COGNITION AND EMOTION 2002 16 3 331 354 Induced processing biases have causal effects on anxiety Andrew Mathews MRC Cognition and Brain Sciences Unit Cambridge UK Colin MacLeod University of Western Australia Nedlands Australia After briefly describing the nature of emotional processing biases associated with vulnerability to anxiety and a model of how they may be produced we review new data on the experimental induction of attentional and interpretative biases We show that these biases can be readily induced in the laboratory in the absence of mood changes However induced biases exert effects on the processing of new information and cause congruent changes in state anxiety when they influence how emotionally significant information is encoded We can therefore conclude that biases have causal effects on vulnerability to anxiety via their influence on how significant events are processed Finally we discuss how our model might account for the acquisition of processing bias and for when they can influence anxiety The nature of anxiety linked processing biases In everyday life we commonly encounter events that have both negative and positive aspects or are ambiguous in their emotional meaning Which aspect we attend to or how we interpret such ambiguity is associated with enduring tendencies to experience positive or negative emotional states such as anxiety as well as current mood for a review see Mathews MacLeod 1994 Thus when two or more stimuli or meanings compete for processing resources people with high levels of trait anxiety particularly when under stress are more likely than are others to attend to a negatively valenced or threatening stimulus Correspondenc e should be addressed to Andrew Mathews MRC Cognition and Brain Sciences Unit 15 Chaucer Road Cambridge CB2 2EF UK e mail andrew mathews mrc cbu ac uk or Colin MacLeod Department of Psychology University of Western Australia Nedlands WA 6907 Australia e mail colin psych uwa edu a u Several of the experiments reported here were carried out by Lynlee Campbell as part of her doctoral research work We thank Bundy Mackintosh and Tom Borkovec for helpful theoretical discussions 2002 Psychology Press Ltd http www tandf co uk journals pp 02699931 html DOI 10 1080 02699930143000518 332 MATHEWS AND MacLEOD and are more likely to adopt a threatening interpretation of ambiguous information For example if two words differing in their valence are displayed simultaneously and are followed by a to be detected target high trait anxious individuals under stress are typically faster to detect targets in the prior location of a threatening word In low trait anxious individuals this effect is typically absent or sometimes reversed even under stressful conditions e g MacLeod Mathews 1988 Of course it would not be adaptive for anyone regardless of trait anxiety to ignore all threatening information because many events require attention and positive action to avert real danger Indeed it seems self evident that our survival depends on responding to major threats to personal safety The distinction between those with high versus low trait anxiety thus refers to their responsiveness to more minor threat cues that do not signal dangers requiring urgent action There are a number of ways in which one could try to account for this difference in responsiveness One possibility is that there is some kind of threat evaluation process with a threshold that must be exceeded before the cognitive system shifts from a mode in which threat cues can be ignored to one in which they are attended If so then it could be the threshold level at which this shift occurs that is associated with vulnerability to anxiety As indicated earlier there is evidence showing that attention to minor threat cues vigilance is revealed by stress in high but not in low trait anxious individuals MacLeod Mathews 1988 Differences between high and low trait anxious individuals are not confined to experiments involving words Equivalent results have been found using displays of two faces varying in emotional expression Mogg Bradley 1999 but see Mansell Clark Ehlers Chen 1999 for discrepant results in social anxiety or with other pictures varying in their threatening content Yiend Mathews 2001 This last study reported experiments in which single pictures varying in emotional valence were presented in one location followed by a target in the same or in another location This method allows a distinction to be drawn between the speeding due to attentional engagement with a picture if the target appears in the same location and slowing when participants must disengage attention to find a target elsewhere Evidence from this last study and from other similar research e g Compton 2000 Fox Russo Bowles Dutton 2001 strongly suggests that these effects arise because anxiety prone individuals have greater difficulty in disengaging attentional resources from threatening information Related interference effects occur when threatening stimuli must be ignored in order to perform a task efficiently For example in highly anxious individuals search for a neutral target is slowed if it is embedded among threatening distracters Mathews May Mogg Eysenck 1990 Anxious individuals also show greater interference when colour naming threatening words the emotional INDUCED PROCESSING BIASES HAVE CAUSAL EFFECTS 333 Stroop effect see Williams Mathews MacLeod 1996 This effect persists even when the words are displayed very briefly and are followed by an obscuring mask to restrict awareness e g MacLeod Rutherford 1992 Mogg Bradley Williams Mathews 1993 Stroop interference from masked words has also been found in one study to predict later distress following a disturbing life event MacLeod Hagan 1992 Both faster detection of targets replacing threatening words and related interference effects may thus arise from the same basic process initiated outside awareness in which mildly threatening stimuli capture attentional resources more readily in high trait anxious individuals When reading about events that are ambiguous in terms of their emotional implications individuals having low scores on measures of trait anxiety are more likely to arrive at positive interpretations whereas those with high scores appear to be more even handed or to favour more threatening interpretations Thus after hearing the sentence The doctor examined little Emma s growth nonanxious controls were more likely than anxious patients to endorse related
View Full Document
Unlocking...