On the Effect of Observed Subject Biases in Apparent Personality Analysis from Audio-visual Signals

TL;DR

This paper investigates how subjective biases like gender, ethnicity, and attractiveness influence automatic apparent personality perception from audio-visual data, proposing a multi-modal neural network to analyze and mitigate these biases.

Contribution

It introduces a multi-modal deep neural network that combines raw audio-visual data with attribute-specific predictions to estimate apparent personality, and provides an analysis of bias sources affecting predictions.

Findings

Achieves state-of-the-art results on the ChaLearn First Impressions dataset.

Identifies key sources of bias impacting personality perception.

Provides an interpretability framework for bias analysis in personality prediction.

Abstract

Personality perception is implicitly biased due to many subjective factors, such as cultural, social, contextual, gender and appearance. Approaches developed for automatic personality perception are not expected to predict the real personality of the target, but the personality external observers attributed to it. Hence, they have to deal with human bias, inherently transferred to the training data. However, bias analysis in personality computing is an almost unexplored area. In this work, we study different possible sources of bias affecting personality perception, including emotions from facial expressions, attractiveness, age, gender, and ethnicity, as well as their influence on prediction ability for apparent personality estimation. To this end, we propose a multi-modal deep neural network that combines raw audio and visual information alongside predictions of attribute-specific models to regress apparent personality. We also analyse spatio-temporal aggregation schemes and the effect of different time intervals on first impressions. We base our study on the ChaLearn First Impressions dataset, consisting of one-person conversational videos. Our model shows state-of-the-art results regressing apparent personality based on the Big-Five model. Furthermore, given the interpretability nature of our network design, we provide an incremental analysis on the impact of each possible source of bias on final network predictions.