Abstract

The study of numerical estimation collectively spans hundreds of papers and hundreds of thousands of citations. Interest in this topic hinges on one assumption: that we can approximate number independently of continuous spatial dimensions (e.g., area). Accordingly, many studies have specifically tried to demonstrate sensitivity specific to number while controlling other dimensions. However, recent work demonstrates that perceived area (based on psychophysical judgments) differs from true area (i.e., a precise pixel count). This difference raises concerns about most past studies of approximate number, by asking if they have systematically controlled for the wrong dimension(s). Building on recent findings that the percept of area may be systematically illusory, the current study examines the relation between perceived area and number. Four experiments reveal that (1) perceived area, but not mathematical area, strongly influences numerosity judgments, (2) perceived area influences perceived number but not the reverse, (3) number acuity is greatly reduced in stimuli controlled for perceived area, and (4) the ability to make area discriminations on the basis of ‘additive area’ but not mathematical area predicts number discrimination ability. Together, these findings highlight a potentially serious confound in prior work, raising new theoretical and methodological challenges for the field.

Highlights

  • The study of numerical estimation collectively spans hundreds of papers and hundreds of thousands of citations

  • This propensity purportedly relies on an evolutionary ancient system—the Approximate Number System—that serves as perceptual foundation for downstream numerical and mathematical ­ability[1,2,3]

  • We may rely on an ‘Additive Area Heuristic’ (AAH), approximating area as the sum of objects’ dimensions rather than their product

Read more

Summary

Introduction

The study of numerical estimation collectively spans hundreds of papers and hundreds of thousands of citations. Recent work demonstrates that perceived area (based on psychophysical judgments) differs from true area (i.e., a precise pixel count) This difference raises concerns about most past studies of approximate number, by asking if they have systematically controlled for the wrong dimension(s). If you were to sum the diameters of the dots in the left panel, that value would be greater than the sum of the diameters in the right display This ‘additive-area’ estimate appears to track area estimations quite accurately, even after accounting for variables like number and contour l­ength[12].

Methods
Results
Discussion
Conclusion
Full Text
Paper version not known

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