Selective recall of adjectives in clinical anxiety

Abstract

The present experiment examined selective recall of adjectives by clinically anxious subjects. Three groups of subjects, anxious, depressed and controls, were required to listen to 80 adjectives and to decide for each one whether it could describe themselves or another named person. There were two recall conditions of the words; an immediate recall and a delayed 24 hour recall. Subjects were also asked to complete a recognition task in the immediate recall condition. Results failed to support the primary prediction of a negative self-referent recall bias in anxiety that was most evident for threatening material. A difficulty in interpreting results was the finding that there were similar levels of state anxiety in all three groups and high levels of depression in both clinical groups. These findings were interpreted in relation to Beck and Rush's (1975) "danger-schema" model and to findings by Mogg, Mathews and Weinman (1987), the study of which the present investigation replicated. Design limitations were discussed and clinical implications considered.