Testing for the “Blues”: Using the Modified Emotional Stroop Task to Assess the Emotional Response of Gorillas

SIMPLE SUMMARY: We presented three adult male zoo-housed gorillas (Gorilla gorilla gorilla) with a modified emotional Stroop task on a computer touchscreen as a way to assess gorillas’ attentional shift for positively versus negatively valenced items. The gorillas learned to respond to a blue border and to withhold responding to a yellow border. Images of items assumed to have a positive or negative valence for the gorillas were placed within the borders. As predicted, gorillas touched the reinforced blue border more slowly when ‘negative’ images appeared within the blue border, compared to when ‘positive’ images appeared within the yellow border. However, gorillas’ accuracy did not vary as a function of which images appeared within the blue border, possibly because of the high levels of performance on all trials. These results validate the procedure to some degree for assessing the emotional valence of familiar stimuli. ABSTRACT: We adapted the emotional Stroop task developed for primates to test whether gorillas would show response slowing for images of ‘negative’ compared to images of ‘positive’ items placed within previously reinforced borders. Three zoo-housed male gorillas participated in six phases of an emotional Stroop paradigm. In Phase One, they learned to select blue borders over yellow borders in a forced choice task presented on the touchscreen. In Phase Two, neutral yellow or blue two-dimensional shapes were placed within the borders. On congruent trials, blue images were presented within both blue and yellow borders. On incongruent trials, yellow images were placed within both blue and yellow borders. We continued to use these trials as control trials in subsequent phases. We predicted that response latencies would be slower and accuracy would be lower on incongruent trials. Although the gorillas responded more quickly to incongruent trials, in contrast to predictions, they were more accurate on congruent trials, consistent with predictions. Therefore, we proceeded with Phase Three in which photographs of images assumed to have positive and negative valences for the gorillas were placed within the borders. On test trials, the same positive or negative image was placed within both borders. In Phase Four, a positive image was paired with a negative image on each trial and the positive image appeared in either the blue (congruent trials) or yellow border (incongruent trials). Phases Five and Six replicated Phases Three and Four with images of novel positive and negative items. The gorillas responded more quickly on congruent trials compared to incongruent trials on test trials but not on control trials throughout Phases 3–6. These findings provide some validation for the emotional Stroop task to test attentional shift with emotionally valenced items.