Visual perspective taking (VPT) refers to the ability to comprehend what another person can see from their viewpoint. It’s a skill we first develop when we master the art of “hide and seek.” When finding the best hiding spot, children need to consider where they will be least likely to be seen. They need to imagine what the “seeker” will see as they walk around looking for the “hider.”

VPT also plays a big role in the world of magic and illusions. In 1983, David Copperfield magically made the Statue of Liberty disappear and then reappear. Surely, one person couldn’t take Liberty from an entire country (including a live audience), but he was able to convince everyone that he did it simply by shifting their perspective.

On a day-to-day basis, we use VPT as an important factor in social interactions. By understanding what other people can see, we have a better understanding of what they know, which can be useful in conversation or other shared tasks. In order to have a successful exchange, sometimes we need to set aside our own perspective and temporarily adopt the other person’s perspective. But how do we do this? Do we create multiple perspectives at the same time, or do we need to switch between different perspectives one at a time?

Dorit Segal (pictured above) tested this question in an article published in Psychonomic Bulletin & Review. Segal used a visual scenario experiment, in which an avatar faces a rectangular box hanging from the ceiling (piñata style). The avatar could stand on any side of the box, and the participant had to judge whether the avatar saw a square or a rectangle (based on the side of the box it could see). The avatar changed positions throughout the experiment. Half the trials required the same response as the previous trial, while the others required a change in response.

Segal measured response time and accuracy. She found that responses for “stay” trials (where the avatar stays in the same position as the previous trial) were significantly faster than “switch” trials (where the avatar moves to a different side). This effect represents a switching cost. That is, switching perspectives takes more time than maintaining the same perspective across trials.

There was also an effect of avatar position. Participants responded fastest and most accurately when the avatar was in front, sharing the same perspective as the participant. Responses were slower for the back, left and right positions (in that order). The right position had the slowest and least accurate responses due to the large orientation disparity that required the participant to make the greatest mental transformation.

Previous research has entertained the idea of co-activation of multiple perspectives. Co-activation could be an efficient default because the participant could have multiple prepared responses to choose from. This is especially helpful in situations with high task uncertainty. For example, will the avatar stay in the same position or switch to a different position? To examine this, Segal compared “stay” trials to “single” trials, which is where the avatar stayed in the same position for an entire block of trials. There was no significant difference between “stay” and “single” trials, suggesting there is no cost to mixing types of trials in the same block.

Together, these results point to the cost of perspective switching, and they suggest that people cannot activate multiple visual perspectives simultaneously. Segal stated, “These findings suggest that a perspective-switching mechanism plays a crucial role in perspective-taking and imply that limitations in switching abilities may hinder the ability to adopt another’s perspective.”

The implications of this study will certainly influence models related to VPT, and future research will benefit from this important insight on switching costs. What’s more, future magicians may benefit from these findings, too.

Source: https://giphy.com/gifs/film-arrested-development-will-arnett-Sf0lxerEx2eNG

 Featured Psychonomic Society article

Segal, D. (2025). The cost of perspective switching: Constraints on simultaneous activation. Psychonomic Bulletin & Review. https://doi.org/10.3758/s13423-024-02633-x