Abstract
Extensive research was and is still conducted regarding symmetry (e.g. Eisenmann, 1967: preference for symmetry). Locher and Nodine (1973) investigated the influences of symmetry or asymmetry on eye movements. In a complexity rating task, they showed that fixation distributions to symmetrical figures were clustered in one-half, whereas there was no bias for asymmetrical figures. Fixation time and numbers did not differ. We hypothesized that eye movement differences to symmetrical or asymmetrical figures depend on the task participants have to fulfil. Twenty participants had to visually inspect same 24 stimuli as in Locher and Nodine (1973) and rate them according to their complexity (Experiment 1) and liking (Experiment 2). In Experiment 3, participants had to memorize and draw each figure. Time until participants rated/started drawing the figures (RTs) and eye movements were recorded. In Experiment 1 participants rated asymmetrical as more complex than symmetrical figures; RTs and fixation numbers did not differ. Ratings in Experiment 2 did not differ, but RTs and fixation numbers for asymmetrical were significantly higher than for symmetrical figures. In Experiment 3 RTs and fixation numbers for asymmetrical were significantly higher (almost doubled) than for symmetrical figures. There was no general preference for symmetrical figures. We could replicate findings by Locher and Nodine (1973), namely that time and number of fixations did not differ in a complexity task, however, time and number of fixations did differ enormously in the liking and drawing task (similar to results by Milisavljevic et al., 2011). Preliminary scan path analysis revealed in none of the tasks a general bias to one-half in symmetrical figures, but there were significantly more trials with biases towards one-half in Experiment 2 compared to the other tasks. Our results showed that different task demands influence eye movement patterns during the perception of symmetrical figures.
Meeting abstract presented at VSS 2014