EMOTIONAL SEMANTIC CONGRUENCY BASED ON STIMULUS DRIVEN COMPARATIVE JUDGEMENTS. FACES IN UPRIGHT/INVERTED ORIENTATION AND NUMBERS

The left-to-right spatial mental representation of emotional valence has been extensively studied in the past decades. Studies based on speeded classification of centrally presented emotions showed a valence-specific lateral bias, characterized by faster left-sided responses for negative emotions (angry face) and faster right-sided responses for positive emotions (happy face). However, it is not clear whether the valence-specific lateral bias occurs in a valence comparison task (CT) between pairs of simultaneously displayed facial expressions (horizontally aligned), with the instruction to choose the most negative/positive. Differently from studies involving single emotion, in the CT there is a lateralization of both the stimuli (one face is presented at the left- and one at the right-side of the screen) and the responses (using left- and right-hands). In the present study, I investigated the possible occurrence of the valence-specific lateral bias in a valence CT pairing facial expressions, belonging to the anger-to-neutral-to-happiness emotional continuum, into two types of stimulus pairs: (1) mixed-facial expressions pairs (i.e, a neutral face paired with a 100% emotional angry/happy face); and (2) complete-facial expressions pairs (i.e., a 50% or 100% emotional face paired with another emotional face of the same emotional intensity, but with the opposite emotional valence). In the Study 1, I demonstrated that in a valence CT (with stimulus self-terminated by the participant response), the lateralized motor reactivity is independent from any valence-specific lateral bias. The motor reactivity resulted to be proportional to the target absolute emotional intensity relative to the cutoff (i.e., the neutral face), irrespective of the response side and the congruency with the spatial arrangement of the pair with the left-to-right spatial mental representation of emotional valence. The occurrence of this bias, namely Emotional Semantic Congruency effect (ESC), is fully consistent with a stimulus-driven theoretical framework and a capture of visual spatial attention due to emotional stimuli. The attentional capture phenomenon is predicted by a direct Speed-Intensity Association (SIA) with a remapping of three source of intensities, all intrinsic in the stimulus pairs, into response speeds. The three sources of intensities are: the target absolute emotional intensity relative to the cutoff, the average emotional intensity of the stimulus pair, and an additive/subtractive constant which formalizes an emotion anisotropy, that produces a general improvement of the performance for relatively positive vs. negative emotion intensities. ESC resulted to be independent on lateralization of emotions and it occurs both under tachistoscopic presentation of stimuli and in indirect task condition in which the valence intensity is task irrelevant. In the Study 2, I demonstrated that ESC is independent from motivational significance of the stimuli, and it is generalizable from the specific domain of emotion (non-symbolic with high motivational significance) to the domain of numbers (symbolic with low motivational significance). In the Study 3, I demonstrated that ESC is independent from the type of stimulus processing (i.e., part-based or holistic) involved in the processing of face. ESC resulted to be independent form inversion of facial expressions (beyond a global slowing down of responses due to the face inversion effect). Furthermore, I investigated the nature of the emotion anisotropy which is cutoff dependent, and it occurs only in the case in which the cutoff need to be extrapolated from image pairs. In conclusion, the present work revealed a general mechanism regulating CT, based on a capture of visual spatial attention by the extremal values of a series. This capture is independent from both the representational domain (emotional vs. numerical) and the type of stimulus processing involved (holistic vs. part-based).