Modulation of Visual Alpha Oscillations in Expert and Novice Surgeons Performing Sutures

BACKGROUND: Surgeons learn to perform highly repetitive movements, improving their speed and precision. Simple movements elicit a synchronization of alpha frequency band (8–12 Hz) in the occipital area, reflecting the inhibition of irrelevant areas. Yet, there is limited evidence on alpha modulation by movement performance and task experience and demands for complex visuo-motor skills. In this study we evaluated the extent of the modulation of the electroencephalogram (EEG) power in the alpha frequency band (8–12 Hz) in the visual areas and its relationship with suture performance to quantify the attentional modulation in expert surgeons and medical students. The EEG based measurements might offer a relevant measure of attentional modulation, to evaluate the progression and outcomes of learning and training surgical programs. Two groups of expert surgeons and medical students performed 6 surgical exercises on a suture pad, under two different task demands. They performed an open suture technique under relaxed conditions and stressed conditions. We obtained the EEG alpha power spectra, using a 20-20 system EEG device while suturing as well as in a baseline, eyes-open, condition as well as the number as sutures as an index of performance.RESULTS: Surgical expertise resulted in twice the number of sutures and greater task demands increased suture performance by 20%. In contrast, alpha power in the occipital areas is greater in surgeons and medical students performing sutures, relative to the baseline, yet it is not modulated by expertise or task demands. Interestingly, the alpha power correlated positively with suture performance in surgeons, but not in the medical students.CONCLUSIONS: The modulation of the EEG alpha power is consistent with the inhibitory-attentional hypothesis of alpha rhythm in a complex visuo-motor task, suggesting that the attentional resources allocated to the visual areas are redistributed in the somatosensory and motor areas, in addition to the visual areas during the suture task relative to the baseline. Furthermore, the association of alpha power with suture performance suggests that, unlike medical students, experts have a gradual redistribution of the inhibitory-attentional resources linked to their suture performance.

Perceptual Grouping Reveals Distinct Roles for Sustained Slow Wave Activity and Alpha Oscillations in Working Memory

Multiple neural signals have been found to track the number of items stored in working memory (WM). These signals include oscillatory activity in the alpha band and slow-wave components in human EEG, both of which vary with storage loads and predict individual differences in WM capacity. However, recent evidence suggests that these two signals play distinct roles in spatial attention and item-based storage in WM. Here, we examine the hypothesis that sustained negative voltage deflections over parieto-occipital electrodes reflect the number of individuated items in WM, whereas oscillatory activity in the alpha frequency band (8–12 Hz) within the same electrodes tracks the attended positions in the visual display. We measured EEG activity while participants stored the orientation of visual elements that were either grouped by collinearity or not. This grouping manipulation altered the number of individuated items perceived while holding constant the number of locations occupied by visual stimuli. The negative slow wave tracked the number of items stored and was reduced in amplitude in the grouped condition. By contrast, oscillatory activity in the alpha frequency band tracked the number of positions occupied by the memoranda and was unaffected by perceptual grouping. Perceptual grouping, then, reduced the number of individuated representations stored in WM as reflected by the negative slow wave, whereas the location of each element was actively maintained as indicated by alpha power. These findings contribute to the emerging idea that distinct classes of EEG signals work in concert to successfully maintain on-line representations in WM.

Enhanced Descending Corticomuscular Coupling During Hand Grip With Static Force Compared With Enhancing Force

The interaction between cortex and muscles under hand motor with different force states has not been quantitatively investigated yet, which to some extent places the optimized movement tasks design for brain-computer interface (BCI) applications in hand motor rehabilitation under uncertainty. Converging evidence has suggested that both the descending corticospinal pathway and ascending sensory feedback pathway are involved in the generation of corticomuscular coupling. The present study aimed to explore the corticomuscular coupling during hand motor task with enhancing force and steady-state force. Twenty healthy subjects performed precision grip with enhancing and static force using the right hand with visual feedback of exerted force. Mutual information and Granger causal connectivity were assessed between electroencephalography (EEG) over primary motor cortex and electromyography (EMG) recordings, and statistically analyzed. The results showed that the mutual information value was significantly larger for static force in the beta and alpha frequency band than enhancing force state. Furthermore, compared with enhancing force, the Granger causal connectivity of descending pathways from cortex to muscle was significantly larger for static force in the beta and high alpha frequency band (10-20 Hz), indicating the connection between the primary motor cortex and muscle was strengthened for static force. In summary, the hand grip with static force resulted in an increasing corticomuscular coupling from EEG over the primary motor cortex to EMG compared with enhancing force, implying more attention was required in the static force state. These results have important implications toward motor rehabilitation therapy design for the recovery of impaired hand motor functions.

Human resting-state electrophysiological networks in the alpha frequency band: Evidence from magnetoencephalographic source imaging

In the resting-state, extended regions of the human cortex engage in electrical oscillations within the alpha-frequency band (7–14 Hz) that can be measured outside the head by magnetoencephalography (MEG). Given the accumulating evidence that alpha oscillations play a fundamental role in attentional processing and working memory, it becomes increasingly important to characterize their cortical organization. Event-related studies have demonstrated that attentional allocation can modulate alpha power selectively within the visual, auditory, and somatosensory cortices, as well as in higher-level regions. Such studies demonstrate the existence of multiple generators by exploiting experimental contrasts and trial-averaging. The identification of alpha generators from resting-state data alone has proven much harder and, consequently, relatively little is known about their organization: Apart from the classical visual, somatosensory, and auditory rhythms, it is unclear how many more generators can be observed with MEG and how they are organized into functional networks. Such knowledge, however, possibly enables to delineate separate cognitive, perceptual, and motor processes that co-occur in the resting-state and is therefore important. In this study we use the resting-state MEG data-set provided by the Human Connectome Project to identify cortical alpha generators and to characterize their organization into functional networks. The large number of subjects (N = 94), multiple scans per subject, and state-of-the-art surface-based cortical registration enable a detailed characterization of alpha in human cortex. By applying non-negative matrix factorization to source-projected power fluctuations, we identify 16 reliable cortical generators in each hemisphere. These include the classical sensory alpha rhythms as well as several additional ones in the lateral occipital and temporal lobes and in inferior parietal cortex. We show that the generators are coordinated across hemispheres and hence form resting-state networks (RSNs), two of which are the default mode network (DMN) and the ventral attention network (VAN). Our study hence provides a further subdivision of RSNs within the alpha frequency band and shows that these RSNs are supported by alpha generators. As such, it links the classical literature on human alpha with more recent research into electrophysiological RNSs.