Effects of State Anxiety on Performance Using a Task-Switching Paradigm: An Investigation of Attentional Control Theory
Derakshan, Nazanin, Smyth, Sinéad, Eysenck, Michael W., Psychonomic Bulletin & Review
Low- and high-anxious participants performed arithmetical tasks under task-switching or nontask-switching conditions. These tasks were low or high in complexity. The task on each trial was either explicitly cued or not cued. We assumed that demands on attentional control would be greater in the task-switching condition than in the nontask-switching condition, and would be greater with high-complexity tasks than with low-complexity ones. We also assumed that demands on attentional control would be greater when cues were absent rather than present. According to attentional control theory (Eysenck, Derakshan, Santos, & Calvo, 2007), anxiety impairs attentional control processes required to shift attention optimally within and between tasks. We predicted that there would be greater negative effects of high state anxiety in the task-switching condition than in the nontask-switching condition. Our theoretical predictions were supported, suggesting that state anxiety reduces attentional control.
The processing efficiency theory (PET; Eysenck & Calvo, 1992) was developed to explain the frequent adverse effects of anxiety on the performance of complex cognitive tasks. According to PET, high levels of state anxiety reduce the efficiency of cognitive processing and often lead to impaired performance. Specifically, anxiety affects the functioning of the working memory system which, according to Baddeley's (1986) original conception, consisted of the central executive (a limited capacity, attention-like system), the phonological loop (used for verbal rehearsal), and the visuospatial sketchpad (used for processing visual and spatial information). According to PET, the component of working memory most affected by anxiety is the central executive.
Support for PET comes from research showing that anxiety impairs performance when two concurrent tasks both use the central executive. Eysenck, Payne, and Derakshan (2005) conducted a systematic investigation of the effects of anxiety on the different components of working memory using the Corsi Blocks Test. This task was performed concurrently with different secondary tasks, each utilizing a different component of working memory. Adverse effects of anxiety on the Corsi task were obtained only when the secondary task required use of the central executive, suggesting that anxiety reduces its available capacity.
In spite of much empirical support for PET, Eysenck, Derakshan, Santos, and Calvo (2007) highlighted various limitations in that theory. Importantly, PET failed to specify the executive function(s) affected by anxiety. There is ongoing debate concerning the number and nature of such functions. However, in a systematic investigation, Miyake et al. (2000) used latent variable analysis to identify three major control functions of the central executive: inhibition, shifting, and updating. The inhibition function often involves using attentional control in a negative way to prevent attentional resources being allocated to taskirrelevant stimuli and responses (Friedman & Miyake, 2004). The shifting function involves using attentional control in a positive way to shift the allocation of attention to maintain focus on task-relevant stimuli. According to attentional control theory (ACT; Eysenck et al., 2007), anxiety impairs the efficiency of the inhibition and shifting functions. There is direct support for the prediction that anxiety impairs the inhibition function (Derakshan, Ansari, Hansard, Shoker, & Eysenck, 2009).
Miyake et al. (2000) reported that performance on taskswitching paradigms loads highly on the shifting function. Task-switching paradigms involve comparing performance on blocks of trials involving one task (A or B) with performance on blocks of trials with 50% of trials involving Task A and 50% Task B. Participants exposed to a task-switching procedure generally demonstrate a switching cost (Monsell, 2003). There is hardly any published research on the effects of anxiety on the shifting function. …