Abstract
The Approximate Number System (ANS) is conceptualized as an innate cognitive system that allows humans to perceive numbers of objects or events (>4) in a fuzzy, imprecise manner. The representation of numbers is assumed to be abstract and not bound to a particular sense. In the present study, we test the assumption of a shared cross-sensory system. We investigated approximate number processing in the haptic modality and compared performance to that of the visual modality. We used a dot comparison task (DCT), in which participants compare two dot arrays and decide which one contains more dots. In the haptic DCT, 67 participants had to compare two simultaneously presented dot arrays with the palms of their hands; in the visual DCT, participants inspected and compared dot arrays on a screen. Tested ratios ranged from 2.0 (larger/smaller number) to 1.1. As expected, in both the haptic and the visual DCT responses similarly depended on the ratio of the numbers of dots in the two arrays. However, on an individual level, we found evidence against medium or stronger positive correlations between “ANS acuity” in the visual and haptic DCTs. A regression model furthermore revealed that besides number, spacing-related features of dot patterns (e.g., the pattern’s convex hull) contribute to the percept of numerosity in both modalities. Our results contradict the strong theory of the ANS solely processing number and being independent of a modality. According to our regression and response prediction model, our results rather point towards a modality-specific integration of number and number-related features.
Highlights
To be able to perceive and process numbers of events or objects, humans as well as other animals require a neuronal grounding
We evaluated a) whether adult participants express typical ratio dependent performance in both tasks, b) the relative weight of factors that lead to the percept of numerosity by means of a regression model of numerosity perception for each modality, c) the test–retest reliability of the haptic dot comparison task (DCT) and the visual DCT, and d) whether participants’ individual and group-wise performances correlate between the haptic and the visual DCT
Participants performed overall better in the haptic DCT compared with the visual DCT, Mhap = 79.5% (SD = 4.10) to Mvis = 77.7% (SD = 3.71)
Summary
To be able to perceive and process numbers of events or objects, humans as well as other animals require a neuronal grounding. The two-component model, distinguishes the number sense into one system that processes small nonsymbolic numbers of distinct individuals that fall into subitizing range (i.e., smaller than five items) and another system for larger numbers beyond four (Dehaene, 2011; Feigenson et al, 2004; Hyde, 2011; Mou & vanMarle, 2014; Olsson et al, 2016; De Smedt et al, 2013) The latter one is often denoted as the Approximate Number System (ANS), a cognitive system that enables people “to represent quantities as imprecise, noisy mental magnitudes without verbal counting or numerical symbols”
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
