ABSTRACT
In early evolutionary history, as organisms developed the ability to move through and act on their environment, and to use perception to guide these actions, advantage was surely conferred on those that could, when appropriate, delay their actions relative to the perceived events that prompted them. For example, in some instances the optimal response to perceiving a source of food may not be to instantaneously pounce, but, rather, to withhold that action until the prey re-emerges from the hole into which it has disappeared. Thus, the ability to control actions, including the ability to guide actions with information no longer accessible to sensory receptors, likely evolved as a core function of perception–action systems. This perspective highlights the fact that such terms as attention, response inhibition, and the topic of this volume, working memory, are constructs that have been developed by scientists trying to reverse engineer a highly evolved neurocognitive system. Although unquestionably useful for the categorization and description of certain types of behaviour and/or neurophysiology, they need not correspond to discrete systems that can be meaningfully understood in isolation from the larger system within which they are observed. Rather, they may be better construed as properties or functions of an integrated perception–action system. The focus of this chapter will be on the short-term retention of spatial information that temporally bridges the offset of the sensory stimulus and the onset of the action that is guided by that sensory information. This is typically operationalized experimentally by studying the delay period of tests of delayed recognition and delayed response.
