Purchase this article with an account.
C. Simion, C. Scheier, E. Shimojo, S. Shimojo; What We See Is What We Like — Intrinsic link between gaze and preference. Journal of Vision 2002;2(7):570. doi: 10.1167/2.7.570.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Deciding whether we like someone or something has profound influences on our attitudes and social behavior. Here we investigate how such cognitive/emotional decisions are related to perceptual-motor behavior such as gaze.
Using eye-tracking to measure gaze direction, our past experiments (VSS 01) revealed a bias when subjects compared two faces for attractiveness (faces in a pair were matched for baseline attractiveness). Specifically, the preferred face was looked at for longer time with the bias gradually and significantly increasing from chance level (50%) to 84% just prior to decision, revealing a “cascade effect”. Thus, a cognitive causal pathway (we see an object more because we like it more) and a perceptual-motor pathway (we like it more because we see it more) seem to form a feedback loop, which precedes and ultimately results in, a conscious preference decision.
The present study examines the robustness and generality of this “cascade effect” in several different directions. We show that the effect is preserved, when (a) subjects are asked to compare faces that are significantly different in base attractiveness, (b) when they compare slightly modified versions of the same face, and when (c) the stimuli to be compared are abstract shapes (Fourier-descriptor generated figures) rather than faces.
The “cascade effect” was reflected in a gaze bias curve. Its shape in all cases could be captured with a 4 parameter sigmoid. Moreover, the saturation level of the sigmoid correlates positively with the difficulty of the task. This suggests that the fewer differential or cognitive cues available, the more subjects need to rely on the cascade mechanism to make their decision. We conclude that cognitive and perceptual-motor processes interact in preference judgments, and that the influence of the latter is more significant when cognitive factors are minimized.
This PDF is available to Subscribers Only