Skip to main content
NCBI home page
Clinical Neurophysiology Practice logoLink to Clinical Neurophysiology Practice
. 2026 Aug 5;11:679–688. doi: 10.1016/j.cnp.2026.07.017

Impact of stress on EEG functional connectivity during procedural learning

Francis L Kirk a, Alexander F Friend a, Sudhanva Suvarna a, Emmett E Whitaker b, Gregory L Holmes c,⁎
PMCID: PMC13495613  PMID: 42631122

Abstract

Objective

In a procedural learning environment, trainees acquire skills for managing life-threatening emergencies, a context that inherently elicits stress. While moderate stress can enhance memory consolidation, excessive stress (distress) may impair learning and performance. Here we elucidated the neurophysiological effects of stress on learners during a procedural learning task.

Methods

We examined the effects of stress on learning by assessing EEG functional connectivity (FC) in medical trainees during endotracheal intubation. The cohort comprised medical trainees who were monitored at baseline (rest) and during two high-fidelity simulation scenarios: a calm condition and a stressful condition (intubation during anesthesia induction). Concurrent data were collected, including heart rate, eye blinks and self-reported measures of perceived stress and cognitive load.

Results

High-fidelity simulation reliably provokes a physiological stress response, predominantly mediated by sympathetic nervous system activation. Across conditions, FC increased relative to baseline, indicating heightened neural network synchronization during task engagement. Notably, under the stressful condition, FC was further elevated compared with the calm scenario, suggesting that stress within a defined range may enhance neural communication efficiency and learning capacity.

Conclusions

These findings imply a potential neurophysiological signature of stress that could be leveraged to design personalized educational interventions, calibrating stress levels to optimize learning while mitigating deleterious effects.

Significance

The study demonstrates the feasibility of using real-time EEG to monitor stress-related neural dynamics in medical trainees and underscores the importance of integrating neurophysiological insights into the development of training environments.

Keywords: Coherence, Phase lag value, EEG

Highlights

  • •

    High-fidelity simulation triggers sympathetic stress responses in trainees.

  • •

    EEG connectivity rises during tasks; highest under stress vs calm scenarios.

  • •

    Real-time EEG may help tailor stress levels to optimize learning outcomes.

1. Introduction

Moderate stress can enhance memory formation and learning, but excessive stress or distress may impair both learning and performance (LeBlanc, 2009; Schwabe and Wolf, 2010; Schwabe et al., 2010; Vogel et al., 2017). Although this relationship is widely recognized, the neurophysiological mechanisms by which stress affects medical trainees in procedural settings remain poorly understood, and the specific neural signatures of distress in high-stakes clinical training are largely unknown.

EEG functional connectivity (FC) provides a window into how distributed neural networks coordinate during cognitive and behavioral tasks. FC reflects statistical dependencies between oscillations across brain regions (Bastos and Schoffelen, 2015), indexing functional rather than anatomical connections, and thus offers a lens on how brain networks reorganize under stress and during complex procedural performance (Friston, 2011). This approach has yielded insights into various cognitive states and disorders, underscoring its utility for investigating stress-related neural dynamics in education, yet no prior studies have combined full-montage EEG during procedurally demanding training in anesthesiology.

We aimed to characterize the impact of stress on EEG power and FC in anesthesiology trainees during endotracheal intubation in a high-fidelity simulator. Trainees underwent resting-state EEG (eyes open and closed) and two intubation scenarios—one calm and one stressful—allowing direct comparison of neural dynamics across stress levels. This framework aims to delineate stress-related neural dynamics in procedural training and to identify biomarkers that can guide adaptive, evidence-based educational interventions, fostering safe and effective clinical learning.

The primary hypotheses contended that the stressful scenario would produce increased EEG power, signaling enhanced cortical engagement, concurrently with decreased FC, indicative of diminished connectivity under greater cognitive demand, in comparison with the calm simulation.

2. Methods

2.1. Institutional review board

The study was approved by the University of Vermont Institutional Review Board and written informed consent was obtained from all subjects. The manuscript adheres to the applicable CONSORT (Consolidated Standards of Reporting Trials) guidelines.

2.2. Overview

A total of twenty-nine medical trainees with varying degrees of training in anesthesiology participated in the study. All studies were conducted between 0800 and 1400 h. Upon arrival to the simulation laboratory, participants were oriented to the high-fidelity simulation laboratory and completed the NASA (National Aeronautics and Space Administration) Task Load Index (NASA-TLX).

Participants then watched an instructional video on orotracheal intubation and then had a wireless 22-channel microEEG® system (BioSignal, Acton, MA) and Hexoskin® smart shirt (Hexoskin, Montreal, Canada) with integrated 3‑lead EKG placed. A baseline EKG and resting EEG with eyes opened and closed were done.

Next, each participant completed two scripted scenarios involving induction of general anesthesia and endotracheal intubation. The first scenario (calm) involved an uneventful induction in a supportive environment with a healthy patient undergoing minor surgery. The second (stressful) scenario featured a complex, ill patient, an abrasive attending anesthesiologist, and dismissive, unhelpful OR staff. The scenarios ended once the intubation was successfully completed. Following each scenario the State-Trait Anxiety Inventory (STAI) administered. The participants were then debriefed and the NASA-TLX post-test was administered. A graphic summarizing the study procedure is shown in Fig. 1.

Fig. 1.

Fig. 1

Open in a new tab

Timeline of study (STAI = State–Trait Anxiety Inventory; NASA-TLX National Aeronautics and Space Administration) Task Load Index).

2.3. Questionnaires

The STAI is a 40-item self-report questionnaire that measures two forms of anxiety: state anxiety (STAI-S; how you feel at the moment) and trait anxiety (STAI-T; how you generally feel). It contains two 20-item scales rated on a 4-point Likert scale, with some items reverse-scored, yielding separate total scores for each form that typically range from 20 to 80. Higher scores indicate greater anxiety on the respective scale, and the STAI is widely used in clinical and research settings with well-established reliability and validity across diverse populations.

The NASA TLX is a subjective workload assessment tool that measures perceived mental workload across six dimensions: Mental Demand, Physical Demand, Temporal Demand, Performance, Effort, and Frustration. Users rate each dimension (or rate after a pairwise weighting procedure to derive a workload weight for each dimension), and an overall workload score is computed on a 0–100 scale. It is widely used in human factors and ergonomics to compare task demands across systems and conditions.

2.4. Autonomic tone

Heart rate was analyzed using Kubios® software (Kuopio, Finland) to assess sympathetic and parasympathetic activity. For each participant, we calculated a parasympathetic index (PNSi) and a sympathetic index (SNSi) during both the calm and stressful scenarios.

2.5. EEG

A miniature, wireless EEG system (microEEG®) was used for data collection (Grant et al., 2014; Omurtag et al., 2012). It digitizes the EEG signals close to the electrodes, transmits the digital data wirelessly to a personal computer located within 10 m, and stores the data on an on-board memory card. A cap with embedded electrodes was employed to record the EEG signals. The standard 10–20 electrode placement system, with 19 electrodes, was used (Fp1/Fp2, F3/F4, F7/F8, T7/T8, C3/C4, P3/P4, P7/P8, O1/O2, Fz/Cz/Pz). EEG was recorded with a common reference (A1/A2) and sampled at a rate of 250 Hz. Waveforms were notched filtered at 60 Hz, band-pass filtered from 0.16 to 70 Hz and transformed from reference to a bipolar montage. EEG segments during eye blinks were manually removed. All EEG analyses involved evaluating all pairs of electrodes (19 × 19). Offline analysis was conducted using BESA® software (BESA Research 7.1, BESA Connectivity 2.0; Gräfelfing, Germany). Time -frequency analysis was performed using complex demodulation with an epoch duration of −1000 to 1000 msec with frequency sampling at 1 Hz and time sampling at 50 msec. Spectral power (absolute and relative), coherence, and phase-locking value (PLV) were calculated for each subject. EEG data were assessed during four 60 s sequential states: eyes closed (EC), eyes open (EO), calm scenario, and stressful scenario.

While the use of a 19-electrode EEG system limits spatial resolution, particularly for detailed characterization of functional connectivity patterns, the standard 10–20 electrode configurations remain widely used in both clinical and research settings and have been employed in numerous studies examining large-scale network dynamics. Sensor-level measures such as coherence and phase-locking value (PLV) derived from these systems provide physiologically meaningful estimates of functional connectivity, reflecting synchronization between distributed neural signals (Ma et al., 2024). While higher-density arrays can improve spatial sampling and support more refined source localization, meaningful connectivity patterns can still be detected using conventional electrode configurations, particularly at the level of broad cortical networks.

2.6. Coherence

Coherence is a measure of the linear relationship between two signals at a specific frequency (Bastos and Schoffelen, 2015; Khaleghi et al., 2024). Coherence is defined as the normalized cross-spectrum: C(f) = 〈Sxy(f)〉√〈Sxx(f)〉〈Syy(f)〉, where Sxy is the cross-spectral density between two signals, and Sxx and Syy are the autospectral densities for signals x and y, respectively. Coherence is defined as the magnitude of coherence: Coh(f) = |Cxy(f)| = |〈Sxy(f)〉√〈Sxx(f)〉〈Syy(f)〉|.

The estimated coherence for a given frequency ranges between 0 and 1. A value of 0 indicating that the two signals are totally uncorrelated, and a value of 1 indicates perfect correlation. Coherence is a bivariate measure, which means it considers only two signals simultaneously. Coherence is also a nondirectional connectivity measure, that is Coh(f) = Cohyx(f) where coherence at a particular frequency (f) between two signals (y) and (x) is the same regardless of the order in which the signals are considered. Numerical values of coherence from each electrode pair at 1–30 Hz were analyzed for all four states (EO, EC, Calm scenario, Stressful scenario).

Fig. 2 shows example coherence at 7 Hz during the eyes-open (EO), calm, and stressful scenarios, displayed in both 3D and circular graph formats. Coherence increases from EO to the calm scenario and increases further in the stressful scenario. Supplementary Figure S1, Supplementary Figure S2, Supplementary Figure S3 provide additional 3D and circular coherence examples at 4, 7, 10, and 16 Hz, showing the same pattern of increasing coherence across electrode pairs from EO to calm to stressful conditions.

Fig. 2.

Fig. 2

Open in a new tab

Example of coherence values during eyes open (A), calm (B) and stressful (C) scenarios. Coherence values are presented for 7 Hz. in 3D (top panel) and circular graph mode (bottom panel). The coherence values are colored coded with the darker red lines showing the highest coherences. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Supplementary Figure S1.

Supplementary Figure S1

Open in a new tab

Example of coherences during eyes opened condition. Coherence values are presented for 4, 7, 10 and 16 Hz. In 3D (top panel) and circular graph mode (bottom panel). The coherence values are colored coded with the darker red showing the highest coherences.

Supplementary Figure S2.

Supplementary Figure S2

Open in a new tab

Example of coherences during calm scenario. Coherence values are presented for 4, 7, 10 and 16 Hz. In 3D (top panel) and circular graph mode (bottom panel). The coherence values are colored coded with the darker red lines showing the highest coherences.

Supplementary Figure S3.

Supplementary Figure S3

Open in a new tab

Example of coherences during stressful scenario. Coherence values are presented for 4, 7, 10 and 16 Hz. In 3D (top panel) and circular graph mode (bottom panel). The coherence values are colored coded with the darker red lines showing the highest coherences.

2.7. Phase locking value (PLV)

PLV is a measure for phase synchronization (Bastos and Schoffelen, 2015). The PLV is defined based on the average of phase angle difference between two signals over time at a specific frequency: PLVxy(f) = |1nΣei∙(φx,t−φy,t)nt=1|, where n is the number of time points, i is the imaginary number that results from taking the square root of −1, and φx and φy are phase angles from signals x and y at frequency f. The PLV takes values between 0 and 1. A value close to 1 means perfect phase locking, while a value close to 0 results from a random phase distribution over time. Like coherence, PLV is nondirectional.

Coherence and PLV provide similar results in many cases. In this study, there was a strong correlation between total coherence and PLV (Pearson r = 0.8934; XY pairs = 965; p < 0.0001). However, coherence is influenced by strong increases or decreases in power as the analysis includes power information of the signals. For instance, if connectivity increases, but the amplitudes of signals simultaneously decrease, estimated coherence values may be biased. PLV is insensitive to the amplitudes of signals and only depends on the phase relationship between two signals. In this study the PLV were generally lower than the coherences. Fig. 3 provides an example from this study showing coherence and PLV at 7 Hz during a stressful scenario.

Fig. 3.

Fig. 3

Open in a new tab

Comparison of coherences and PLV for 7 Hz theta activity during the stress scenario. 3 DV (left) and Circular Graph View (right) of coherence values (A) and PLV (B). Right hand panel shows example of coherences (A) and PLV (B) from the T7-T8 electrode pairs across frequencies. Note the PLV had lower values than the coherence values. Coherence and PLV are color coded with darker colors indicative of higher values.

Total coherence and PLV refer to all frequencies (1–30 Hz) for all electrode pairs. The total coherence, PLV and absolute power does not consider regional measures of connectivity or power but does provide a broad measure of connectivity and power. Each band (delta [0- < 4 Hz], theta [4- < 8 Hz], alpha [8- < 13 Hz] and beta [13–30 Hz]) were also measured. For connectivity measures brain regions were also assessed: 1. Frontal (Fp1/Fp2, F3/F4); 2. Temporal (F7/F8, T7/T8); 3. Central (C3/C4); 4. Parietal (P3/P4, P7/P8); 5. Occipital (O1/O2); 6. Midline (Fz/Cz/Pz).

2.8. Spectral power

After a single taper with the Hamming windowing function, the fast Fourier transform (FFT) of the EEG was calculated from 0 to 30 Hz using a sliding window (window = 1 s, overlap = 0.5 s). Both absolute and relative power of delta, theta, alpha and beta were calculated for all states.

2.9. Eye blinks

Blinking is an automatic behavior sensitive to affect, cognition, and arousal (Stern et al., 1984). The eye blink rate has been proposed as a noninvasive biomarker of the dopamine system (Demiral et al., 2022) and with increased attention and mental load eye blink rate decreases (Brookings et al., 1996; Wiberg et al., 2015). However, increases in arousal are associated with increases in endogenous blink production (Stern et al., 1984). To examine the effect of the two scenarios on eye blinking the EEG was used to count eye blinks for 5 min during the EO resting state, Calm scenario and Stressful scenario.

2.10. Sample size calculations

Based on preliminary studies we calculated sample size by an anticipated difference in coherence (target difference [delta]) between the calm and stressful scenario of 0.4 with a standard deviation of differences of 0.5 with a two-sided alpha of 0.05 and power of 0.80 which resulted in a minimal sample size of 14 paired subjects. To account for dropouts and robust variance more subjects were recruited.

2.11. Statistics

For the EEG measures each participant's data was compared during the four states (EO, EC, Calm scenario, Stressful scenario). Eye blinks were compared in the EO, Calm Scenario and Stressful scenario). For all group comparisons the ANOVA with repeated measures was used followed by the paired t-test for individual components. Pearson r was used to assess correlations between coherences and PVL, eye blinks with coherences and total coherences and PLV values with time to intubation in the calm and stressful scenarios. For p values a false discovery rate using the two-stage linear step-up procedure of Benjamnini, Krieger and Yekutieli (BKY) correction was used to generate q values (Benjamini, 2006). All measures are presented as the means ± standard deviations to evaluate the variability between individuals in the population mean (Nagele, 2003). All EEG data was analyzed using GraphPad Prism version 10.0.0. Due to the large number of statistical analyses, all statistical results are included in tabular form in the Supplemental tables.

3. Results

Of the 29 subjects who participated in the study eight subjects were medical students (2nd year = 6; 3rd year = 2), three were interns and 18 were anesthesiology residents (1st year = 7; 2nd year = 5; 3rd year = 6). There were 10 females and 18 males. The mean age was 30.7 ± 6.22 yrs.

3.1. Questionnaires

The STAI survey results following the stressful scenario (38.11 ± 12.23) had a significantly higher mean score compared to the calm scenario (31.17 ± 6.297)(t = 3.094, df = 27, p = 0.0046). In the NASA TLD the mental, physical, temporal demands and performance, effort and frustration scales were significantly higher following the stressful scenario than during the baseline testing (all q values ≤ < 0.001)(Suppl. Table S1). The STAI and NASA TLD demonstrated that the participants that the participants felt stress following the endotracheal stimulations and that the stressful stimulation produced more stress than the calm scenario.

3.2. Heart rate/autonomic tone

Autonomic data on heart rate was obtained in only 14 patients due to equipment failure. SNSi was significantly lower during the calm scenario (2.146 ± 1.785) than the stressful scenario (3.256 ± 2.980; paired t-test = 2.259, df = 13). The PNSi was significantly higher in the calm scenario (−1.463, df = 0.237) than the stressful scenario (−1.889; paired t-test = 2.458, df = 13). These findings indicate that the two scenarios differed in effect on the autonomic nervous system supporting the premise that the stressful scenario led to increase in sympathetic activity as evidenced by increased heart rate.

3.3. EEG coherences and PLV for all electrodes pairs

Across all participants, there were significant differences in both coherence and phase-locking value (PLV) across the four states—eyes closed (EC), eyes open (EO), calm, and stressful scenarios—when considering all electrode pairs and frequency bands (delta, theta, alpha, and beta)(Fig. 4). These differences were confirmed by ANOVA, with all groups showing q-values less than 0.0001 (see Suppl. Tables S2 and S3).

Fig. 4.

Fig. 4

Open in a new tab

Across all participants, there were significant differences in both coherence (A) and phase-locking value (PLV) across the four states—eyes closed, eyes open, calm, and stressful scenarios—when considering all electrode pairs and frequency bands (delta, theta, alpha, and beta). (* p ≤ 0.0001; # p < 0.001; + p < 0.01; ^ p < 0.05).

Overall, coherence and PLV were lowest in the delta band and highest in the alpha band. Both measures, across all frequencies, were higher during the stressful scenario compared to the calm scenario, as determined by paired t-tests (Suppl. Tables S2 and S3). Additionally, coherence and PLV were greater during EC than during EO in total and most frequency bands. An increase in coherence and PLV was observed between the EO state and the calm scenario, particularly in the total and beta bands.

Summary connectivity measures for each participant are presented in Table 1. Most participants demonstrated a stepwise increase in both coherence and phase-locking value (PLV) across conditions, with higher values observed from the eyes-open (EO) state to the calm scenario, and further increases from the calm to the stressful scenario.

Table 1.

Mean (± SD) total coherence and PLV values for each subject. Yellow cells indicate increases from the EO condition to the calm scenario, whereas green cells indicate increases from the calm to the stressful scenario.

3.3.

Open in a new tab

These findings indicate that coherence and PLV vary systematically with experimental state, supporting the interpretation that functional connectivity is state-dependent. In particular, the consistent elevation of both measures under stress suggests that stress is associated with increased functional connectivity across electrode pairs and frequency bands (MD = Missing Data).

3.4. Relationship between FC and performance

Time to intubation in the calm and stressful scenario were available for 25 of the trainees. The time to successful intubation was longer than in the calm (343.2 ± 165.2 s) scenario than the stressful scenario (218.9 s; t = 3.810, df = 23, p < 0.0009). Despite the stressful environment, 84% (21/25) of the trainees had shorter times to intubation than during the calm scenario. However, there was no significant correlation between total coherence or PLV to time to intubation during either the calm or stressful scenario (data not shown).

3.5. Regional EEG coherences and PLV

Regional EEG coherences (Fig. 5) and PLV (Fig. 6) were assessed for frontal, temporal, central, parietal, midline and occipital electrodes. As with the analysis of all electrode pairs, electrodes pairs in the frontal, temporal, central, parietal and midline electrodes showed a similar pattern to the analysis of all electrode pairs. Coherences and PLV were highest in the alpha band and lowest in the delta band. Coherences and PLV were higher during EC than during EO. During both the calm and stress scenarios coherences and PVL were higher in the stressful scenario than the calm scenario in all brain regions except the occipital region (Suppl. Tables T4-T15).

Fig. 5.

Fig. 5

Open in a new tab

Coherence values for frontal (A), central (B), temporal (C), parietal (D), midline (E) and occipital (F) electrode pairs during the four conditions across all bands (total, delta, theta, alpha, beta). (* p ≤ 0.0001; # p < 0.001; + p < 0.01; ^ p < 0.05).

Fig. 6.

Fig. 6

Open in a new tab

Phase locking values for frontal (A), central (B), temporal (C), parietal (D), midline (E) and occipital (F) electrode pairs during the four conditions across all bands (total, delta, theta, alpha, beta). (* p ≤ 0.0001; # p < 0.001; + p < 0.01; ^ p < 0.05).

Occipital electrodes differed from other brain regions. For both total and all frequency bands, coherences and PLV were lower than other brain regions. The increase in coherence and PLV from EO to calm to stressful scenario seen in other brain regions was not present in the occipital region (Figs. 5F, 6F; Suppl. Tables T14-T15) The PLV was lower during the stressful scenario than during the calm scenario in total and all four bands (Fig. 6F, Suppl. Table T15). As described below, eye blinks increased from the EO state to the calm and stressful scenario. Since it is known that 50 ms before and during an eye blink assimilation of information is impaired (Stern et al., 1984), the relationship between eye blink frequency and occipital coherence was assessed. There was a negative correlation between eye blink frequency and occipital electrode pairs (Pearson r = −0.3278, p = 0.0120, XY pairs = 58).

Taken together these results show that there are region specific changes in FC connectivity during state changes from EO to calm to stressful scenarios. All brain regions except the occipital region showed increases in coherence and PLV from the EO to calm to stressful scenarios, The reduced coherences and PLV in the occipital regions may related to the impairment of visual information caused by eye blinks.

3.6. Spectral power

Absolute and relative spectral power was compared in the four conditions. Because of the wide variability in the spectral power, it was graphed using a log10 scale. As expected, alpha power was high during the EC state. Compared to the EO state, spectral power was higher in both the calm and stressful scenarios in total power and power in the delta and beta bands (Fig. 7; Suppl. Table 16). Regarding relative power, like absolute power, alpha was high during EC. Delta relative power was higher than EO during the calm scenario and beta power was highest during the stressful scenario (Fig. 7; Suppl. Table 17). These findings show that there is an increase in total, delta, theta and beta spectral power with both calm and stressful scenarios compared to the EO state. The increases were greater for the stressful scenario than the calm scenario. Relative changes in power were less apparent except for relative beta being increased in the stressful scenario.

Fig. 7.

Fig. 7

Open in a new tab

Absolute and relative spectral power across the four states—eyes closed, eyes open, calm, and stressful scenarios. (* p ≤ 0.0001; # p < 0.001; + p < 0.01; ^ p < 0.05).

3.7. Eye blinks

There were significant increases in eye blinks between EO (81.93 ± 31.62) in both the calm (126.2 ± 83.87) and stressful scenario (161.6 ± 51.76) as well as an increase from the calm scenario to the stressful scenario (ANOVA and paired t comparisons (p < 0.0001; Suppl. Table S18). There was a strong correlation between the number of eye blinks during the calm and stressful scenario (Pearson r = 0.8421; XY pairs = 29; p < 0.0001). However, there was not a correlation between eye blinks and coherence in either the calm (Pearson r = 0.0992; XY pairs = 29; p = 0.6084) or stressful scenario (Pearson r = 0.1279; XY pairs = 29; p = 0.5086). This finding shows that while eye blinks increased during the stressful scenario there was not a direct link between eye blinks and coherence.

3.8. Comparison of medical students and interns and residents in EEG measures and eye blinks

To determine if there were differences between the eight medical students and 21 interns and residents, the following comparisons were made between the two groups during the calm and stressful scenarios: i) total coherence and PLV across all electrode pairs; ii) total absolute spectral power; and iii) eye blinks. Statistical significance was not noted between the two groups in any of these measures.

4. Discussion

The study assessed FC and EEG spectral power in medical students and residents, during simulated endotracheal intubation in a laboratory setting. Participants completed two scenarios: one in a calm, supportive environment and another in an emergency situation characterized by chaos and a lack of support. The NASA TLX and STAI questionnaires, heart rate and eye blink data indicated that the two scenarios increased stress, with the stressful scenario producing more stress than the calm scenario. Although endotracheal intubation is inherently stressful, it was expected that stress levels would be significantly higher during the emergency scenario. Based on selected studies (Hag et al., 2021; Lee et al., 2014; Saffari et al., 2023) and preliminary findings, the original hypothesis was that EEG FC would decrease (be impaired) under stress. However, this hypothesis was not supported; instead, both scenarios showed increased FC, with higher values observed during the stressful scenario compared to the calm one.

Regional differences were noted in the connectivity measures: coherence and phase-locking value (PLV) were higher in the theta and alpha bands than in delta and beta bands. Both coherence and PLV were generally lower in the occipital region compared to other brain areas, and the expected increases from eyes open (EO) to calm and stress scenarios were not observed in the occipital region.

Regarding spectral power, absolute beta power was higher during the stressful scenario across total, delta, theta, and beta bands. Relative power in the beta band was also highest during the stressful condition. These findings suggest that, contrary to initial expectations, FC increases under stress, with specific regional and spectral variations.

4.1. Coherence and PLV as measures of FC

The synchronization and coordination of waveforms between different brain regions are essential for nearly all cognitive functions and have been a major focus in both basic and clinical neuroscience. Oscillatory activity patterns in brain networks are thought to organize various computations, enabling the brain to either integrate or segregate information within oscillatory cycles depending on cognitive demands (Galinsky and Frank, 2023). Measures such as coherence and phase-locking value (PLV) quantify the degree of synchrony between brain regions (Chiarion et al., 2023; Srinivasan et al., 2007; Thatcher, 2012; Thatcher et al., 2008). Coherence, in particular, reflects dynamic coupling between areas necessary for integrating distributed information (Varela et al., 2001).

Changes in neural coherence are associated with different cognitive processes. For example, coherence between the hippocampus and cingulate cortex (Remondes and Wilson, 2013), the medial prefrontal cortex (Hyman et al., 2005; Jones and Wilson, 2005; Siapas et al., 2005), and the amygdala (Terada et al., 2013) has been observed during decision-making and reward processing. Enhanced coherence between the medial prefrontal cortex and striatum has been linked to motor behavior and learning (Antzoulatos and Miller, 2014; DeCoteau et al., 2007; von Nicolai et al., 2014). Neural information transfer is reflected in the timing of action potentials, which can occur in bursts organized by the phase of ongoing oscillations, such as theta rhythms. The phase of these oscillations can encode information; for instance, in rats performing a Y-maze task, the preferred firing phase of prefrontal cortex action potentials spikes relative to hippocampal theta shifts near decision points (Benchenane et al., 2010). The hippocampus dynamically combines spatial navigation computations driven by external cues with internally generated memory sequences (McNaughton et al., 1983; Pastalkova et al., 2008; Wang et al., 2015). By accessing specific hippocampal information during certain EEG phases, cortical neurons fire precisely when a decision is required (Benchenane et al., 2010).

4.2. Stress and FC

Assessing the impact of stress on FC is challenging due to the diversity of studied disorders, the variety of tests used, the brain regions examined, the types of stressors applied, and the outcome measures employed (Katmah et al., 2021; Vanhollebeke et al., 2023). Additionally, multiple methods exist for estimating FC from EEG data (Bastos and Schoffelen, 2015; Puthanmadam Subramaniyam and T, 2025).

For example, Kim et al. (Kim et al., 2017) investigated how stress and task difficulty influenced working memory and cortical communication in 38 adolescents performing easy and hard memory tasks under low- and high-stress conditions. They analyzed accuracy, reaction time, and inter- and intrahemispheric EEG coherence. Their findings showed that increased task difficulty and stress decreased cortical communication between memory-related areas, with notable differences between coherences in the left and right hemispheres. Similarly, Woo et al. (Woo and Kim, 2017) examined cortical coherence in 14 air pistol athletes during practice and competition. They found higher interhemispheric coherence during aiming in competition compared to practice, but no significant differences in intrahemispheric coherence between the two conditions.

Across several studies, the impact of stress on functional connectivity (FC) measured by EEG has yielded mixed findings. Alonso et al. (Alonso et al., 2015) reported that stress induced by the Stroop task and sleep deprivation produced varied coherence changes: beta coherence increased under stress, theta coherence rose during the Stroop task, and alpha coherence decreased after sleep deprivation. Al-Shargie et al. (Al-Shargie et al., 2023) found that mental stress triggered by the Stroop task increased connectivity between frontal and parietal regions. In contrast, Saffari et al. (Saffari et al., 2023) observed that higher mental stress—specifically, watching a video while performing a digit span task—was associated with decreased FC (via phase-locking value, PLV) in frontocentral, frontoparietal, and centroparietal areas. Hag et al. (Hag et al., 2021) used a time-pressured mental arithmetic task with negative feedback and reported a significant reduction in FC under stress, assessed with EEG PLV. Collectively, these findings illustrate inconsistencies in how stress affects FC, likely reflecting differences in stress type, cognitive load, and measurement approaches.

In summary, studies on stress and FC have produced variable results, likely due to differences in stress types, load, and outcome measures, making it a complex area of research.

4.3. Clinical significance of the results

In this study, the observed increase in coherence and phase-locking value during both the calm and stressful scenarios suggested that trainees were able to enhance brain connectivity in response to different conditions. These findings indicate that, despite the stress, trainees could increase communication between various brain regions (Srinivasan et al., 2007). Although it is challenging to directly link increased connectivity with improved performance, the implication is that heightened connectivity may lead to higher-quality execution. Successfully performing endotracheal intubation requires coordinated activity across multiple brain areas, including the sensory, visual, and motor cortex, as well as subcortical structures like the striatum, and higher-level regions such as the frontal lobes and hippocampus involved in memory. Interestingly, in this context, stress appeared to stimulate greater connectivity between these regions, which could be beneficial.

Interestingly, post hoc analyses revealed no differences in functional connectivity measures between anesthesiology residents and medical students/interns. However, the study was not powered to detect differences in coherence or PLV across these groups, which represent varying levels of intubation experience. Given that stress responses would be expected to differ between novice learners and more experienced residents, a larger study with adequate representation of both groups is warranted to more definitively evaluate these potential differences.

4.4. Imitations of the study

Stress augmented functional connectivity (FC) during procedural learning; however, the present findings remain fundamentally correlational and do not support a causal linkage between FC dynamics and performance outcomes. This limitation is consistent with broader neuroimaging literature, in which commonly used FC metrics such as coherence and phase-locking value reflect statistical inter-regional associations rather than directed or mechanistic interactions, thereby constraining causal inference (Reid et al., 2019). Although trainees demonstrated reduced time to intubation under stressful conditions, the absence of significant correlations between FC measures and performance suggests that FC changes may not directly mediate skill execution in this context, or that the selected metrics lack sufficient sensitivity to capture behaviorally relevant network dynamics.

An alternative explanation for improved performance is a practice or order effect. Because the calm scenario uniformly preceded the stressful condition, participants may have benefited from repetition-dependent motor learning, which is known to induce rapid neuroplastic changes and performance gains independent of contextual modifiers such as stress (Bassett et al., 2015). Indeed, motor skill acquisition is strongly influenced by cumulative practice exposure, and early improvements in efficiency may reflect task familiarization rather than stress-enhanced learning per se. Additionally, stress exerts heterogeneous effects on motor learning and performance, with evidence suggesting that individual variability in physiological reactivity and task context can differentially facilitate or impair skill acquisition (LaFollette et al., 2026). These interacting factors may obscure simple linear relationships between FC and behavioral outcomes.

Future investigations should adopt more rigorous experimental designs to clarify these relationships. Randomized or counterbalanced task sequencing would mitigate order effects, while incorporation of effective connectivity approaches or perturbational methods (e.g., neurostimulation) may better address causal brain–behavior relationships. Furthermore, examining FC–performance coupling across diverse instructional paradigms—such as variable practice, interleaving, or stress inoculation—may provide greater insight into how training design influences both neural network organization and skill transfer.

A notable limitation of this study is the substantial loss of cardiac data, affecting more than 50% of participants due to equipment failure. This reduces statistical power and limits the integration of autonomic markers with neural and behavioral measures. Nevertheless, the available cardiac data, together with STAI, NASA-TLX, and ocular blink metrics, consistently demonstrate that participants experienced stress during the scenarios, with higher stress levels observed in the stressful condition compared to the calm condition. Prior work suggests that physiological stress responses can modulate neural connectivity and influence learning trajectories, underscoring the importance of incorporating multimodal measures in future studies. Prior work suggests that physiological stress responses can both modulate neural connectivity and influence learning trajectories, further underscoring the importance of integrating multimodal measures in future studies (Andersen et al., 2018).

An additional methodological limitation is the fixed sequencing of experimental conditions. Without randomization, it is not possible to disentangle the independent effects of stress from those of prior exposure or learning consolidation. Future studies employing randomized or crossover designs, along with larger sample sizes and more robust physiological data acquisition, will be essential to isolate the specific contribution of stress to psychomotor skill acquisition and its neural correlates.

5. Conclusion

This study demonstrates that stress modulates neural connectivity during procedural learning in a high-stakes medical setting. Using concurrent EEG recordings in medical trainees performing endotracheal intubation across baseline, calm, and stressful simulation conditions, we found that FC increased relative to rest during all task conditions, consistent with broad engagement of neural networks during skill acquisition. Notably, FC was further elevated in the stressful condition, suggesting that, within a certain range, stress may enhance neural communication and support learning processes.

The stress response—largely mediated by sympathetic activation—was reliably elicited using high-fidelity simulation. This was supported by convergent physiological and behavioral measures, including heart rate, ocular blink rate, and subjective assessments of stress and cognitive load, reinforcing the functional relevance of the observed neural changes.

Together, these findings suggest the presence of a measurable neurophysiological signature of stress that may be leveraged to inform personalized educational strategies. Calibrating stress exposure during training could help optimize learning while minimizing potential negative effects. In addition, this study supports the feasibility of real-time EEG monitoring to capture stress-related neural dynamics in medical training environments, highlighting the potential value of incorporating neurophysiological metrics into simulation-based curriculum design.

Future studies should investigate causal relationships between FC changes and performance outcomes, as well as evaluate adaptive training approaches that individualize stress exposure. The use of higher-density EEG systems and source-space analyses may further improve the spatial resolution of connectivity patterns and extend these findings.

The following are the supplementary data related to this article.

Supplementary material
mmc4.docx (73.2KB, docx)

Authors contribution

FLK, EEW and GLH participated in the study design. FLK and AFF conducted the study. FLK, SS and GLH performed the statistical analysis of the results. The first draft of the paper was conducted by FLK and GLH. All authors reviewed the final manuscript.

Declaration of generative AI and AI-assisted technologies in the manuscript preparation process

During the preparation of this work the authors used AI CHATBOX and Doximity in order to improve grammar and typographical errors. After using this tool/service, the authors reviewed and edited the content as needed and takes full responsibility for the content of the published article.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This research was supported by a grant to FLK from the Society for Education in Anesthesia (SEA) titled The Electroencephalographic Signature of Distress During Procedural Learning in Anesthesiology Residents.

References

  1. Alonso J.F., Romero S., Ballester M.R., Antonijoan R.M., Mañanas M.A. Stress assessment based on EEG univariate features and functional connectivity measures. Physiol. Meas. 2015;36(7):1351–1365. doi: 10.1088/0967-3334/36/7/1351. [DOI] [PubMed] [Google Scholar]
  2. Al-Shargie F., Badr Y., Tariq U., Babiloni F., Al-Mughairbi F., Al-Nashash H. Classification of mental stress levels using EEG connectivity and convolutional neural networks. Annu Int Conf IEEE Eng Med Biol Soc. 2023;2023:1–5. doi: 10.1109/embc40787.2023.10340398. [DOI] [PubMed] [Google Scholar]
  3. Andersen J.P., Di Nota P.M., Beston B., Boychuk E.C., Gustafsberg H., Poplawski S., et al. Reducing lethal force errors by modulating police physiology. J. Occup. Environ. Med. 2018;60(10):867–874. doi: 10.1097/jom.0000000000001401. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Antzoulatos E.G., Miller E.K. Increases in functional connectivity between prefrontal cortex and striatum during category learning. Neuron. 2014;83(1):216–225. doi: 10.1016/j.neuron.2014.05.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bassett D.S., Yang M., Wymbs N.F., Grafton S.T. Learning-induced autonomy of sensorimotor systems. Nat. Neurosci. 2015;18(5):744–751. doi: 10.1038/nn.3993. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Bastos A.M., Schoffelen J.M. A tutorial review of functional connectivity analysis methods and their interpretational pitfalls. Front. Syst. Neurosci. 2015;9:175. doi: 10.3389/fnsys.2015.00175. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Benchenane K., Peyrache A., Khamassi M., Tierney P.L., Gioanni Y., Battaglia F.P., et al. Coherent theta oscillations and reorganization of spike timing in the hippocampal- prefrontal network upon learning. Neuron. 2010;66(6):921–936. doi: 10.1016/j.neuron.2010.05.013. [DOI] [PubMed] [Google Scholar]
  8. Benjamini Y.K., Krieger A.M., Yekutieli D. Adaptive linear step-up procedures that control the false discovery rate. Biometrika. 2006;93:491–507. [Google Scholar]
  9. Brookings J.B., Wilson G.F., Swain C.R. Psychophysiological responses to changes in workload during simulated air traffic control. Biol. Psychol. 1996;42(3):361–377. doi: 10.1016/0301-0511(95)05167-8. [DOI] [PubMed] [Google Scholar]
  10. Chiarion G., Sparacino L., Antonacci Y., Faes L., Mesin L. Connectivity analysis in EEG data: a tutorial review of the state of the art and emerging trends. Bioengineering (Basel) 2023;10(3) doi: 10.3390/bioengineering10030372. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. DeCoteau W.E., Thorn C., Gibson D.J., Courtemanche R., Mitra P., Kubota Y., et al. Oscillations of local field potentials in the rat dorsal striatum during spontaneous and instructed behaviors. J. Neurophysiol. 2007;97(5):3800–3805. doi: 10.1152/jn.00108.2007. [DOI] [PubMed] [Google Scholar]
  12. Demiral Ş.B., Manza P., Biesecker E., Wiers C., Shokri-Kojori E., McPherson K., et al. Striatal D1 and D2 receptor availability are selectively associated with eye-blink rates after methylphenidate treatment. Commun. Biol. 2022;5(1):1015. doi: 10.1038/s42003-022-03979-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Friston K.J. Functional and effective connectivity: a review. Brain Connect. 2011;1(1):13–36. doi: 10.1089/brain.2011.0008. [DOI] [PubMed] [Google Scholar]
  14. Galinsky V.L., Frank L.R. Critically synchronized brain waves form an effective, robust and flexible basis for human memory and learning. Sci. Rep. 2023;13(1):4343. doi: 10.1038/s41598-023-31365-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Grant A.C., Abdel-Baki S.G., Omurtag A., Sinert R., Chari G., Malhotra S., et al. Diagnostic accuracy of microEEG: a miniature, wireless EEG device. Epilepsy Behav. 2014;34:81–85. doi: 10.1016/j.yebeh.2014.03.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Hag A., Handayani D., Pillai T., Mantoro T., Kit M.H., Al-Shargie F. EEG mental stress assessment using hybrid multi-domain feature sets of functional connectivity network and time-frequency features. Sensors (Basel) 2021;21(18) doi: 10.3390/s21186300. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Hyman J.M., Zilli E.A., Paley A.M., Hasselmo M.E. Medial prefrontal cortex cells show dynamic modulation with the hippocampal theta rhythm dependent on behavior. Hippocampus. 2005;15(6):739–749. doi: 10.1002/hipo.20106. [DOI] [PubMed] [Google Scholar]
  18. Jones M.W., Wilson M.A. Phase precession of medial prefrontal cortical activity relative to the hippocampal theta rhythm. Hippocampus. 2005;15(7):867–873. doi: 10.1002/hipo.20119. [DOI] [PubMed] [Google Scholar]
  19. Katmah R., Al-Shargie F., Tariq U., Babiloni F., Al-Mughairbi F., Al-Nashash H. A review on mental stress assessment methods using EEG signals. Sensors (Basel) 2021;21(15) doi: 10.3390/s21155043. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Khaleghi N., Hashemi S., Peivandi M., Ardabili S.Z., Behjati M., Sheykhivand S., et al. EEG-based functional connectivity analysis of brain abnormalities: a systematic review study. Informatics in Medicine Unlocked. 2024;47 doi: 10.1016/j.imu.2024.101476. [DOI] [Google Scholar]
  21. Kim Y., Woo J., Woo M. Effects of stress and task difficulty on working memory and cortical networking. Percept. Mot. Skills. 2017;124(6):1194–1210. doi: 10.1177/0031512517732851. [DOI] [PubMed] [Google Scholar]
  22. LaFollette K.J., Frank D.J., Burgoyne A.P., Macnamara B.N. Task, person, and experiential characteristics drive the transfer of learning. Commun. Psychol. 2026;4(1) doi: 10.1038/s44271-026-00408-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. LeBlanc V.R. The effects of acute stress on performance: implications for health professions education. Acad. Med. 2009;84(10 Suppl):S25–S33. doi: 10.1097/ACM.0b013e3181b37b8f. [DOI] [PubMed] [Google Scholar]
  24. Lee S.H., Yoon S., Kim J.I., Jin S.H., Chung C.K. Functional connectivity of resting state EEG and symptom severity in patients with post-traumatic stress disorder. Prog. Neuro-Psychopharmacol. Biol. Psychiatry. 2014;51:51–57. doi: 10.1016/j.pnpbp.2014.01.008. [DOI] [PubMed] [Google Scholar]
  25. Ma X., Qi Y., Xu C., Weng Y., Yu J., Sun X., et al. How well do neural signatures of resting-state EEG detect consciousness? A large-scale clinical study. Hum. Brain Mapp. 2024;45(4) doi: 10.1002/hbm.26586. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. McNaughton B.L., Barnes C.A., O’Keefe J. The contributions of position, direction, and velocity to single unit activity in the hippocampus of freely-moving rats. Exp. Brain Res. 1983;52(1):41–49. doi: 10.1007/BF00237147. [DOI] [PubMed] [Google Scholar]
  27. Nagele P. Misuse of standard error of the mean (SEM) when reporting variability of a sample. A critical evaluation of four anaesthesia journals. Br. J. Anaesth. 2003;90(4):514–516. doi: 10.1093/bja/aeg087. [DOI] [PubMed] [Google Scholar]
  28. Omurtag A., Baki S.G., Chari G., Cracco R.Q., Zehtabchi S., Fenton A.A., et al. Technical and clinical analysis of microEEG: a miniature wireless EEG device designed to record high-quality EEG in the emergency department. Int. J. Emerg. Med. 2012;5(1) doi: 10.1186/1865-1380-5-35. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Pastalkova E., Itskov V., Amarasingham A., Buzsáki G. Internally generated cell assembly sequences in the rat hippocampus. Science. 2008;321(5894):1322–1327. doi: 10.1126/science.1159775. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Puthanmadam Subramaniyam N., T C.T. A novel method for estimating functional connectivity from EEG coherence potentials. Sci. Rep. 2025;15(1) doi: 10.1038/s41598-025-94076-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Reid A.T., Headley D.B., Mill R.D., Sanchez-Romero R., Uddin L.Q., Marinazzo D., et al. Advancing functional connectivity research from association to causation. Nat. Neurosci. 2019;22(11):1751–1760. doi: 10.1038/s41593-019-0510-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Remondes M., Wilson M.A. Cingulate-hippocampus coherence and trajectory coding in a sequential choice task. Neuron. 2013;80(5):1277–1289. doi: 10.1016/j.neuron.2013.08.037. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Saffari F., Norouzi K., Bruni L.E., Zarei S., Ramsøy T.Z. Impact of varying levels of mental stress on phase information of EEG signals: a study on the frontal, central, and parietal regions. Biomedical Signal Processing and Control. 2023;86 doi: 10.1016/j.bspc.2023.105236. [DOI] [Google Scholar]
  34. Schwabe L., Wolf O.T. Learning under stress impairs memory formation. Neurobiol. Learn. Mem. 2010;93(2):183–188. doi: 10.1016/j.nlm.2009.09.009. [DOI] [PubMed] [Google Scholar]
  35. Schwabe L., Wolf O.T., Oitzl M.S. Memory formation under stress: quantity and quality. Neurosci. Biobehav. Rev. 2010;34(4):584–591. doi: 10.1016/j.neubiorev.2009.11.015. [DOI] [PubMed] [Google Scholar]
  36. Siapas A.G., Lubenov E.V., Wilson M.A. Prefrontal phase locking to hippocampal theta oscillations. Neuron. 2005;46(1):141–151. doi: 10.1016/j.neuron.2005.02.028. [DOI] [PubMed] [Google Scholar]
  37. Srinivasan R., Winter W.R., Ding J., Nunez P.L. EEG and MEG coherence: measures of functional connectivity at distinct spatial scales of neocortical dynamics. J. Neurosci. Methods. 2007;166(1):41–52. doi: 10.1016/j.jneumeth.2007.06.026. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Stern J.A., Walrath L.C., Goldstein R. The endogenous eyeblink. Psychophysiology. 1984;21(1):22–33. doi: 10.1111/j.1469-8986.1984.tb02312.x. [DOI] [PubMed] [Google Scholar]
  39. Terada S., Takahashi S., Sakurai Y. Oscillatory interaction between amygdala and hippocampus coordinates behavioral modulation based on reward expectation. Front. Behav. Neurosci. 2013;7:177. doi: 10.3389/fnbeh.2013.00177. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Thatcher R.W. Coherence, phase differences, phase shift, and phase lock in EEG/ERP analyses. Dev. Neuropsychol. 2012;37(6):476–496. doi: 10.1080/87565641.2011.619241. [DOI] [PubMed] [Google Scholar]
  41. Thatcher R.W., North D.M., Biver C.J. Development of cortical connections as measured by EEG coherence and phase delays. Hum. Brain Mapp. 2008;29(12):1400–1415. doi: 10.1002/hbm.20474. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Vanhollebeke G., Kappen M., De Raedt R., Baeken C., van Mierlo P., Vanderhasselt M.-A. Effects of acute psychosocial stress on source level EEG power and functional connectivity measures. Sci. Rep. 2023;13(1):8807. doi: 10.1038/s41598-023-35808-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Varela F., Lachaux J.P., Rodriguez E., Martinerie J. The brainweb: phase synchronization and large-scale integration. Nat. Rev. Neurosci. 2001;2(4):229–239. doi: 10.1038/35067550. [DOI] [PubMed] [Google Scholar]
  44. Vogel S., Klumpers F., Schröder T.N., Oplaat K.T., Krugers H.J., Oitzl M.S., et al. Stress induces a shift towards striatum-dependent stimulus-response learning via the mineralocorticoid receptor. Neuropsychopharmacology. 2017;42(6):1262–1271. doi: 10.1038/npp.2016.262. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. von Nicolai C., Engler G., Sharott A., Engel A.K., Moll C.K., Siegel M. Corticostriatal coordination through coherent phase-amplitude coupling. J. Neurosci. 2014;34(17):5938–5948. doi: 10.1523/jneurosci.5007-13.2014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Wang Y., Romani S., Lustig B., Leonardo A., Pastalkova E. Theta sequences are essential for internally generated hippocampal firing fields. Nat. Neurosci. 2015;18(2):282–288. doi: 10.1038/nn.3904. [DOI] [PubMed] [Google Scholar]
  47. Wiberg H., Nilsson E., Lindén P., Svanberg B., Poom L. Physiological responses related to moderate mental load during car driving in field conditions. Biol. Psychol. 2015;108:115–125. doi: 10.1016/j.biopsycho.2015.03.017. [DOI] [PubMed] [Google Scholar]
  48. Woo M., Kim Y. Inter- and Intrahemispheric EEG coherence and visuomotor performance during shooting competition and practice. Percept. Mot. Skills. 2017;124(4):830–845. doi: 10.1177/0031512517709150. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary material
mmc4.docx (73.2KB, docx)

Articles from Clinical Neurophysiology Practice are provided here courtesy of Elsevier

RESOURCES