The attentional deployment task is an ingenious methodology which involves participants giving a neutral response to a neutral stimulus. In the initial version of the task, the participants are first presented with two words on a computer screen, one above the other, for a fixed period of time, although more recent versions of the task have also included faces and pictures. The task with verbal stimuli is to read aloud the top word, while ignoring the lower one. After the screen clears, one of two things happens: either there is a pause and the next trial begins with two more words being presented, or a small dot appears in the space vacated by either the top or the bottom word. If this is the case, the participants have to respond as rapidly as possible by pressing a button. The latency to detect this dot probe is used to assess the location of attentional deployment, the rationale being that participants will be faster to respond to a dot probe appearing in a screen location to which they are currently attending.
The prototypical study using the attentional deployment paradigm with GADs was carried out by MacLeod et al. (1986). MacLeod et al. (1986) presented the task to groups of people with GAD, clinically depressed participants, and normal controls. The results showed that: "clinically anxious, but not clinically depressed, subjects consistently shifted attention towards threat words, resulting in reduced detection latencies for probes appearing in the vicinity of such stimuli. Normal control subjects, on the other hand tended to shift attention away from such material" (p. 15).
Interpretations of such data as these tend to emphasise shifts in attention. However, there are some potential problems with this approach as, by the time the probe stimulus appears (after 525 ms), there has been ample scope for several shifts of attention. Consequently, all we can really say about the results from this task is that participants are more likely to be attending to a screen location where a threat word appeared 525 ms earlier. On this basis an alternative interpretation offers itself; that all participants immediately shifted attention to the location of the threat word but that anxious participants did not shift their attention away again because they continued to attend to it. Because of such criticisms, more recent studies have manipulated the
Stimulus Onset Asynchrony (SOA) between threat and probe. Such studies suggest that GAD individuals show a fast orientation towards threat (Mogg, Millar, & Bradley, 2000), but high-anxious participants with better attentional control then shift their attention away (Derryberry & Reed, 2002) so that by 1250 ms they show primarily avoidance of threatening stimuli (Koster, Verschuere, Crombez, & van Damme, 2005). Attentional bias anxiety models therefore now suggest that GAD and normals differ in their thresholds for initial threat detection such that anxious individuals detect milder threat before subsequent avoidance processes occur (e.g., Mackintosh & Mathews, 2003).
Was this article helpful?