Language interference was elicited by electrical stimulation of the dominant basal temporal region in 8 out of 22 cases and in none of 7 cases with subdural electrodes implanted over the nondominant temporal lobe. Language interference was elicited by stimulation of electrodes placed over the fusiform gyrus 3-7 cm from the tip of the temporal lobe. Electrical stimulation of the basal temporal language area produced a global receptive and expressive aphasia with speech arrest at high stimulus intensities. Other higher cortical function, for example copying complex designs or memory of nonverbal information was intact, in spite of the total inability to process verbal information. At lower stimulus intensities partial aphasias with a predominant receptive component occurred. Surgical resection of the basal temporal language area produces no lasting language deficit.

We report on a 38-year-old patient with intractable complex partial seizures originating in the dominant left medial temporal region. In the work-up for seizure surgery, arrays of subdural electrodes were placed, and electrical stimulation revealed marked language interference in a 2 X 2-cm area in the left basal temporal fusiform gyrus (3.5 to 5.5 cm posterior to the temporal tip). Complete receptive and expressive aphasia, inability to repeat, agraphia, and alexia were elicited, but visual memory was preserved, and no constructional apraxia was noted. Stimulation of the basal temporal gyrus at lower stimulus intensities produced a relatively selective and severe anomia.

In three patients stimulation of the frontal speech area resulted in one or more of the following symptoms: speech arrest, writing arrest, or impaired rapid alternating movements of the tongue, fingers or toes. Speech arrest could be altered at individual points either with or without impairment of rapid movements or writing, but writing was not impaired without concomitant difficulties with either speech or rapid finger movements. Our data suggest that the frontal speech area may function to integrate complex motor functions, some speech related and others not. We also confirm previous conclusions that Exner's writing centre is not separate from Broca's area and that the writing defect in Broca's aphasia can occur without involvement of the motor strip.

No abstract