Prediction of childhood intelligence from habituation in early infancy

Abstract

The present study investigated whether measures of habituation and dishabituation in early infancy predicted later ( age 4 1 2 years) intelligence. An infant-controlled habituation procedure was used and each infant was tested on three separate occasions. Statistically reliable correlations between the two ages were obtained, and these were specific to: (a) verbal components of the child intelligence test scores; (b) measures derived from the familiarization of habituation phase of the infant testing, rather than from subsequent dishabituation or novelty response scores. The predictive infant measures were only those which minimally satisfied both of two psychometric criteria: (1) There are consistent changes with age; (2) there is test-retest reliability, with age partialled out. The results are compared with those from other comparable studies, and it is argued that the psychometric acceptability of infant cognitive measures needs to be demonstrated before they can be considered to be potential predictors.