Date of Award

Spring 5-1-1987

Document Type


Degree Name

Doctor of Philosophy (PhD)


Philosophy & Religion


Antes (1974) described two stages of typical picture viewing behavior. First, subjects scanned, making long saccades to highly informative areas which were briefly fixated. Second, informative areas become "bases of operation" from which shorter saccades were made to less informative regions which were fixated for a longer time. The present study addressed anxiety's effects on this pattern. Research suggests that increased arousal, of which anxiety is one type, narrows attention to the most task salient or informative cues (Hockey, 1970). The pattern of attention described by Antes whereby details of progressively less informativeness are attended might therefore be affected by anxiety level. Effects of both between-subjects differences in anxiety (trait anxiety) and within-subjects differences in anxiety (state anxiety) were investigated since research suggests they are distinct. Higher anxiety was predicted to increase (1) fixations to regions rated highly informative by an independent group of 20 subjects and (2) interfixation distance. It was predicted to decrease (1) the total number of slide areas fixated and (2) fixation duration. Forty undergraduates scoring high or low on the trait portion of the Spielberger State Trait Anxiety Inventory (STAI) viewed slides while their eye movements were recorded. Then they described the slides. There were two sessions. One was an hour before a scholastic examination, to increase state anxiety (HISTATE). One was when there was no impending examination (LOSTATE). STAI scores indicated successful manipulation of state anxiety, but neither it nor trait anxiety affected the dependent variables. Analyses were rerun only on the data of subjects reporting increased anxiety in HISTATE. Both STAI scores and pupil diameter data indicated successfully manipulated state anxiety. State anxiety significantly decreased interfixation distance rather than, as predicted, perpetuating the first stage of typical viewing involving long saccades between highly informative areas. Possible explanations for this finding were discussed, including; (1) methodological artifacts; (2) anxiety constricting the visual field and; (3) increased efficiency of viewing whereby shorter saccades more economically scan picture areas. Procedures for exploring these alternatives were discussed. Additionally, discussion included methodological changes for further research of this kind, such as improvements in anxiety measurement, stimulus preparation, and subjects' task.