Figure 1.
Illustration of the trial structure (Experiment 1).
Figure 2.
Frequency histograms showing the distribution of saliency ratios (average saliency in the eye region divided by the average saliency in the mouth region) for fearful, happy and neutral facial expressions.
Table 1.
Experiment 1: Proportions of correct responses in the emotion classification and the gender discrimination task as a function of presentation time, initial fixation and emotional expression.
Figure 3.
Proportions of fixation changes towards the other major facial feature as a function of task, presentation time, facial expression and initial fixation (Experiment 1).
Error bars indicate standard errors of the mean.
Figure 4.
Illustration of the modulatory effect of facial expression on fixation changes (A,C) and fixation durations (B,D) across all experimental tasks and presentation times (Experiment 1).
The upper panels (A,B) show the values for the whole stimulus set whereas the lower panels (C,D) only depict the respective values for a subset of faces with a comparable saliency ratio in the eye as compared to the mouth region. Error bars indicate standard errors of the mean.
Figure 5.
Heat maps illustrating the normalized fixation time on different face regions for the long presentation time of Experiment 1 (A) and the normalized distribution of saliency (B) as derived from a computational model of bottom-up visual attention [13], [14] for fearful, happy and neutral facial expressions.
Figure 6.
Proportion of time spent fixating either the eye or the mouth region in relation to the time subjects spent fixating the overall face in the long presentation time condition (Experiment 1).
Mean proportions for fixations on the eye or mouth region are shown as a function of task, initial fixation and facial expression with error bars indicating standard errors of the mean. The regions of interest that were used to define fixations in the eye or mouth region, respectively, are shown on the right side.
Table 2.
Experiment 2: Proportions of correct responses in the emotion classification task as a function of vertical placement and emotional expression.
Figure 7.
Proportions of fixation changes towards the eye and the mouth region as a function of the position in the visual field and the facial expression (Experiment 2).
Results are depicted separately for a stimulus set with a similar saliency ratio of the eye as compared to the mouth region across facial expressions (stimulus set 1) and for a stimulus set with a saliency ratio of approximately 1 (stimulus set 2). The regions of interest that were used to define whether saccades targeted the eye or mouth region, respectively, are shown on the right side. Error bars indicate standard errors of the mean.
Figure 8.
Proportion of time spent fixating either the eye or the mouth region in relation to the time subjects spent fixating the overall face (Experiment 2).
Mean proportions for fixations on the eye or mouth region are shown as a function of the position in the visual field and the facial expression. Results are depicted separately for a stimulus set with a similar saliency ratio of the eye as compared to the mouth region across facial expressions (stimulus set 1) and for a stimulus set with a saliency ratio of approximately 1 (stimulus set 2). The regions of interest that were used to define fixations in the eye or mouth region, respectively, are shown on the right side. Error bars indicate standard errors of the mean.