The present investigation used ERPs to detect the activation of implicit stereotypical representations associated to different ethnic groups, by means of an implicit paradigm. 285 sentences were presented to 20 Italian Caucasian participants while EEG signals were recorded from 128 scalp sites. Sentences could either violate (Incongruent condition), non-violate (Congruent condition) or be neutral (Neutral condition) with regard to stereotypical concepts concerning non-Caucasian ethnic groups (e.g. Asians, Africans, Arabs). No awareness or judgment about stereotypes was required. Participants were engaged in a fictitious task, ignoring the overall study's purpose. The results showed that Incongruent terminal words elicited a greater anterior N400 response (300-500 ms) compared to Congruent and Neutral words, reflecting a difficulty in integrating the information incongruent with pre-existing stereotypical knowledge. The participant's individual amplitude values of the N400-Difference Wave (Incongruent-Congruent), showed a direct correlation with the individual racism scores obtained at the Subtle and Blatant Prejudice Scale, administered at the end of the experimental session. Intra-cortical sources explaining the N400 involved areas of the social cognition network such as the medial frontal cortex (BA10) and the inferior temporal gyrus (BA20) which are known to support processing of information about other people and impression formation. Moreover, Congruent terminal words evoked a greater P300 response (500-600 ms) compared to the other conditions, possibly reflecting the merging of incoming inputs with anticipated semantic information. A late post-N400 frontal positivity (650-800 ms) was found to be larger to sentences concerning other-race characters (ether congruent or incongruent) compared to sentences involving own-race characters (neutral). The study corroborated the effectiveness of neurophysiological measures to assess implicit complex semantic representations and circumventing social desirability-related problems.

The present study used EEG/ERPs to detect the activation of implicit stereotypical representations associated to other-race (OR) people and the modulation of such activation through the previous presentation of positive vs. neutral social information. Electrophysiological signals were recorded in 40 Italian Caucasian participants, unaware of the overall study’s purpose. They were presented with 285 sentences that could either violate, non-violate (e.g., “the Roma girl was involved in a robbery) or be neutral with regard to stereotypical concepts concerning other-race people (e.g. Asians, Africans, Arabic). ERPs were time-locked to the terminal words. Prior to the sentence reading task, participants were exposed to a 10 minutes colourful video documentary. While the experimental group was presented a video containing images picturing other-race characters involved in “prestigious” activities that violated stereotypical negative assumptions (e.g. a black neurosurgeon leading a surgery team), the control group viewed a neutral documentary about flora and fauna. EEG signals were then recorded during the sentence reading task to explore whether the previous exposure to the experimental video could modulate the detection of incongruence in the sentences violating stereotypes, as marked by the N400 response. A fictitious task was adopted, consisted in detecting rare animal names. Indeed, only the control group showed a greater N400 response (350–550 ms) to words incongruent with ethnic stereotypes compared to congruent and neutral ones, thus suggesting the presence of a racial bias. No N400 response was found for the experimental group, suggesting a lack of negative expectation for OR individuals. The swLORETA inverse solution, performed on the prejudice-related N400 showed that the Inferior Temporal and the Superior and Middle Frontal Gyri were the strongest N400 intra-cortical sources. Regardless of the experimental manipulation, Congruent terminal words evoked a greater P300 response (500–600 ms) compared to incongruent and neutral ones and a late frontal positivity (650–800 ms) was found to be larger to sentences involving OR than own-race characters (either congruent or incongruent with the prejudice) thus possibly indicating bias-free perceptual in-group/out-group categorization processes. The data showed how it is possible to modulate a pre-existing racial prejudice (as reflected by N400 effect) through exposure to positive media-driven information about OR people. Further follow-up studies should determine the duration in time, and across contexts, of this modulatory effect.

The aim of the study was to investigate the neural processing of congruent vs. incongruent affective audiovisual information (facial expressions and music) by means of ERPs (Event Related Potentials) recordings. Stimuli were 200 infant faces displaying Happiness, Relaxation, Sadness, Distress and 32 piano musical pieces conveying the same emotional states (as specifically assessed). Music and faces were presented simultaneously, and paired so that in half cases they were emotionally congruent or incongruent. Twenty subjects were told to pay attention and respond to infrequent targets (adult neutral faces) while their EEG was recorded from 128 channels. The face-related N170 (160-180 ms) component was the earliest response affected by the emotional content of faces (particularly by distress), while visual P300 (250-450 ms) and auditory N400 (350-550 ms) responses were specifically modulated by the emotional content of both facial expressions and musical pieces. Face/music emotional incongruence elicited a wide N400 negativity indicating the detection of a mismatch in the expressed emotion. A swLORETA inverse solution applied to N400 (difference wave Incong. -Cong.), showed the crucial role of Inferior and Superior Temporal Gyri in the multimodal representation of emotional information extracted from faces and music. Furthermore, the prefrontal cortex (superior and medial, BA 10) was also strongly active, possibly supporting working memory. The data hints at a common system for representing emotional information derived by social cognition and music processing, including uncus and cuneus.

This work presents a supervised machine-learning approach to build an expert system that provides support to the neuroscientist in automatically classifying ERP data and matching them with a multisensorial alphabet of stimuli. To do this, two different approaches are considered: a hierarchical tree-based algorithm, XGBoost, and feedfoward neural networks, highlighting the pros and cons of both approaches in the different steps of the classification task. Moreover, the sensitivity of the classification capabilities of the tool as a function of the number of available electrodes is also studied, highlighting what can be achieved by applying the method using commercial, wearable EEG systems. The main novelty of this work consists in significantly enlarging the pool of stimuli that the expert system can recognize and comprising different, possibly mixed, sensorial domains. The obtained results open the way to the design of portable devices for augmented communication systems, which can be of particular interest for the development of advanced Brain-Computer Interfaces (BCI) for communication with different types of neurologically impaired patients.

No abstract