Discrimination of Large and Small Numerosities by Human Infants


Department of Psychology, Harvard University, 1120 William James Hall, 33 Kirkland Street, Cambridge, MA 02138. E-mail: lipton@wjh.harvard.edu


Six experiments investigated infants' sensitivity to numerosity in auditory sequences. In prior studies (Lipton & Spelke, 2003), 6-month-old infants discriminated sequences of 8 versus 16 but not 8 versus 12 sounds, and 9-month-old infants discriminated 8 versus 12 but not 8 versus 10 sounds, when the continuous variables of rate, sound duration, and sequence duration were controlled. The current studies investigated whether infants' numerical discrimination is subject to the signature ratio limit of adults' numerosity discrimination. Four experiments at 6 and 9 months provided evidence for this signature limit, suggesting that common mechanisms underlie numerosity discrimination in infants and adults. In further experiments, infants failed to discriminate 2 versus 4 or 2 versus 3 sounds when tested under the same conditions as with large numbers. These findings accord with studies using visual-spatial arrays (e.g., Clearfield & Mix, 1999) and suggest that separate systems underlie infants' representation of small and large numerosities.