Abstract
Eye movements can be used as a predictor for various ocular and cognitive conditions. In this study, we propose a new paradigm that can be used to analyze the impact of cognitive load on oculomotor behavior. Participants (N=30) viewed a sequence of 100 scenes freely for 10 seconds. After each scene, they identified a target object from a previous scene among 3 similar distractors from unviewed scenes in a 4 alternative forced choice task. A staircase 2 down 1 up controlled N-back - the number of images back in the sequence from which the target was selected. Our paradigm was successful in actively engaging working memory, as subjects demonstrated more difficulty in recalling correct responses as N-back increased (r(2998)=0.292, p<.001). When comparing the maximum N-back achieved between subjects, there were no significant differences between the number of fixations (F(8,21)=0.848, p=0.572), duration of fixations (F(8,21)=0.693, p=0.694), number of saccades (F(8,21)=0.709, p=0.681), or duration of saccades (F(8,21)=0.279, p=0.966). Similarly, oculomotor behavior did not act as a predictor of correct/incorrect responses with increasing demand from the N-back task. There was no significant interaction between N-back and response accuracy for the number of fixations (F(9)=0.128, p=0.999), the duration of fixations (F(9)=0.385, p=0.943), the number of saccades (F(9)=0.309, p=0.972), or the duration of saccades (F(9)=1.350, p=0.205). When analyzing the total area of each scene viewed, we found no significant differences across groups of maximum N-back reached (F(8)=0.448, p=.878), and proportion of image viewed was not a significant predictor of accuracy (F(9)=0.803, p=0.613). These results suggest that oculomotor strategies generally do not change as a result of higher cognitive demand.