Inference of personality traits and affect schedule by analysis of spontaneous reactions to affective videos

Abstract

This paper presents a method for inferring the Positive and Negative Affect Schedule (PANAS) and the BigFive personality traits of 35 participants through the analysis of their implicit responses to 16 emotional videos. The employed modalities to record the implicit responses are (i) EEG, (ii) peripheral physiological signals (ECG, GSR), and (iii) facial landmark trajectories. The predictions of personality traits/PANAS are done using linear regression models that are trained independently on each modality. The main findings of this study are that: (i) PANAS and personality traits of individuals can be predicted based on the users' implicit responses to affective video content, (ii) ECG+GSR signals yield 70%±8% F1-score on the distinction between extroverts/introverts, (iii) EEG signals yield 69%±6% F1-score on the distinction between creative/non creative people, and finally (iv) for the prediction of agreeableness, emotional stability, and baseline affective states we achieved significantly higher than chance-level results.

Survey on RGB, 3D, Thermal, and Multimodal Approaches for Facial Expression Recognition: History, Trends, and Affect-Related Applications

“…Usually studies about personality are based on the BigFive personality trait model which is organized along five factors: openness, conscientiousness, extraversion, agreeableness, and neuroticism. While there are works on detecting personality and mood from FEs only [252], [253] the dominant approach is to use multimodality either by combining acoustic with visual cues [252], [258] or physiological with visual cues [259]. Visual cues can refer to eye gaze [260], [261], frowning, head orientation, mouth fidgeting [260], primary FEs [252], [253] or characteristics of primary FEs like presence, frequency or duration [252].…”

First Impressions: A Survey on Vision-Based Apparent Personality Trait Analysis

“…Similarly as the case of personality perception, personality recognition have been addressed in the literature using different data modalities, i.e., still images [50], [96], [97], image sequences [98], [99], audiovisual (with [11], [59], [79], [85], [86], [100], [101], [102], [103] or without [88], [104] interactions) and multimodal [105], [106], [107], [108].…”

“…However, disregarding the analysis performed on the YouTube vlog [55] dataset, any visualbased analysis was performed in relation to the other sources. In [108], they studied the relation between Big-Five traits and implicit responses of people to affective content (i.e., emotional videos), combining features obtained from electroencephalogram, peripheral physiological signals and facial landmark trajectories with a linear regression model.…”

A Survey on Big Five Personality Traits Prediction Using Tensorflow

“…A version of the Brunswick lens indicates the ability of the character traits to expect both activity and interview performance in general when using the two data channels. [17] Previous research recommended generalizing the cultural movement from a five-element structure to personality traits.…”

“…(2) The peripheral physiological functions have much better reciprocal events with the range of the Big Five trends and PANAS than other suggested features (3) Due to the linear strength of the direct family members between (a) EEG and openness and (b) peripheral and diastolic physiological warnings, we ended up with implicit high scores in F1 (about 70%) in predicting a diastole / high/low with simple linear technique [Abadi et al 2015].…”