The relative salience of numerical and non-numerical dimensions shifts over development: A re-analysis of Tomlinson, DeWind, and Brannon (2020)

Abstract

Visual displays of objects include information about number and other magnitudes such as cumulative surface area. Despite the confluence of cues, a prevalent view is that number is uniquely salient within multidimensional stimuli. Consistent with this view, Tomlinson, DeWind, and Brannon (2020) report that, in addition to greater acuity for number than area among both children and adults, number biases area judgments more than the reverse, at least in childhood. However, a failure to consider perceived area, undermines these results. To address this concern, we used an index of perceived area when assessing acuity and bias of number and area. In this context, number and area were comparable in acuity among children and adults. Bias, however, differed across development. Although adults showed greater bias of number on area judgments than the reverse, children experienced greater area bias on number judgments. Thus, contra Tomlinson et al., when differences in mathematical and perceived area are accounted for, area is more salient than number early in development. However, number does become the more salient dimension by adulthood, suggesting a role for experience with symbolic number and education in drawing attention to number within multidimensional visual stimuli.

Numerosity, area-osity, object-osity? Oh my

“…Recent evidence from our lab, however, goes against the uniqueness claim. For example, we found that when perceptual discriminability between number and cumulative area was matched, area biased children's number judgments more than the reverse (Aulet & Lourenco, 2021b) and children sorted visual stimuli according to area, not number (Aulet & Lourenco, 2021c), suggesting greater intrinsic salience for nonnumerical magnitude, and consistent with others who have argued against the uniqueness of number Newcombe, Levine, & Mixs, 2015). Similarly, Testolin, Dolfi, Rochus, and Zorzi (2020) found that the internal encoding of "mature" computational networks, trained to discriminate stimuli according to number, treated total perimeter and convex hull as comparable to number.…”

A theory of perceptual number encoding.

“…However, even in the face of statistical learning, non-numerical magnitudes remain more salient, and thus, must be inhibited in order to successfully attend to, and discriminate, number. Although there is accumulating evidence that area may be more salient than number early in development (Aulet & Lourenco, 2021a, 2021cHenik, Gliksman, Kallai, & Leibovich, 2017)…”

Do children estimate area using an ‘Additive-Area Heuristic’?

“…Yet a reanalysis of those data accounting for additive area suggested this may not be the case. Indeed, when accounting for additive area, the opposite may be true: area biases number judgments (Aulet & Lourenco, 2021).…”