Attention mediates the flexible allocation of visual working memory resources.

Abstract

Though it is clear that it is impossible to store an unlimited amount of information in visual working memory (VWM), the limiting mechanisms remain elusive. While several models of VWM limitations exist, these typically characterize changes in performance as a function of the number of to-be-remembered items. Here, we examine whether changes in spatial attention could better account for VWM performance, independent of load. Across 2 experiments, performance was better predicted by the prioritization of memory items (i.e., attention) than by the number of items to be remembered (i.e., memory load). This relationship followed a power law, and held regardless of whether performance was assessed based on overall precision or any of 3 measures in a mixture model. Moreover, at large set sizes, even minimally attended items could receive a small proportion of resources, without any evidence for a discrete-capacity on the number of items that could be maintained in VWM. Finally, the observed data were best fit by a variable-precision model in which response error was related to the proportion of resources allocated to each item, consistent with a model of VWM in which performance is determined by the continuous allocation of attentional resources during encoding. (PsycINFO Database Record

Stochastic sampling provides a unifying account of visual working memory limits

“…Here we have focused on experiments in which each item in an array is equally likely to be tested and so has equal priority for representation. However, the view of WM as a limited resource has also been motivated by studies showing that prioritized items can be represented more precisely in WM, at the cost of decreased precision for other items [7,[19][20][21][22]. The population coding model has previously been shown to reproduce data from tasks that encourage unequal allocation of resources (e.g.…”

“…3). Importantly, discrete representations are compatible with an underlying continuous memory resource that can be distributed according to behavioral goals [25][26][27]: indeed in the stochastic model the integer number of samples available for each item at retrieval is unpredictable, and so cannot be the basis of prioritization. Instead, the resource distributed between items corresponds to the mean or expected total number of samples, which is constant and continuous-valued -in the neural model [2] this is equated with the instantaneous firing rate or membrane potential, while decoding is based on the expression of this rate in discrete spikes.…”

Recent evidence that attention is necessary, but not sufficient, for conscious perception

“…Many influential attention papers suggest that the principal function of attention is to be a filter on sensory input, narrowing incoming stimuli down to a manageable stream of information that can be processed effectively with the brain's limited resources . The precise resource being competed for, however, has not been thoroughly characterized . Attention has also been proposed to be a mechanism that supports the flexible routing of information through the brain .…”

“…[39][40][41] The precise resource being competed for, however, has not been thoroughly characterized. 42 Attention has also been proposed to be a mechanism that supports the flexible routing of information through the brain. 43 Alongside its ability to select information for elaborated processing, attention is also thought to have the ability to suppress distracting or unneeded information.…”

Electrophysiological correlates of the flexible allocation of visual working memory resources

“…In this way, the proportion of attentional resources allocated to any 1 given memory item can continuously vary anywhere between 0 and 100%. In these past 2 studies 18,19 , it was found that working memory performance (i.e., raw error = 1/precision) was 3 best predicted by the likelihood that an item would be probed on a given trial, independent of the 4 overall memory load. Importantly, this relationship between probe likelihood and memory 5 precision, which followed a power-law, could also account for changes in performance across 6 loads; for example, the change in precision from one to two items was consistent with each item 7 only receiving half as many attentional resources.…”