The effect of lamotrigine (LTG) as add-on therapy on electroencephalogram (EEG) background activity was studied in 11 patients with refractory partial seizures with or without secondary generalization. The computerized EEG study was performed at rest with eyes closed (EC), during blocking reaction (BR), fixation (FIX), and mental arithmetic (MA) tasks. EEG spectral values were analyzed statistically using three-way ANOVA. The neuropsychological evaluation included a battery of six tests. Epileptic patients before LTG therapy, compared with control subjects, displayed at rest condition EEG changes consisting of higher delta and theta relative power coupled with lower alpha and beta power. During performance of attentive (BR) and cognitive (FIX) tasks, a decrease in alpha reactivity associated with a decrease of beta 1 and beta 2 power was found. The addition of LTG to previous therapy induced changes, although subtle, consisting of an increase in both alpha reactivity and beta power to attentive task. Neuropsychological evaluation did not evidence any impairment of cognitive functions. During LTG therapy, a decrease in seizure frequency occurred in 9 of the 11 patients whereas no changes were observed in the remaining 2. On the basis of these neurophysiologic and neuropsychological findings, LTG as add-on therapy does not seem to produce adverse side effects on mental activity; moreover, EEG data indicate a slight improvement in attentional processes.