George Zouridakis

Language dominance determined by magnetic source imaging A comparison with the Wada procedure

Objective: To evaluate the validity of data derived from magnetic source imaging (MSI) regarding cerebral dominance for language in patients with intractable seizure disorder.Method:— The authors performed functional imaging of the receptive language cortex using a whole-head neuromagnetometer in 26 consecutive epilepsy patients who also underwent the intracarotid amobarbital (Wada) procedure. During MSI recordings, patients engaged in a word recognition task. This task was shown previously to activate language areas in normal adults as well as in patients who undergo intraoperative language mapping, allowing confirmation of MSI findings. Language laterality indices were formed for both the Wada and the MSI procedures. In addition, clinical judgments regarding cerebral dominance for language were made using the two methods by independent raters. Results: Cluster analysis indicated excellent agreement between the quantitative MSI and Wada indices. Rater judgments showed almost complete agreement as well. Conclusion: MSI is a promising method for determining cerebral dominance for language.

Multiple Bilaterally Asymmetric Cortical Sources Account for the Auditory N1m Component

The hypothesis that the N1, the major negative component of the cortical evoked response to auditory stimuli, originates from the primary auditory cortex has been supported by several studies. In a previous study we showed that, when monaural stimulation with pure tones is used, the distribution of the N1 peak over the scalp could be accounted for by successive activation of adjacent sources on the floor of the Sylvian fissure. In an attempt to establish the generality of the phenomenon, in this study we investigated further the generation of the N1 component using a variety of auditory stimuli, including pure tones, complex sounds (musical notes), and words, as well as binaural stimulus presentation. Additionally, we used a new recording system which allows recording of the distribution of the magnetic flux over the entire head simultaneously, thus eliminating the need for multiple recording sessions and the related problems of habituation and of changes in attention level. We found that a series of single dipolar sources could account for the entire duration of the N1m component. The location of the sources fell within the primary auditory cortex and, during the evolution of the component, they followed a posterior-anterior, medial-lateral, superior-inferior trajectory, bilaterally, along the superior surface of the temporal lobes. Additionally, the distribution of N1 sources on the two hemispheres showed a marked asymmetry, with the right hemisphere sources covering a larger area. The established consistency of successive source excitation across subjects, studies, types of stimuli, and recording systems, as well as the newly demonstrated hemispheric asymmetry of source extent, suggest the presence of a reliable phenomenon indicative of the functional organization of the auditory cortex.

Test-retest reliability of the P50 mid-latency auditory evoked response

Attenuation in mid-latency auditory evoked responses (MLAERs) can be used to study sensory gating. If paired-click stimuli (S1 and S2) are used, lower amplitude in response to S2 vs. S1 (attenuation) is considered evidence for intact sensory gating. However, the need for reliable measurements of MLAER amplitude and attenuation is a recognized problem. Ten normal volunteers were studied six times each. An S1 amplitude test-retest reliability coefficient (r) of 0.585 was obtained when means of two recordings were used vs. reliability coefficients as high as 0.809 for means of six recordings. Averaging a higher number of runs (120 vs. 60) resulted in a reliability coefficient of 0.677/recording. Similar values were obtained for S1 and S2 latencies. Reliability coefficients for S2 attenuation (S2/S1) were not nearly as high (a value of 0.138 when means of all six recordings were used). The S1 amplitude as measured in this study (with 120 averages) appears to be a reliable psychophysiologic measurement, but the S2/S1 attenuation measure is more variable, perhaps reflecting a greater sensitivity of the S2/S1 to uncontrolled variables in this study. Further research to identify such variables is necessary.

The P50 component of the auditory evoked potential and subtypes of schizophrenia

Decreased amplitude of the P50 component of the averaged evoked potential has been reported in schizophrenic patients. In an attempt to determine the relationship of this decrease to subtype diagnosis, we compared P50 amplitudes in 24 neuroleptic-free schizophrenic patients with paranoid (n = 13) versus nonparanoid (n = 11) subtype diagnoses. Eleven normal subjects and 11 cocaine users served as control groups. The schizophrenic patients were studied again after they had been treated with neuroleptics for 2 weeks. The control groups were studied again at least 2 weeks later. At baseline, the nonparanoid patients had significantly lower P50 amplitudes than did the normal subjects. The paranoid patients did not differ from the normal control subjects. The cocaine users had significantly decreased P50 amplitudes as compared with the normal control subjects. Neuroleptic treatment had no effect on P50 amplitudes in the paranoid patients but normalized amplitudes in the nonparanoid patients. The data suggest that P50 measurements may be useful in identifying subtypes of schizophrenia.

Lateralization of Cerebral Activation in Auditory Verbal and Non-Verbal Memory Tasks Using Magnetoencephalography

The magnetic flux normal to the scalp surface was measured with a whole-head neuromagnetometer while right-handed subjects (N = 15) were engaged in either an auditory word- or a tone-recognition task. Sources of the recorded magnetic fields were modeled as equivalent current dipoles at 4 ms intervals and the number of sources in the later portion of the magnetic response was used as an index of the degree of brain activation. Significantly more sources were found in the left as compared to the right hemisphere in the word but not the tone task on a group basis. On an individual basis, 13/15 subjects had more sources in the left as compared to the right hemisphere during the word task, while in the tone task 3/10 subjects showed this pattern. Sources of activity were found in the left superior and middle temporal gyri in all subjects with available MRI scans. Sources were also found in the supramarginal gyrus and in medial temporal areas, including the hippocampus, in the majority of cases. MEG appears to be a promising tool for detecting activity in cerebral areas specialized for language and memory function.

Atypical temporal lobe language representation: MEG and intraoperative stimulation mapping correlation

Functional brain imaging techniques hold many promises as the methods of choice for identifying areas involved in the execution of language functions. The success of any of these techniques in fulfilling this goal depends upon their ability to produce maps of activated areas that overlap with those obtained through standard invasive procedures such as electrocortical stimulation. This need is particularly acute in cases where active areas are found outside of traditionally defined language areas. In the present report we present two patients who underwent mapping of receptive language areas preoperatively through magnetoencephalography (MEG) and intraoperatively through electrocortical stimulation. Language areas identified by both methods were located in temporoparietal regions as well as in less traditional regions (anterior portion of the superior temporal gyrus and basal temporal cortex). Importantly there was a perfect overlap between the two sets of maps. This clearly demonstrates the validity of MEG-derived maps for identifying cortical areas critically involved in receptive language functions.

A fuzzy clustering approach to EP estimation

The problem of extracting a useful signal (a response) buried in relatively high amplitude noise has been investigated, under the conditions of low signal-to-noise ratio. In particular, the authors present a method for detecting the true response of the brain resulting from repeated auditory stimulation, based on selective averaging of single-trial evoked potentials. Selective averaging: is accomplished in two steps. First, an unsupervised fuzzy-clustering algorithm is employed to identify groups of trials with similar characteristics, using a performance index as an optimization criterion. Then, typical responses are obtained by ensemble averaging of all trials in the same group. Similarity among the resulting estimates is quantified through a synchronization measure, which accounts for the percentage of time that the estimates are in phase. The performance of the classifier is evaluated with synthetic signals of known characteristics, and its usefulness is demonstrated with real electrophysiological data obtained from normal volunteers.

Identification of reliable spike templates in multi-unit extracellular recordings using fuzzy clustering.

A method for extracting single-unit spike trains from extracellular recordings containing the activity of several simultaneously active cells is presented. The technique is particularly effective when spikes overlap temporally. It is capable of identifying the exact number of neurons contributing to a recording and of creating reliable spike templates. The procedure is based on fuzzy clustering and its performance is controlled by minimizing a cluster-validity index which optimizes the compactness and separation of the identified clusters. Application examples with synthetic spike trains generated from real spikes and segments of background noise show the advantage of the fuzzy method over conventional template-creation approaches in a wide range of signal-to-noise ratios.

Assessment of functional cerebral laterality for language using magnetoencephalography.

The purpose of the study was to examine the feasibility of using magnetoencephalography (MEG), a noninvasive functional brain imaging technique, to assess cerebral laterality for language. The magnetic flux normal to the scalp surface was measured with a whole-head neuromagnetometer while subjects (n = 16) were engaged in a word-matching and a tone-matching task. The effect of hemisphere and task on the number of satisfactory equivalent current dipole (ECD) solutions obtained during the late portion of the responses to the word and tone stimuli was examined. An interhemispheric ECD laterality index was also computed. Satisfactory ECD solutions were localized in perisylvian cortices during both tasks. A greater number of ECDs was found in the left hemisphere in 14 (87%) of 16 of the subjects in the word-matching task, a proportion that approaches the reported incidence of left-hemisphere dominance among right-handers. A similar proportion of subjects also showed a clear asymmetry in the number of ECDs favoring the left hemisphere in the language task as compared to the nonlanguage task. These findings suggest that MEG is a promising tool for laterality assessment. Magnetoencephalography-based functional asymmetry data are currently being compared against invasive presurgical procedures (i.e., intracarotid amobarbital procedure).

Lateralization of activity associated with language function using magnetoencephalography: a reliability study.

Summary This study was conducted to investigate the reliability of magnetoencephalography in lateralizing and localizing brain activity associated with receptive language function. Sixteen right-handed adults with no history of neurologic disorder engaged in a continuous recognition memory task for visually presented words in two separate sessions. The magnetic flux normal to the scalp surface was measured with a whole-head neuromagnetometer during task performance. Using the total number of acceptable activity sources as an index, overall activation was greater in the left compared with the right hemisphere for all 16 subjects in both sessions. Sources of activity were consistently found in the temporoparietal areas of the left hemisphere in all subjects. Moreover, clusters of activity sources in this region either overlapped spatially or were found in close proximity across sessions. Medial and basal temporal lobe activity was also observed in most subjects during at least one session, and tended to be lateralized to left hemisphere. These results suggest that magnetoencephalography is a promising tool for determination of cerebral dominance for language and localization of temporal lobe language areas.