The role of affective awareness in a social categorization task: behavioural and autonomic data

Affective neuroscience research posits that non-conscious stimuli evoke emotion states and physiological indexes also when individuals are not visually aware of them. In this study we tested in 34 Italian participants whether subliminal and supraliminal affective priming may influence the attribution of faces to a social group. Participants' face temperature (peri-orbital region and nasal tip) during the task was measured by means of functional infrared thermal imaging (fITI). This technique allows skin temperature recording by tracking changes in facial temperature with high thermal resolution. The emotional visual stimuli used for the affective priming were taken from the International Affective Picture System and the neutral faces employed in the social categorization task were taken from three validated face sets. To prevent visual stimuli to be consciously perceived, we employed the forward and backward masking technique: the target image was presented for 33 ms and it was preceded and followed by visual masks created by scrambling the target itself. During the supraliminal block the target image was presented for 500 ms. In the subsequent social categorization task, participants were asked to report whether the neutral face belonged to an in-group (i.e., Italian) vs out-group (i.e., Romanian) individual. We fit behavioral and thermal data in a mixed model logistic regression predicting the probability of categorizing the face as in/out-group. A significant two-way interaction between negative valence and temperature (200-600 ms time bin) was present. In the negative subliminal condition the increase in orbital temperature - indexing the engagement of the ANS sympathetic division - predicted out-group categorization, while its decrease - indexing the engagement of the ANS para-sympathetic division - predicted in-group categorization. By contrast, in the negative supraliminal condition, the increase in orbital and nose temperature predicted the categorization of faces as in-group, while its decrease predicted out-group categorization. In keeping with assimilation and contrast effects in priming research, target categorization in the subliminal block was valence-driven, whereas in the supraliminal block, target categorization was opposite respect to the valence of the prime. As an emotional regulation index, we computed the difference in orbital temperature when categorizing the face as out-group vs in-group in the negative supraliminal condition. The difference was negatively correlated with the frequency of in-group choices suggesting that higher sympathetic activity may underpin the aversion to include strangers in one's own social group. Inclusion behavior seems to be affected by differences in emotion regulation: the tendency to transfer priming-driven affective reactions to unrelated social decisions is in fact more pronounced in individuals who show stronger increases of peri-orbital temperature when facing threat-related stimuli.