Attention, Not Performance, Correlates With Afterdischarge Termination During Cortical Stimulation

Abstract

Cortical stimulation has been used for brain mapping for over a century, and a standard assumption is that stimulation interferes with task execution due to local effects at the stimulation site. Stimulation can however produce afterdischarges which interfere with functional localization and can lead to unwanted seizures. We previously showed that (a) cognitive effort can terminate these afterdischarges, (b) when termination thus occurs, there are electrocorticography changes throughout the cortex, not just at sites with afterdischarges or sites thought functionally important for the cognitive task used, and (c) thresholds for afterdischarges and functional responses can change among stimulation trials. We here show that afterdischarge termination can occur prior to overt performance of the cognitive tasks used to terminate them. These findings, taken together, demonstrate that task-related brain changes are not limited to one or a group of functional regions or a specific network, and not limited to the time directly surrounding overt task execution. Discrete locations, networks and times importantly underpin clinical behaviors. However, brain activity that is diffuse in location and extended in time also affect task execution and can affect brain mapping. This may in part reflect fluctuating levels of attention, engagement, or motivation during testing.

Pan-cortical electrophysiologic changes underlying attention

“…We calculated WxCoh for all electrode contact pairs over a range of wavelet scales with peak frequencies from 7.13 to 40.05 Hz. These in our previous papers had been significantly associated with AD termination [5][6][7] , and spectral analysis had shown these to contain the majority of the WxCoh changes 7 . For these calculations, a scale can be thought of as a surrogate for frequency.…”

“…Our results are consistent with these conclusions but expand them: we found coherence changes throughout the entire cortex. In addition, we have shown that afterdischarge termination correlated with ongoing attention throughout the AST trial period, and not to the portion of the trial during which the patient answered the posed question 6 (Fig. 8).…”

“…Coherence decreases in the frequency ranges reported in our patients have previously been found in the setting of attention increases or increased cognitive load [69][70][71][72][73][74][75][76] . It is possible that AD termination was related to the patient answering the posed question, however, as described above, AD termination in response to AST was better correlated with attention to and consideration of the task request rather than the actual performance of the task 6 . More broadly, however, the quality of executive function, the capacity of working memory, and the level of patient engagement with the task and motivation for answering the posed question could be factors influencing whether ADs ended in response to the task setting 77 .…”

“…Figure 8. Question and answer latencies6 . The figure shows the latency between (blue bars) question start and afterdischarge termination vs. (red bars) answer start and afterdischarge termination.…”