Age of avatar modulates the altercentric bias in a visual perspective-taking task: ERP and behavioral evidence

Abstract

Despite being able to rapidly and accurately infer their own and other peoples' visual perspectives, healthy adults experience difficulty ignoring the irrelevant perspective when the two perspectives are in conflict; they experience egocentric and altercentric interference. We examine for the first time how the age of an observed person (adult vs. child avatar) influences adults' visual perspective-taking, particularly the degree to which they experience interference from their own or the other person's perspective. Participants completed the avatar visual perspective-taking task, in which they verified the number of discs in a visual scene according to either their own or an on-screen avatar's perspective (Experiments 1 and 2) or only from their own perspective (Experiment 3), where the two perspectives could be consistent or in conflict. Age of avatar was manipulated between (Experiment 1) or within (Experiments 2 and 3) participants, and interference was assessed using behavioral (Experiments 1-3) and ERP (Experiment 1) measures. Results revealed that altercentric interference is reduced or eliminated when a child avatar was present, suggesting that adults do not automatically compute a child avatar's perspective. We attribute this pattern to either enhanced visual processing for own-age others or an inference on reduced mental awareness in younger children. The findings argue against a purely attentional basis for the altercentric effect, and instead support an account where both mentalising and directional processes modulate automatic visual perspective-taking, and perspective-taking effects are strongly influenced by experimental context.

Keywords: Altercentric interference; ERPs; Self/other; Theory of Mind; Visual perspective-taking.

Fig. 1

Examples of the visual stimuli, showing the age of avatar manipulation, and different configurations of discs on the walls. Note that the avatar’s gender always matched the participant’s gender

Fig. 2

Schematic trial sequence of visual displays presented to participants in the visual perspective-taking task

Fig. 3

Mean response accuracy and response times for each condition in Experiment 1. Error bars show standard errors

Fig. 4

Grand average ERPs over the central occipital lobe elicited by the target image for other consistent, other-inconsistent, self-consistent and self-inconsistent conditions, showing the P100 and P200 for the adult avatar (left panel) and the child avatar (right panel), in Experiment 1

Fig. 5

Grand average ERPs over the central parietal lobe elicited by the target image for other-consistent, other-inconsistent, self-consistent and self-inconsistent conditions, showing the P300 for the adult avatar (left panel) and the child avatar (right panel), in Experiment 1

Fig. 6

Grand average ERPs over the left (left panels) and right (right panels) frontal lobes elicited by the target image for other-consistent, other-inconsistent, self-consistent and self-inconsistent conditions, showing the LFSW for the adult avatar (top panel) and the child avatar (bottom panel), in Experiment 1

Fig. 7

Topographic maps show the ERP waveform for each component of interest. Data for the P100 shows the age of avatar effect, averaged over condition. Data for the P200, P300 and LFSW show the consistency effect (i.e. inconsistent minus consistent), separately for each avatar and perspective condition

Fig. 8

Mean response accuracy and response times for each condition in Experiment 2. Error bars show standard errors

Fig. 9

Mean response accuracy and response times for each condition in Experiment 3. Error bars show standard errors