The role of the human thalamus in language and memory: evidence from electrophysiological studies.

The data reviewed here indicate that electrical stimulation of the dominant ventrolateral thalamus can produce deficits in language processing that are not seen after similar stimulation of the nondominant ventrolateral thalamus. The nature of the language deficit produced varies, depending upon the rostrocaudal location of the stimulation site. Stimulation of the anterior left ventrolateral thalamus in right-handed patients resulted in production of a repeated erroneous word, stimulation of the medial ventrolateral thalamus evoked perseveration, and stimulation of the posterior ventrolateral thalamus and anterior pulvinar resulted in misnaming and omissions. Additional studies have examined the effect of electrical thalamic stimulation on verbal and nonverbal short-term memory. Left (but not right) ventrolateral thalamic stimulation during verbal memory input greatly decreased subsequent recall errors, while stimulation during verbal memory retrieval increased recall errors. This finding contrasted with those obtained from studies on nonverbal memory, in which right ventrolateral stimulation during memory input decreased recall errors, while left thalamic stimulation at the same stage increased recall errors. Left pulvinar stimulation disrupted verbal memory processing, while right pulvinar stimulation disrupted nonverbal memory processing. Limited evidence suggests that the effects of thalamic electrical stimulation on verbal memory may persist for several days after the stimulation has ended. The lateralization of thalamic functions also affects the motoric aspects of speech production. Left (but not right) ventrolateral thalamic stimulation disrupted speech articulation and increased the expiratory phase of respiration. The fact that these motor effects were evoked from the same general area of the thalamus that produced the language deficits discussed above raises the possibility that the thalamus is involved in coordinating the cognitive and motoric aspects of language production. A model of thalamic function is discussed in which defined regions of the thalamus operate as a "specific alerting response," increasing the input to memory of category-specific material while simultaneously inhibiting retrieval from memory.

High-resolution inter-ictal SPET and phased-array MRI in partial epilepsy: an imaging comparison with video/EEG and outcome correlation.

To assess the clinical utility of high-resolution inter-ictal single photon emission tomography (SPET) of regional cerebral perfusion and high-resolution magnetic resonance imaging (MRI) of the brain with a phased-array temporal lobe coil, 35 patients with presumed partial epilepsy were evaluated prospectively by these techniques in addition to prolonged video/electroencephalographic (EEG) monitoring. Twenty of these patients had surgical treatment of partial epilepsy with outcome determinations spanning from 12 months to 3 years at follow-up. There were four categories of imaging findings as compared to scalp/sphenoidal EEG localization. Category I included 12 patients (34% of total) in whom there was complete imaging and EEG concordance. Category II included 4 patients (11%) in whom MRI and EEG were concordant but SPET was divergent or normal. Category III included 13 patients (37%) in whom SPET and EEG were concordant but MRI was divergent or normal. Category IV included 4 patients (11%) in whom neither SPET nor MRI was concordant with EEG. In this study, the relative sensitivities of SPET and MRI for localization of partial epilepsy based on prolonged scalp/sphenoidal video/EEG recordings were 76% and 49%, respectively. We conclude that these neuroimaging techniques (phased-array MRI and inter-ictal cerebral perfusion SPET) are complementary and useful in the pre-operative evaluation of patients with partial epilepsy.

Optical imaging of epileptiform and functional activity in human cerebral cortex.

Optical imaging of animal somatosensory, olfactory and visual cortices has revealed maps of functional activity. In non-human primates, high-resolution maps of the visual cortex have been obtained using only an intrinsic reflection signal. Although the time course of the signal is slower than membrane potential changes, the maximum optical changes correspond to the maximal neuronal activity. The intrinsic optical signal may represent the flow of ionic currents, oxygen delivery, changes in blood volume, potassium accumulation or glial swelling. Here we use similar techniques to obtain maps from human cortex during stimulation-evoked epileptiform afterdischarges and cognitively evoked functional activity. Optical changes increased in magnitude as the intensity and duration of the afterdischarges increased. In areas surrounding the afterdischarge activity, optical changes were in the opposite direction and possibly represent an inhibitory surround. Large optical changes were found in the sensory cortex during tongue movement and in Broca's and Wernicke's language areas during naming exercises. The adaptation of high-resolution optical imaging for use on human cortex provides a new technique for investigation of the organization of the sensory and motor cortices, language, and other cognitive processes.

Enhancement of memory with human ventrolateral thalamic stimulation: effect evident on a dichotic listening task.

Previous experience with left ventrolateral thalamic (VL) stimulation during visually presented language and verbal memory tasks has shown that stimulation at the time information enters memory increases the accuracy of subsequent recall. The present study investigated the effects of VL stimulation on an auditory dichotic listening task. A similar effect was identified with significantly more words presented during left VL stimulation subsequently correctly reported, compared to words presented in the absence of stimulation, or with right VL stimulation. No significant effects on the ratios of correct responses from opposite ears were observed.

Common cortical and thalamic mechanisms for language and motor functions.

Evidence for common mechanisms in the human brain for motor and language functions is reviewed, particularly evidence derived from electrical-stimulation mapping during cortical and thalamic operations in awake patients. Several systems in the dominant hemisphere are identified where language and motor function share common mechanisms, including a lateral thalamic attentional system and a lateral perisylvian cortical system common to sequential movement and speech sound identification, where precise timing may be a common mechanism.

Intracellular records from chronic alumina epileptogenic foci in the monkey.

Neuronal records from the primate alumina focus revealed synaptic potentials with bursts of action potentials occurring on depolarizing potentials. These depolarizing potentials were not synchronous with electrocorticographic epileptic patterns. The configuration of the membrane potential changes was consistent with a dendritic generator for epileptic bursts. It is proposed that the genesis of these depolarizing potentials is postsynaptic.

Aphasia and left thalamic hemorrhage.

Left thalamic hemorrhage as a cause of aphasia has not been widely recognized. Large thalamic hemorrhages cause coma, making speech examination impossible; smaller thalamic hemorrhages were difficult to document until recent diagnostic advances. Nine cases of thalamic hemorrhage with aphasia have been described in the literature. This report presents four additional cases. These patients had acute onset of aphasia, supranuclear paralysis of upward gaze, right hemisensory deficits, and mild right hemiparesis. Three of the four patients responded to ventriculostomy drainage with rapid clearing of the supranuclear paralysis of upward gaze, and two later required placement of permanent ventricular shunts. After 1 year, two patients exhibited no clinically detectable speech malfunction and the other two were severely aphasic. The hemiparesis, hemisensory deficits, and ocular pareses all cleared. These cases are discussed with respect to present models of the role of the thalamus in speech.