Implicit Association Task as measure of threat-related information processing in social phobia and panic disorder

Elbers, Diony (Stellenbosch : University of Stellenbosch, 2005-03)


The main objective of the study was to adapt the Implicit Association Task (IAT) to asses implicit self-relevant fear associations in individuals suffering from social phobia and panic disorder. This involved the development of computerbased word stimuli classification tasks in which participants were expected to classify individually presented words belonging to one of four word categories, namely self-related ‘me’ words, other-related ‘not-me’ words, threat-related words (physical or social threat) and corresponding safety-related words. Two response keys on the computer were to be used, each representing two word categories during a specific trial (e.g., the one representing ‘me’ and ‘threat’, and the other ‘not-me’ and ‘safety’ words). The demanded task was to classify the presented words as quickly and accurately as possible. This resulted in the construction of the Physical Threat Implicit Association Task (PIAT) and the Social Threat Implicit Association Task (SIAT). Both IAT versions were administered to a group of 17 participants diagnosed with social phobia, 17 diagnosed with panic disorder, and 17 ‘normal’ controls. Fear-domain specific self-threat association biases were expected for the social phobics on the SIAT, for the panickers on the PIAT, as well as significant differences with the performances of the control group on the IAT tasks. A secondary objective of the study was to investigate the relationships between the IATs and performances on a variety of self-report scales, namely the Social Phobia Inventory, the Panic Disorder Severity Scale, the State-Trait Anxiety Inventory, and the Beck Depression Inventory-II. In contrast to what was expected, the results of both the PIAT and SIAT did not demonstrate a facilitation of the classification task during trials when ‘me’ and ‘threat’ words were allocated to one category (i.e., response key), and ‘notme’ and ‘safety’ to the other. On the contrary, all three participant groups demonstrated significant effects in the opposite direction. Furthermore, the differences between the groups on both IATs were insignificant. With the exception of a significant, negative correlation between the results on the SIAT and the Social Phobia Inventory for the social phobia group, all the other IAT and self-report scale correlations were insignificant. The results were explained in terms of a newly proposed ‘two-forces’ cognitive theory. It was speculated that the IAT effects might have been the result of two opposing forces operating at different stages of the information processing system. This is namely (a) a disruption of performance by attention diversion during an early pre-attentive stage of processing, versus (b) a facilitation of the classification task by implicit association during later elaborative stages of processing, with the former apparently making the major contribution to the final IAT effect. This may be a phenomenon unique to anxiety disorders. The implications for future research of the findings and the newly proposed theory were also discussed.

Please refer to this item in SUNScholar by using the following persistent URL:
This item appears in the following collections: