Emotion Words' Effect on Visual Awareness and Attention of Emotional Faces

Abstract

To explore whether the meaning of a word changes visual processing of emotional faces (i.e., visual awareness and visual attention), we performed two complementary studies. In Experiment 1, we presented participants with emotion and control words and then tracked their visual awareness for two competing emotional faces using a binocular rivalry paradigm. Participants experienced the emotional face congruent with the emotion word for longer than a word-incongruent emotional face, as would be expected if the word was biasing awareness toward the (unseen) face. In Experiment 2, we similarly presented participants with emotion and control words prior to presenting emotional faces using a divided visual field paradigm. Emotion words were congruent with either the emotional face in the right or left visual field. After the presentation of faces, participants saw a dot in either the left or right visual field. Participants were slower to identify the location of the dot when it appeared in the same visual field as the emotional face congruent with the emotion word. The effect was limited to the left hemisphere (RVF), as would be expected for linguistic integration of the word with the face. Since the task was not linguistic, but rather a simple dot-probe task, participants were slower in their responses under these conditions because they likely had to disengage from the additional linguistic processing caused by the word-face integration. These findings indicate that emotion words bias visual awareness for congruent emotional faces, as well as shift attention toward congruent emotional faces.

Keywords: binocular rivalry; divided visual field; emotion; emotion words; emotional faces; semantic priming; visual attention; visual awareness.

FIGURE 1

Stimuli used in Experiment 2 (includes stimuli for Experiment 1).

FIGURE 2

Paradigm and timing for Experiment 1. Frowning face on the left and scowling face on the right. In separate trials, the same two faces also followed the word “sad” and a control word. Each face appeared on the right and left, across trials. Stimuli images are from IAS Lab face set (http://www.affective-science.org) and depict adult participants from Boston College. Participants in this face set gave written and informed consent, including explicit consent to be photographed and their likeness to be reproduced.

FIGURE 3

Paradigm and timing for Experiment 2. Scowling face on the left and frowning face on the right. In separate trials, the same two faces followed the word “sad” and a control word. Each face appeared on the right and left, across trials. In addition, the dot appeared on the right and left, across trials. Stimuli images are from IASLab face set (http://www.affective-science.org) and depict adult participants from Boston College. Participants in this face set gave written and informed consent, including explicit consent to be photographed and their likeness to be reproduced.

FIGURE 4

Reaction time differences between pair types by visual field for Experiment 2. There was a significant effect for the primed trials but only in the RVF (left-hemisphere) between emotion words and control words, in line with the hypothesis that participants would be slower to respond due to elaborative linguistic processing to trials preceded by an emotion word that is able to be integrated with the face.

FIGURE 5

Reaction time differences between pair types by visual field for Experiment 2. There was a significant effect for the primed trials but only in the RVF (left-hemisphere) between emotion words and control words, in line with the hypothesis that participants would be slower to respond due to elaborative linguistic processing to trials preceded by an emotion word that is able to be integrated with the face.