Small-large number dissociation : the effect of visual nesting
No Thumbnail Available
Authors
Meeting name
Sponsors
Date
Journal Title
Format
Thesis
Subject
Abstract
Previous studies show a limit in infants' number abilities: although infants can reliably discriminate two large numbers ([greater than or equal to] 4) or two small ([less than] 4) numbers, they often fail to discriminate one small number from one large number. This "cross-boundary failure" is assumed to be caused by a dissociation between the capacity-limited Object Tracking System (OTS) that tracks sets of up to 3 items with precision, and the Analogue Magnitude System (AMS) that estimates numerosities of any size with noisy magnitudes. The first goal of current study is to test if infants can cross the boundary when the OTS fails to process a small number of objects by using nested items rather than discrete items (i.e., dot array) as stimuli. The successful small-large number comparison would support that the cross-boundary effect is caused by the dissociation of OTS and AMS. Infant habituation study and change detection study together showed that 5-11 months old infants were not able to discriminate the numerosity of two sets of nested items with a ratio of 1:2 no matter if the boundary of four is crossed or not. This indicates that nested items are not suitable for the purpose of blocking access to the OTS while allowing access to the AMS. The second goal is to test whether adults' ability to make cross-boundary comparison is supported by their improved ability to choose the most task appropriate system (AMS) (H1) or their improved ability to make cross-system comparison (H2). Dot array comparison task showed that people tended to use the AMS to process both small number and large number in cross-boundary comparison and thus supported H1. Besides findings related to our major goals, current study also found that whether visual nesting led to overestimation or underestimation was dependent on the response format, which casted doubt on widely accepted account for how nested stimuli are processed (i.e., FINST theory). Moreover, adults' performance in the nested items vs. discrete items comparison was worse than the performance in the incongruent convex hull condition of a standard AMS comparison task. This suggests the standard control of continuous dimensions (including congruent, matched, incongruent conditions) may not be as efficient as expected.
Table of Contents
DOI
PubMed ID
Degree
Ph. D.