A comparative study of ERP correlates of psychometric and Piagetian intelligence measures in normal and hyperactive children.

Verbal and performance scores of the Wechsler Intelligence Scale for Children-Revised (WISC-R 1981) and of a Piagetian battery, the Cognitive Development Scale for Children (EDC 1984), were obtained on 30 normal control and 19 hyperactive 6-8-year-old children. Amplitudes and latencies of a fronto-central P250 and of the parieto-occipital N250, P350 and P500 were measured concurrently in 4 categorization tasks derived from tests of the WISC-R and EDC batteries. Spearman correlations were computed between the intelligence and the ERP factor scores. Results showed that age-related and age-corrected Wechsler's scores were correlated with similar ERP changes (reduced amplitude, decreased latency). With regard to the amplitude changes, each type of intelligence was associated with a specific ERP pattern. The verbal scores were correlated with the P350 and the P500 amplitudes, and the performance scores with the frontal P250 and occipital N250 amplitudes. By contrast, Piagetian development and intelligence scores yielded ERP correlates in the opposite direction: P500 amplitude was negatively correlated with raw EDC scores, but positively with scaled EDC scores. In addition, Piagetian intelligence was not related to the general peak latency decrease with age. In hyperactive children, additional negative correlations were found between P250 amplitude and the subjects' verbal test scores. Correlations with some performance tests that were negative in normal controls, were positive in hyperactive children. In addition, latency-based correlations found in normal controls were lacking in hyperactive children. These findings provide strong evidence that intelligence comprises different components related to different subsets of cognitive processes, as indexed by different ERP waves. They also suggest that the development and intelligence do not always rely on the same changes, and that intelligence forms may not be referred to the same use of the same processes in hyperactive and normal children.