Measuring the Emotional Conflict with a Word-Face Stroop Task

Abstract

Event Abstract Back to Event Measuring the Emotional Conflict with a Word-Face Stroop Task Zeynep Basgöze1* and D. Gökçay1 1 Informatics Institute, Middle East Technical University, Turkey Background: Conflict resolution is essential for humans to survive. Stroop Task is a cognitive task designed to measure conflict resolution based on congruent and incongruent situations. Cognitive conflict resolution occurs in DLPFC and caudal ACC parts of the brain [1]. On the other hand, when emotions are involved, the limbic areas such as rostral & subgenual ACC and amygdala take over [1, 2]. Currently, emotional conflict is being measured by creating a conflict between emotional words and their color or emotional words and their number. However, these tasks fail to generate emotional conflict consistently, because they activate both cognitive and emotional parts of the system. To be able to measure emotional conflict, one needs to create a task where the conflict activates emotional networks exclusively. The word-face Stroop [1], which we conducted in this study, is such a task where the conflict is between “emotional” words and “emotional” faces. In congruent situations, positive words are shown on positively affective faces; negative words are shown on negatively affective faces; whereas in incongruent situations positive words are shown on negatively affective faces and vice versa. The novelty of our task in this study is due to the use of a brand-new Turkish affective word list, TUDADEN [4], and evaluation of valence as well as concreteness of the words. Method: 19 healthy right handed subjects (8 F, 11 M, age 25±3.8) voluntarily participated in the experiment. Emotionally normed words were selected from TUDADEN database [4]. 96 words are used: 32 positive, 32 negative, 32 neutral; 16 abstract and 16 concrete words existed for each category. Positive and negative words had 16 congruent and 16 incongruent situations each. Neutral words had no incongruent situation, serving as a baseline. Faces in the background (4 happy, 4 sad, 4 neutral, each from 2 M 2 F) were chosen from “The Productive Aging Lab Face Database” [3]. The subjects were asked to make a decision about the valence (positive, negative or neutral) of the emotional words which appeared on a background showing these affective faces. Each subject responded to three types of trials: congruent cases, incongruent cases and neutral cases (neutral word on a neutral face). They first saw a fixation point during 1500 ms, and then observed the stimuli during 2000 ms. Results: A 3×2 ANOVA with factors of congruency, valence, concreteness (2×2×2) showed that subjects differed significantly in RT’s of congruency [F(1,18)=9.467, p<0.01] and valence [F(1,18)=5.476, p<0.05]. A barely significant effect of interaction between congruency, valence and concreteness is also found [F(1,18)=3.886, p=0.064], but there is no significant effect of words’ concreteness, nor of responses’ correctness. Conclusion: Our task showed that healthy people are slower in reacting to incongruent stimuli compared to the congruent ones, just as in classical Stroop, but here, in an emotional state. Moreover, reaction times also differ according to the valence of the words: People are slower when the words are negative compared to positively valenced words.