Discrete fixed-resolution representations in visual working memory

Abstract

As well as holding a vast store of long-term memories, the human brain creates short-term memories that last only a few seconds and are essential for performing tasks such as adding two numbers or comparing the attractiveness of two faces. We know that only a limited amount of information can be stored in short-term memory, but whether we store high-quality representations of a small number of items, or a potentially infinite number of 'low resolution' items is the subject of much debate. A new study of visual working memory resolves the matter in favour of the 'high resolution' option: short-term information storage does not discard quality in favour of quantity, but stores a relatively small number of objects, as discrete fixed-resolution representations. Only a limited amount of information can be stored in short-term memory, but it is unclear whether we store high-quality representations of a small number of items or a larger number of items whose representation is of lower quality. Visual working memory is studied, particularly both the number of representations and the resolution of each representation, with the results favouring the idea that we store a smaller number of objects, with relatively discrete, fixed-resolution representations. Limits on the storage capacity of working memory significantly affect cognitive abilities in a wide range of domains1, but the nature of these capacity limits has been elusive2. Some researchers have proposed that working memory stores a limited set of discrete, fixed-resolution representations3, whereas others have proposed that working memory consists of a pool of resources that can be allocated flexibly to provide either a small number of high-resolution representations or a large number of low-resolution representations4. Here we resolve this controversy by providing independent measures of capacity and resolution. We show that, when presented with more than a few simple objects, human observers store a high-resolution representation of a subset of the objects and retain no information about the others. Memory resolution varied over a narrow range that cannot be explained in terms of a general resource pool but can be well explained by a small set of discrete, fixed-resolution representations.