Although interictal spikes are thought to share pathophysiological mechanisms with partial-onset seizure discharges, positron emission tomographic studies of the interictal state have paradoxically shown focal hypometabolism whereas seizures produce hypermetabolism. To address this question, we performed functional mapping studies in an interictal spiking model in the rat. Recording screw electrodes were inserted through the skull bone so as to depress underlying cortex. Interictal spiking was subsequently induced by systemic administration of bicuculline methiodide. 2-deoxy[14C]glucose studies revealed increased glucose utilization in superficial and middle cortical layers at spiking screw sites. Nonspiking screw sites in the same animals and in controls did not show increased uptake. Convulsive seizures caused additional 2-deoxy[14C]glucose uptake at screw sites and in widespread forebrain areas. c-fos immunoreactivity occurred in superficial cortex at interictal spiking, but not nonspiking, sites. Convulsive seizures induced widespread forebrain c-fos immunoreactivity. These data suggest interictal epileptiform activity occurs in cells adjacent to cortical injury; these activate deeper layers via local connections. Interictal and ictal epileptiform states share common mechanisms, as both induce glucose hypermetabolism and immediate-early gene product activation. Possible reasons for failure to detect hypermetabolism in interictal human subjects are discussed.