From low invasiveness to high control: how artificial intelligence allows to generate a large pool of standardized corpora at a lesser cost

The use of corpora represents a widespread methodology in interpersonal perception and impression formation studies. Nonetheless, the development corpus using traditional approach involves procedure that is both time- cost-intensive might lead to methodological flaws (e.g., high invasiveness). This turn lower internal external validities Drawing on technological advances artificial intelligence machine learning, we propose an innovative based deepfake technology develop while tackling challenges approach. makes it possible generate synthetic videos showing individuals doing things they have never done. Through automatized process, this allows create large scale at lesser cost short time frame. method characterized by low degree invasiveness given requires minimal input from participants (i.e., single image or video) video person. Furthermore, control over content videos. As first step, referent created which actor performs desired behavior. Then, participant input, will compose are generated specific class learning algorithms such either facial features behavior exhibited transposed face body another In present paper, apply field social skills more specifically studies provide technical information researchers who interested developing technology.