a consequence of participants’ failure to infer a model for data in visual short term memory, and reflects the nature of the internal representations used in the task. Significance statement Human bias towards more recent events is a common and well-studied phenomenon which occurs even when past events contain no information about the future. In this paper we investigate why such suboptimal behaviour should occur and what it tells us about the underlying mechanisms of perception and memory. Our results reveal that the internal model underlying the recency bias is a weighted mixture of past events. We suggest that this reflects the nature of representations in visual short term memory.