Brain Oscillations in Sport: Toward EEG Biomarkers of Performance

Abstract

Brain dynamics is at the basis of top performance accomplishment in sports. The search for neural biomarkers of performance remains a challenge in movement science and sport psychology. The non-invasive nature of high-density electroencephalography (EEG) recording has made it a most promising avenue for providing quantitative feedback to practitioners and coaches. Here, we review the current relevance of the main types of EEG oscillations in order to trace a perspective for future practical applications of EEG and event-related potentials (ERP) in sport. In this context, the hypotheses of unified brain rhythms and continuity between wake and sleep states should provide a functional template for EEG biomarkers in sport. The oscillations in the thalamo-cortical and hippocampal circuitry including the physiology of the place cells and the grid cells provide a frame of reference for the analysis of delta, theta, beta, alpha (incl.mu), and gamma oscillations recorded in the space field of human performance. Based on recent neuronal models facilitating the distinction between the different dynamic regimes (selective gating and binding) in these different oscillations we suggest an integrated approach articulating together the classical biomechanical factors (3D movements and EMG) and the high-density EEG and ERP signals to allow finer mathematical analysis to optimize sport performance, such as microstates, coherency/directionality analysis and neural generators.

Keywords: EEG; biomarkers; brain rythms; cortical activity; sport.

Figure 1

Illustration of an integrated neurophysiological approach in sport. The sleep push or drag flicking in field hockey sport is here illustrated as an example of complex movement involving the whole body. It is an attacking movement mainly performed within the penalty corner but also realized outside the penalty corner situation. Here in our experimental setup, the player was placed in front of the laboratory wall with the field of action in its back. He was asked to relax in this position before turning in front of the action field. Then he performed the sleep push movement at its self-paced mode in order to put the ball inside of the goal. (A) The player was equipped with a 128 electrodes cap (ANT), the cables of the cap were connected to the amplifier which were placed in a traveling box piloted by the experimenter in order to assume the follow-up of the player's running. (B) Surfaces electrodes () were placed on muscles of interest and allowed the electromyographic (EMG) recording (see, G). (C) Retro-reflective passive markers were fixed on the skin overlying the following bony landmarks on both sides of the body: LM, lateral malleolus; MT, fifth metatarsal head; LE, lateral condyle of the knee; GT, greater trochanter; and IL, tubercule of the anterosuperior iliac crest then a stick diagram of the whole body representation was constructed. (D) Kinematics of the sleep push movement, successive pictures of the main phase are presented and 3 markers placed on the head and 3 other placed on the distal part of the stick are joined by continuous lines (blue) giving a global representation of the movement signature.

Figure 2

(A) Recording configuration (simultaneous EEG, LFP, and whole-cell recordings) in the mouse somatosensory cortex (Barrel cortex) and reconstruction of one excitatory pyramidal neuron (black) of the layer 2/3. (B) From top to bottom, movement of the whisker (green) defining quiet and whisking periods, EEG signals (red), LFP recording (blue), and membrane potentials (black). (C) Reconstruction of two layer 2/3 pyramidal neurons (blue and red) simultaneously recorded. (D) From top to bottom, whisking movement (green) and superimposition of the membrane potentials of the two pyramidal neurons with their cross-correlation during quiet and whisking movement (modified from Poulet and Petersen, , with permission). (E) Simultaneous recording of a thalamic neuron and cortical LFP showing the increase of the action potential firing rates during the active desynchronized cortical state (a). Behavioral difference of the membrane potential of a pyramidal neuron in presence (b), or in absence of the thalamic influence (c) during quiet and whisking states (see text for more details, modified from Poulet et al., , with permission).

Figure 3

(A) Top traces: representative 16 s electroencephalogram (EEG) traces from the Fp1 channel during early sleep. (B) Average power spectra in non-rapid eye movement (NREM) sleep during episodes 1 and 2 (early sleep, black) and episodes 3 and 4 (late sleep, gray) for Fp1 channel (mean ± SEM, n = 7). Triangles indicate significant bins based on SnPM (P < 0.05, single threshold corrected). (C) Slow-wave activity (SWA; 0.5–4.0 Hz) profile in NREM sleep during the night for an individual subject (average 1-min values, % of the mean of 4 NREM episodes) rapid eye movement (REM) episodes are indicated by hatched areas. Early and late sleep (including REM episodes) are color-coded in black and gray, respectively. Note that 69% of the sleep time corresponds to NREM, 23% of REM and the rest of time refers to waking after sleep onset. In addition, about 17% of the NREM correspond to the SWS (adapted from Riedner et al., , with permission).

Figure 4

SPW-R is composed by a large and sharp amplitude depolarization (SPW) of about 40-100 ms recorded in the hippocampus CA1 stratum radiatum and most often associated by brief and fast LFP oscillation (~110–200 Hz) known as “ripples” (R) in the CA1 pyramidal layer. The SPW-R represents one of the most synchronous network patterns implicating synchronicity of 10–20% of hippocampus neurons (Chrobak and Buzsáki, 1994). (A) Fast field oscillation in the CA1 region of the dorsal hippocampus. Simultaneous recordings from the CA1 pyramidal layer [electrode 1, wide-band recording (1 Hz–10 kHz)] and stratum radiatum (electrode 2, unit activity 500 Hz–10 kHz and fast field oscillation (100–400 Hz). The second and third traces are digitally filtered derivatives of the wide-band trace (1). Note simultaneous occurrence of fast field oscillations, unit discharges, and sharp amplitude depolarization (electrode 2). Calibrations: 0.5 mV (trace 1), 0.25 mV (traces 2 and 3), and 1.0 mV (trace 4) (from Buzsáki et al., , with permission). (B) Depth profile of averaged sharp-wave ripples (SPW-R) in a freely moving mouse (n = 961 events; vertical site separation: 100 mm). Voltage traces (light gray) are superimposed on current-source density (CSD) map. Black trace: site of maximum amplitude ripple; heavy gray trace: site of maximum amplitude SPW (adapted from Stark et al., , with permission). (C) Depth Profile of theta and gamma oscillations in the rat recorded during exploration (left) with a 16-site silicon probe introduced in the CA1-dentate gyrus axis. Numbers indicate recording sites (100 μm spacing). Note the gradual shift of theta phase from str. oriens to str. lacunosum-moleculare (right) and the emergence of gamma waves superimposed on theta oscillation mainly in the granule cell layer and the hilus. Vertical bar: 1 mV (from Bragin et al., , with permission). (D) Schematic representation of the brain showing location of γ oscillations in different cortical areas (i–iv) and Θ oscillation in the hippocampus (HI) entorhinal cortex (EC). These brain oscillations can influence each other within and across networks by modulating the phase and/or the amplitude of the oscillations (modified from Buzsáki and Watson, , with permission). Abbreviations: p or pyr, pyramidal layer; lm, str. lacunosum-moleculare; o, str. oriens; r, str. radiatum; lm, str. lacunosum-moleculare; g, granule cell layer; h, hilus.

Figure 5

(A) Place cell in the hippocampus (adapted from Moser et al., , with permission) and (B) grid cell in the medial entorhinal cortex. The locations where the neurons discharge (red) are superimposed on the rat's trajectory (black) in the recording enclosure black. Whereas, most of the place cells have a single firing location, the of a grid cell form a periodic triangular matrix of the firing fields covering the entire environment available to the animal. (C–E) Theta oscillations revealed by intracranial EEG and MEG in the human. (C) Navigation in virtual reality-rendered multiple T-junction mazes in which key presses move the subject in the VR maze (F, forward; R, right; and L, left). (D) Sample intracranial EEG recorded in the inferior frontal gyrus (upper trace); theta oscillation is emphasizing in an enlarged view of the boxed region (lower trace). The average power in iEEG for one trial through the VR maze during 21 s is represented by the power frequency plot (green) showing a peak in the theta band (4–12 Hz) (adapted from Kahana et al., , with permission). (E) Movement-related MEG time-frequency effects and related fMRI effects. Plots show MEG signal as baseline corrected log normalized difference scores (dB, z axis) against time (x-axis, seconds) and frequency (y-axis, Hz), averaged across 18 participants. The effect of movement initiation during navigation (left panel) is showed at an exemplary single sensor of interest, (posterior right middle temporal location). Coronal slices showing right hippocampal activation for movement initiation compared to stationary periods (middle panel). Percent signal change at right hippocampal peak-voxel averaged across 14 participants for movement and stationary periods (mean 6 SEM) (right panel) (adapted from Kaplan et al., , with permission).

Figure 6

(A) Raw EEG recordings during the arrest reaction. Fourteen EEG channels referenced to linked mastoid (from F7 top) to O2 (bottom). The white arrow indexes to the order to open the eyes. The black arrow points to the onset of the eye movement artifact related to eye opening, mainly recorded by frontal electrodes (F7, F3, FZ, F4, F8). Note that the amplitude of mu rhythm [recorded by central electrodes (C3, CZ, C4)] is reduced before eye opening (red arrow “1”) whereas alpha rhythm (P3, P4, PZ, O1, O2) is only reduced after this movement when the eyes are closed. Event-related spectral perturbation (ERSP) time-locked to the order of eyes opening. (B) ERSP-image plot of the color-coded single trials of the EEG spectrum recorded in P3 channel in one cosmonaut on Earth before the flight (a), in weightlessness (b), and on Earth after the flight (c). The trials are time-locked to the order to open the eyes (stripped line). Blue color indicates a decrease in power. (C) ERSP-image plot for the same electrode, condition and participant as in (B), but time-locked to the order of eyes closing. Red color indicates an increase in power (modified from Cheron et al., , with permission).

Figure 7

Schematic representation of the network models of Dipoppa and Gutkin (2013) dedicated to a working memory task. (A) Two neuronal populations B (in blue) and R (in red) of recurrent network receiving input by sources modulated by the shared background oscillation. The phases of operations are outlined in a white box showing the succession of the gating modes and operations: (I) Gate-in mode: the sample stimulus (blue arrow) activates population B (load). (II) Selective-gating: the distractor stimulus (red arrow) is not able to activate population R persistently (block distractor) and the memory in population B is held (maintain). (III) Gate-out mode: upon match-stimulus presentation, persistent activity is deactivated in the blue population (clear). Input oscillations enabling the gating modes: beta–gamma band ensures the gate-in mode at the beginning of the task, theta band ensures the selective-gating mode during the delay period (memory maintenance together with protections from the distractors), and alpha band ensures the gate-out mode at the task completion (memory is rapidly cleared). (B–D) Adaptive suggestion of this model into sport context (passing ball in soccer) (see main text for more details). (E–G) Frequency-dependent influence of oscillations on the reverberant spiking persistent state. The network responses (black) to a transient excitatory external stimulus (t = 50–150 ms) (blue bar and signal) depend on the frequency content of the background oscillatory input. First and third rows show the raster histogram of 30 representative neurons; second and fourth rows show average population input from recurrent connections (black), background activity (red), and external stimulus (blue) in arbitrary/normalized units. For each frequency the background oscillation is switched on either after the stimulus presentation (first and second rows) or before (third and fourth rows). (E) Beta–gamma-band oscillations (45 Hz) are compatible with persistent state maintenance: neither erasing nor blocking is seen. (F) Theta-band oscillations (6.5 Hz) maintain an a priori persistent state while blocking de novo activations. (G) Alpha-band oscillations (here 10 Hz) inhibit reverberant activity: the persistent state is deactivated by oscillations onset and is prevented from being activated by the transient stimulus (adapted from Dipoppa and Gutkin, , with permission).