Purchase this article with an account.
Eve Isham, Joy Geng; Visual fixation parameters predict decisional outcomes better than preference. Journal of Vision 2013;13(9):798. doi: https://doi.org/10.1167/13.9.798.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Although visual fixations are commonly used to index stimulus-driven or internally-determined preference, recent evidence suggests that visual fixations can also be a source of decisional bias that moves selection toward the fixated object (Krajbich et al, 2010). These contrasting results raise the question of whether visual fixations always index comparative processes during choice-decision tasks, or whether they might better reflect internal preferences when the decision does not carry any economic or corporeal consequences (Rangel et al, 2008). In two experiments, participants were shown pairs of novel black and white patterns and asked to choose which was more aesthetically pleasing (Exp.1; N=20) or appeared more "organic" (Exp.2; N=20). Participants also provided independent aesthetic (i.e., preference) ratings of the stimuli. The behavioral responses were used to classify each stimulus as chosen or unchosen, and as being rated higher, lower, or the same as its pair. Total fixation durations were subjected to a 2 (chosen vs unchosen) x 3 (higher, lower, same) within-subjects ANOVA. Our results demonstrated that longer fixation durations were correlated with the chosen item (p<.001), but not the more preferred item (p>.1). Moreover, a significant difference in cumulative fixation times on the subsequent choice was observed as early as 1000 to 2000 ms before the choice was actually indicated with a button press. Similarly, final fixations were more likely to be on the chosen object than the unchosen one (p<.001). Collectively, these results suggest that fixation parameters such as total looking durations and final fixations are a better index of choice than of aesthetic preference when participants must make an explicit choice.
Meeting abstract presented at VSS 2013
This PDF is available to Subscribers Only