Sustained and Transient Attention in the Continuous Performance Task

Abstract

One of the most frequently applied methods to study abnormal cognition is the Continuous Performance Task (CPT). It is unclear, however, which cognitive functions are engaged in normal CPT performance. The aims of the present study were to identify the neurocognitive functions engaged in the main variants of the CPT and to determine to what extent these variants differentially engage these functions. We hypothesized that the main CPT versions (CPT-X, CPT-AX, CPT-Identical Pairs) can be distinguished by whether they demand sustained or transient attention and sustained or transient response preparation. Transient attention to objects like letters or digits, that is, the need to switch attention to different objects from trial to trial, impairs target detection accuracy relative to sustained attention to a single object. Transient response preparation, that is, the possibility to switch response preparation on and off from trial to trial, improves response speed relative to having to sustain response preparation across all trials. Comparison of task performance and Event-Related brain Potentials (ERPs) of healthy participants obtained in the main CPT variants confirmed these hypotheses. Behavioral and ERP measures indicated worse target detection in the CPT-AX than in the CPT-X, consistent with a higher demand on transient attention in that task. In contrast, behavioral and ERP measures indicated higher response speed in the CPT-AX than in the CPT-X, associated with more response preparation in advance of the targets. This supports the idea of increased transient response preparation in the CPT-AX. We conclude that CPTs differ along at least two task variables that each influences a different cognitive function.

Notes

¹We use the concept of selective attention here in the sense of voluntary orienting of attention to a location in space or an object feature, on the basis of a verbal instruction or a cuing stimulus. Recent studies have shown that voluntary selective attention to a particular stimulus and its demands produces prolonged activity in parietal areas of cortex in advance of stimulus presentation, both for verbally specified stimuli and for stimuli specified by a cuing stimulus (Corbetta & Shulman, 2002). This activity was furthermore independent both of activity produced by the cue and that produced by the target stimulus, and therefore reflects the activity of top-down control processes. If instructions vary from trial to trial, selective activity is varied due to top-down control mechanisms, and therefore is in a transient mode. When instructions are the same across all trials of a task, selective activity is maintained as a constant across trials, and therefore is in a sustained mode.

²In terms of models of attention (for example, Fan, McCandliss, Sommer, Raz, & Posner, 2002), variation of the transient/sustained attention dimension would influence the voluntary orienting of attention, the executive system, and possibly also the vigilance system, because different types of stimuli (brightness versus color) may differ in alertness-evoking capacity. In terms of models of working memory (Baddeley, 2001), transient and sustained attention would influence the central executive, but also the visual spatial scratch pad and phonological short term memory systems. Transient attention may lead to switching between activity patterns in all three of these working memory components. Response preparation is a result of executive working memory mechanisms that make use of the information maintained to plan future behavior.