Figure 1.

Schematic overview of five different kinds of face stimuli used to investigate dynamic face perception with their respective characteristics. Characteristics include (from left to right): Naturalness of facial form and motion varying between high (e.g., videos), intermediate (e.g., synthetic facial animation), and low (e.g., point-light faces); control of form and motion varying between high (e.g., synthetic facial animation), intermediate (e.g., photo-realistic rendering for form and image-based morphing for motion) or low (e.g., videos); potential for separating motion from form information (e.g., synthetic facial animation); and technical demand varying from low (e.g., videos), to high (e.g., photo-realistic rendering). For ease of comparison, advantages are colored green, intermediate in yellow and disadvantages in orange. Stimuli are listed in no particular order. While the first four kinds of stimuli are commonly used in face perception research, photo-realistic rendering is the most recent advancement and has not yet entered face perception research. [Sources of example stimuli: Videos: (Skerry and Saxe, 2014); Point-light faces: recorded with Optitrack (NaturalPoint, Inc., Corvallis, OR, USA); Image-based morphing: (Ekman and Friesen, 1978); Facial animation: designed in Poser 2012 (SmithMicro, Inc., Watsonville, CA, USA); Photo-realistic rendering: (Suwajanakorn et al., 2017)].