Cognitive Event-Related Brain Potentials in Children with Attention and Reading Deficits

ERPs were recorded from four groups of children: reading disabled, attentional deficit disorder with and without hyperactivity, and normal controls. Subjects pressed a button to a low probability nonsense syllable (target, p= .168) and ignored all other events, which included a high probability nonsense syllable (nontarget, p= .664) and either low probability (category, p= .168) symbols (Block 1) or 3-letter words (Block 2). The amplitudes of several late ERP components and the latency of the P3 component were examined. The overall amplitude of P3 was significantly smaller in all clinical groups than in controls, but the difference in P3 amplitude between targets and nontargets was smaller only in the two attentional deficit groups. Reading disabled children had smaller P3 and Pc components to words than to symbols, while controls had equivalent values. The N2 component had a different scalp distribution for words and symbols, but did not differentiate reading disabled children from controls. P3 latency was significantly longer in the three clinical groups than in controls, but only the attentional deficit groups showed an increase in P3 latency across blocks of the task. The results are discussed within the framework of recent cognitive models dealing with attentional processes.