Lana Milman¹, Ian Phillips¹, Chaz Firestone¹; ¹Johns Hopkins University

In addition to physical actions (e.g., climbing a staircase, or grasping an object), we also perform mental actions (e.g., counting objects in our head, or shifting attention). Recent work in philosophy of mind proposes that, just as we can appreciate whether and how easily we can execute various physical actions (physical affordances), we can also do the same for mental actions — appreciating in advance how effectively we will be able to execute a certain cognitive operation before actually carrying it out (the “mental affordance hypothesis”; McClelland, 2020). Here, we explore this hypothesis for counting and its corresponding mental affordance “countability” — i.e., how quickly and accurately an array of objects can be precisely counted. Subjects briefly (500ms or 2500ms) saw two “cookies” (circles) containing within them a number of “M&Ms” (dots). The M&Ms varied in size (large, small, or mixed), color (single or mixed), and opacity (full or partial). Subjects selected whichever cookie seemed easier to count (causing it to reappear onscreen), and then went ahead and counted that cookie’s M&Ms. Results showed that even a 500ms preview was sufficient for subjects to accurately predict many aspects of their own counting performance on a given display, including that larger M&Ms would be easier to count than smaller M&Ms, that opaque M&Ms would be easier to count than semitransparent M&Ms, and so on. However, they also “misperceived” countability: Subjects preferred to count cookies with mixed-size M&Ms over smaller-sized M&Ms, even though counting performance was better on the latter. Our results suggest that naive observers can rapidly form impressions of a mental affordance and use it to guide behavior. Moreover, like physical affordances, we may be imperfectly calibrated to our actual capabilities.

Acknowledgements: Hopkins Office of Undergraduate Research Bloomberg Distinguished Professor Summer Program Award