Cortical gamma-oscillations modulated by auditory-motor tasks-intracranial recording in patients with epilepsy

Abstract

Human activities often involve hand-motor responses following external auditory-verbal commands. It has been believed that hand movements are predominantly driven by the contralateral primary sensorimotor cortex, whereas auditory-verbal information is processed in both superior temporal gyri. It remains unknown whether cortical activation in the superior temporal gyrus during an auditory-motor task is affected by laterality of hand-motor responses. Here, event-related γ-oscillations were intracranially recorded as quantitative measures of cortical activation; we determined how cortical structures were activated by auditory-cued movement using each hand in 15 patients with focal epilepsy. Auditory-verbal stimuli elicited augmentation of γ-oscillations in a posterior portion of the superior temporal gyrus, whereas hand-motor responses elicited γ-augmentation in the pre- and postcentral gyri. The magnitudes of such γ-augmentation in the superior temporal, precentral, and postcentral gyri were significantly larger when the hand contralateral to the recorded hemisphere was required to be used for motor responses, compared with when the ipsilateral hand was. The superior temporal gyrus in each hemisphere might play a greater pivotal role when the contralateral hand needs to be used for motor responses, compared with when the ipsilateral hand does.

Figure 1

Time‐frequency analysis of γ‐oscillations relative to “the onset of button‐press” in patient 13. (A) Subdural electrodes were placed on the right hemisphere. (B and C) In the auditory–motor task when the subject was required to provide responses using the left hand, significant γ‐augmentation (red) involving 90–130 Hz was noted at electrode A39 located in the right superior temporal gyrus at 360–210 ms prior to the onset of button‐press. Subsequently, significant γ‐augmentation involving 75–165 Hz was noted at electrode A62 located over the right central sulcus between −110 ms and +100 ms relative to the onset of button‐press. Furthermore, significant γ‐augmentation involving 80–135 Hz was noted in the same site between 660 and 990 ms following the onset of button press; this late γ‐augmentation may be associated with button‐release movement. Neurostimulation of this electrode site elicited movement of the left‐sided fingers. (D and E) In the task with the right hand used for motor responses, no significant γ‐augmentation was noted at electrode A39 or A62. (F and G) In the task for the left hand, “γ‐range amplitude” was increased by 77% at electrode A39 in the right superior temporal gyrus at −300 ms relative to the onset of button‐press and increased by at least 70% at multiple sites overlying the central sulcus and postcentral gyrus at the onset of button press. (H and I) In the task for the right hand, “γ‐range amplitude” was increased by 43% at electrode A39 at −300 ms relative to the onset of button press but the magnitude of γ‐augmentation did not reach significance defined in this study. No significant γ‐augmentation was noted in the precentral or postcentral gyrus. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Figure 2

Time‐frequency analysis of γ‐oscillations relative to “the onset of button‐press” in Patient 10 with a diagnosis of MR nonlesional focal epilepsy. “Gamma‐range amplitudes” at the onset of button‐press using the left hand (A and B) and the right hand (C and D) are shown. “Gamma‐range amplitudes” were increased in each Rolandic area, to a greater extent, when the hand contralateral to the recorded hemisphere was used for motor‐responses. It also seems as if the left‐hand movement elicited rather bilateral γ‐augmentation, whereas the right‐hand movement elicited unilateral γ‐augmentation confined to the left Rolandic area. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Figure 3

The magnitude of γ‐augmentation elicited in the superior temporal gyrus. The maximum “γ‐range amplitude” (unit: %) in “the center of auditory event‐related γ‐augmentation” is shown by a bar graph. ECoG analysis relative to the onset of motor responses showed that the maximum “γ‐range amplitude” was 82% on average when the hand contralateral to the recorded hemisphere was used for motor responses and 63% on average when the ipsilateral hand was used. AVE, average across the 15 hemispheres (error bars shown).

Figure 4

The magnitude of γ‐augmentation elicited in the Rolandic area. The maximum “γ‐range amplitude” (unit, %) in “the center of movement‐related γ‐augmentation” is shown by a bar graph; thereby, “the center of movement‐related γ‐augmentation” was defined as the site showing the largest “γ‐range amplitude” elicited by the contralateral hand movement. The maximum “γ‐range amplitude” was 178% on average when the hand contralateral to the recorded hemisphere was used for motor responses and 42% on average when the ipsilateral hand was used. The maximum “γ‐range amplitude” elicited by the ipsilateral hand movement was noted in a site different from “the center of movement‐related γ‐augmentation” in two patients (Patients 5 and 11). “The center of movement‐related γ‐augmentation” was localized in the postcentral gyrus in Patient 5 and in the site over the central sulcus in Patient 11; the maximum “γ‐range amplitude” elicited by the ipsilateral hand movement was noted in the precentral gyrus in both patients. Difference in the maximum “γ‐range amplitude” between the ipsi‐ and contra‐lateral hand responses was larger in the Rolandic area, compared to that in the superior temporal gyrus (see also Fig. 3). AVE, average across the 16 hemispheres (error bars shown); Lt, left; Rt, right.