Level‐2 perspectives computed quickly and spontaneously: Evidence from eight‐ to 9.5‐year‐old children

Abstract

It has been widely assumed that computing how a scene looks from another perspective (level-2 perspective taking, PT) is an effortful process, as opposed to the automatic capacity of tracking visual access to objects (level-1 PT). Recently, adults have been found to compute both forms of visual perspectives in a quick but context-sensitive way, indicating that the two functions share more features than previously assumed. However, the developmental literature still shows the dissociation between automatic level-1 and effortful level-2 PT. In the current paper, we report an experiment showing that in a minimally social situation, participating in a number verification task with an adult confederate, eight- to 9.5-year-old children demonstrate similar online level-2 PT capacities as adults. Future studies need to address whether online PT shows selectivity in children as well and develop paradigms that are adequate to test preschoolers' online level-2 PT abilities. Statement of Contribution What is already known on this subject? Adults can access how objects appear to others (level-2 perspective) spontaneously and online Online level-1, but not level-2 perspective taking (PT) has been documented in school-aged children What the present study adds? Eight- to 9.5-year-olds performed a number verification task with a confederate who had the same task Children showed similar perspective interference as adults, indicating spontaneous level-2 PT Not only agent-object relations but also object appearances are computed online by eight- to 9.5-year-olds.

Time pressure disrupts level-2, but not level-1, visual perspective calculation: A process-dissociation analysis

“…Several noteworthy findings emerged in Experiment 1: The error-rate analyses provided unambiguous evidence for the presence of altercentric interference in both the L1-VPT task, replicating many previous studies, and the L2-VPT task, replicating several prior studies (Elekes et al, 2016(Elekes et al, , 2017. The error-rate analyses also provided suggestive, though ultimately inconclusive, evidence that time pressure had opposing effects on altercentric interference in the two tasks.…”

“…Altercentric-interference effects have also been reported for this L2-VPT task. Here, participants have more difficulty reporting the numeral's identity from their own perspective when it has a different identity from the avatar's perspective (Elekes et al, 2016(Elekes et al, , 2017; but see Surtees, Butterfill, & Apperly, 2012). Because participants' focal task goal is simply to report the numeral's identity from their own perspective, 1 any incidental processing of the avatar's perspective should interfere with this objective.…”

“…Because participants' focal task goal is simply to report the numeral's identity from their own perspective, 1 any incidental processing of the avatar's perspective should interfere with this objective. Thus, altercentric-interference effects in indirect measures of L2-VPT have been interpreted as reflecting the spontaneous calculation of the way something looks to the avatar (e.g., Elekes et al, 2016Elekes et al, , 2017.…”

“…Nevertheless, in Experiments 3A and 3B, we modified the L2-VPT task so that the stimulus set comprised four numerals instead of two. Specifically, along with asymmetrical numerals (6, 9) that have different identities when viewed from one's own versus the avatar's physical vantage point, we also included symmetrical numerals (1, 8) that have the same identity when viewed from either perspective (Elekes et al, 2016(Elekes et al, , 2017Surtees, Samson, et al, 2016, Experiment 2). Experiment 3A, like Experiment 1, used a within-subjects manipulation of response deadline, whereas Experiment 3B, like Experiment 2, used a betweensubjects manipulation of response deadline.…”

“…Although it has traditionally been assumed that successfully recognizing another agent's divergent view of the world requires active and effortful deliberation, demonstrations of so-called "implicit mentalizing" in adults, children, and even infants suggest otherwise (for reviews, see Baillargeon, Scott, & He, 2010;Schneider, Slaughter, & Dux, 2015;Schneider, Slaughter, & Dux, 2017;Scott & Baillargeon, 2017; but see Kulke, von Duhn, Schneider, & Rakoczy, 2018). For example, multiple studies have reported evidence that people display behavior on indirect measures indicative of having incidentally calculated both whether an object is or is not visible to another person-level-1 perspective taking (Flavell, Everett, Croft, & Flavell, 1981)-and how that object appears (i.e., the object's identity) to the other person-level-2 perspective taking (Flavell et al, 1981)-in the absence of deliberately attempting to do so (e.g., Elekes, Varga, & Király, 2016;Elekes, Varga, & Király, 2017;Qureshi, Apperly, & Samson, 2010;Samson, Apperly, Braithwaite, Andrews, & Bodley Scott, 2010;.…”