Next Article in Journal
A Risk Evaluation Framework in System Control Subject to Sensor Degradation and Failure
Previous Article in Journal
A Review of Machine Learning Approaches for the Personalization of Amplification in Hearing Aids
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Sensors
Volume 24
Issue 5
10.3390/s24051548
sensors-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Auditory Cue Effects on Gait-Phase-Dependent Electroencephalogram (EEG) Modulations during Overground and Treadmill Walking

by
Kittichai Tharawadeepimuk
1,
Weerawat Limroongreungrat
1,
Metaneeya Pilanthananond
2 and
Ampika Nanbancha
1,*
1
College of Sports Science and Technology, Mahidol University, Nakhon Pathom 73170, Thailand
2
Department of Physiology, Faculty of Science, Mahidol University, Bangkok 10400, Thailand
*
Author to whom correspondence should be addressed.
Sensors 2024, 24(5), 1548; https://doi.org/10.3390/s24051548
Submission received: 12 December 2023 / Revised: 2 February 2024 / Accepted: 21 February 2024 / Published: 28 February 2024
(This article belongs to the Section Biomedical Sensors)
Browse Figures
Versions Notes

Abstract

:
Walking rehabilitation following injury or disease involves voluntary gait modification, yet the specific brain signals underlying this process remains unclear. This aim of this study was to investigate the impact of an auditory cue on changes in brain activity when walking overground (O) and on a treadmill (T) using an electroencephalogram (EEG) with a 32-electrode montage. Employing a between-group repeated-measures design, 24 participants (age: 25.7 ± 3.8 years) were randomly allocated to either an O (n = 12) or T (n = 12) group to complete two walking conditions (self-selected speed control (sSC) and speed control (SC)). The differences in brain activities during the gait cycle were investigated using statistical non-parametric mapping (SnPM). The addition of an auditory cue did not modify cortical activity in any brain area during the gait cycle when walking overground (all p > 0.05). However, significant differences in EEG activity were observed in the delta frequency band (0.5–4 Hz) within the sSC condition between the O and T groups. These differences occurred at the central frontal (loading phase) and frontocentral (mid stance phase) brain areas (p < 0.05). Our data suggest auditory cueing has little impact on modifying cortical activity during overground walking. This may have practical implications in neuroprosthesis development for walking rehabilitation, sports performance optimization, and overall human quality-of-life improvement.

Graphical Abstract

1. Introduction

Ambulation or walking is a fundamental motor behavior requiring intricate coordination between skeletal muscles and neural pathways across the brain [1,2,3]. While the subcortical brain traditionally governs basic locomotion, recent studies highlight the significant role of the cerebral cortex in this process [4,5,6]. Patients with conditions impairing walking abilities, such as stroke, Parkinson’s disease, and Huntington’s disease, often benefit from rehabilitation techniques aimed at improving ambulation [6]. Auditory cues have emerged as a promising tool in walking rehabilitation, offering various benefits including improved physical fitness, posture, stress reduction, and navigation [7,8]. These cues can range from exercise technique instruction, motivational prompts, entertainment features, and safety alerts to guide walking behavior. Notably, auditory cueing has been linked to enhance cortical activity during walking recovery, with increased engagement observed across various cortical regions [9,10].
Previous research has demonstrated that rhythmic auditory cueing leads to increased cortical activity during walking recovery [11]. In a study involving stroke patients, the integration of visual and sensorimotor information across various cortical regions was observed [10]. Researchers used functional near-infrared spectroscopy (fNIRS) to measure changes in oxygenated hemoglobin concentrations in different brain regions. However, due to fNIRS’s limited spatial resolution and challenges in signal interpretation, it is unsuitable for examining changes in cortical activation across different phases of the gait cycle. In contrast, electroencephalography (EEG) directly records neuronal electrical activity, offering superior spatial resolution. Moreover, EEG’s millisecond-level temporal resolution permits detailed information about the timing of neuronal events [12]. Recent EEG research has examined the fundamental aspects of brain activity during various ambulatory conditions [13]. For example, in stroke survivors undergoing robot-assisted gait therapy, the EEG delta frequency band (0.5–4 Hz) has been associated with sensorimotor information, deep sleep states, and the execution of movement [14,15,16,17,18]. While cortical modulations are known to be linked to the gait cycle phase, individual components such as the swing and stance phases are not commonly investigated [19,20,21,22].
In rehabilitation settings, individuals may undergo ambulation either overground or on a treadmill, adjusting their pace based on self-selection or following predetermined cues like a metronome. Currently, only one study [23] has compared EEG activity between these two surfaces during the gait cycle, finding increased cortical activity in the theta frequency band with self-selected speeds on both surfaces. However, it remains unclear whether overground walking with rhymical auditory cues, serving as an anticipatory and external frame reference [24], results in different EEG brain activity compared to treadmill walking at a constant speed with minimal engagement [25]. The potential variations in sensorimotor-related information, motor intention, and cognitive demands may influence cortical activity across different walking conditions. Therefore, investigating EEG amplitude modulation of sensorimotor processing during varied walking tasks could offer novel rehabilitation interventions to effectively modulate gait-related brain activity. Integrated EEG analysis from gait studies may contribute methodologically to establishing new standards for studying the neural correlates of locomotion.
Accordingly, we aimed to harness the technological advances offered by EEG amplitude analysis to investigate brain activity modulation during overground (O) and treadmill (T) walking across the gait cycle. We specifically investigated two speed control conditions: self-selected speed control (sSC) and speed control (SC), representing walking without and with an auditory guidance, respectively. Furthermore, instead of pooling electrodes, we utilized EEG-normalized amplitude for individual electrodes to improve spatial resolution [26], enabling a more detailed analysis of brain activity during walking [17]. We hypothesized that distinct differences would emerge within and between groups in the EEG delta frequency band, reflecting the impact of auditory cued walking versus normal walking.

2. Materials and Methods

2.1. Participants

Twenty-four healthy volunteers (12 males and 12 females; age: 25.7 ± 3.8 years; height: 166.7 ± 7.5 cm; body mass: 64.6 ± 14.2 kg) participated in this study. All participants were right-leg and right-arm dominant. Participant inclusion criteria were: (1) normal joint alignment and gait pattern, (2) Q-angle within 10°–20°, and (3) leg length discrepancy not exceeding 2 cm [27]. Exclusion criteria were (1) knee joint deformity, (2) feet supination or overpronation, (3) pelvic side shift, (4) recurrent or recent injuries of the lower extremities, (5) musculoskeletal problem on either leg within the past three months, and (6) history of neurological disease or vestibular/visual disturbance. The mini-mental state examination (MMSE) and the participant’s medical history were evaluated prior to recruitment. All participants provided written informed consent prior to their participation in this study. The experiments were approved and performed in accordance with the research ethics committee of Mahidol University—Central Institutional Review Board (COA MU-CIRB 2022/060.2005).
This study used a pseudonym for recording the wearable sensor data, thereby anonymizing the identity of the participants. In accordance with informed consent, the data were collected for non-commercial research purposes. The dataset is available at https://osf.io/2R9BA/, accessed on 9 January 2024. Any use or sharing of the data, as well as any derived use, must strictly adhere to the same usage guidelines (for non-commercial research purposes, without attempts at identification, and with limited disclosure). Violations of these usage regulations will be subject to legal consequences for the individual users, as the dataset publisher bears no responsibility.

2.2. Experimental Setup and Design

Employing a between-subject repeated-measures design, 24 participants were randomly allocated into either an O (n = 12, 6 males, 6 females) or T (n = 12, 6 males, 6 females) walking group. The participants in both groups were asked to perform the same two walking conditions (sSC, SC). The sSC condition acted as the control condition within each group (i.e., non-auditory cue), and required participants not to exceed a walking speed of 0.8 m/s (2.88 km/h), as previous research has indicated movement artifacts become more pronounced beyond this threshold [28]. In the SC condition, the participants were asked to walk in time with a metronome (auditory cue) set at 0.55 m/s (1.98 km/h). The participants in the O group were instructed to walk directly forwards 100 m and not to turn around. Each walking sequence corresponded to walking for a duration no greater than 3 min. This approach can minimize EEG artifacts during walking [17,28,29,30]. The test instruments, tasks, and experimental procedures were explained and demonstrated to the participants prior to data collection.

2.3. Data Collection

EEG signals were recorded during walking using a portable EEG system (eego™mylab; ANT Neuro) with a 32-electrode EEG cap. The EEG electrode positions included Fp1, Fp2, Fpz, F3, F4, F7, F8, Fz, FC1, FC2, FC5, FC6, C3, C4, Cz, T7, T8, CP1, CP2, CP5, CP6, P3, P4, P7, P8, Pz, POz, O1, O2, Oz, M1, and M2, along with the reference electrode (REF = CPz) and the ground electrode (AFz), as shown in Figure 1. The electrode impedances were maintained below 20 kΩ, as recommended [28]. The EEG data were sampled at 512 Hz. In addition, inertial measurement unit (IMU) signals were recorded using a wireless IMU system (Noraxon; Ultium Motion) with 7 sensors placed on the sacrum, L|R thigh, L|R shank, and L|R second metatarsal (MyoMotion ISB-compliant modeling). The IMU data were sampled at 200 Hz, and smoothed via a Kalman filter approach [31]. EEG and IMU data of each participant were recorded for 3 min together in the sSC and SC conditions.

2.4. EEG Analysis

All data were analyzed using custom-written routines programs in MATLAB (v2020b, MathWorks, Natick, MA, USA). The raw EEGs were bandpass-filtered (within the range of 0.5–100 Hz for delta band, 4.5–100 Hz for theta band, 8–100 Hz for alpha band, and 13–100 Hz for beta band, using a fourth-order Butterworth filter), and notch-filtered (at 50 Hz, using a fourth-order Butterworth filter). These two filters were used for preparation of EEG data before using ICA algorithms, which greatly improved the quality of artifact separation [32,33]. Next, to identify brain and non-brain sources mixing in the EEGs, the preprocessed EEG data were decomposed using ICA using a “fastICA” algorithm. The procedural sequence in the fastICA algorithm encompassed the subsequent steps: centering (remean), decorrelation (covariance matrix), principal component analysis: PCA (pcamat), normalization (whitening), and ICA (fpica). Using the fastICA algorithm, five ICs were estimated. This process can remove the artifact ICs, which included two ICs originating from ocular, one IC from glossokinetic, and two ICs from muscular activity during walking-related movement [33,34]. Then, the EEG data were low-pass-filtered (using 4 Hz cutoff for delta band, 7.5 Hz cutoff for theta band, 12 Hz cutoff for alpha band, and 30 Hz cutoff for beta band, using a fourth-order Butterworth filter).
To analyze EEG signals within each gait cycle, IMU data were used to identify initial contact, defined as the maximal right thigh angle (Noraxon MyoMotion-Segments Thigh RT-orientation-y (Figure 2). EEG and IMU signals were collected simultaneously and synchronized with digital markers. In the IMU signal, each gait cycle was defined as the time elapsed between two subsequent initial contact points and divided into the stance phase and swing phase using the minimum right thigh angle. The duration of each subphase was then expressed as a percentage of the gait cycle [35,36]. EEG data corresponding to each gait cycle were subsequently selected from the locations corresponding to the initial phase. All within-stride EEG data were then amplitude-normalized to the local peak value and time-normalized to 100 data points. Additionally, the data were ensemble-averaged across gait cycles for each participant. The processed EEG data yielded a matrix of 30 rows, corresponding to the 30 active EEG electrodes, and 100 columns corresponding to the 100 time points (percentage of gait cycle).
Furthermore, the EEG data of each active electrode were ensemble-averaged across participants for each group and walking condition (TsSC: treadmill walking with self-selected speed control, TSC: treadmill walking with speed control, OsSC: overground walking with self-selected speed control, and OSC: overground walking with speed control) to represent typical traces at each location. These data were used to display the direction and magnitude of the changes in gait-phase-related brain activities, which are represented in a colormap using an arbitrary unit (a.u.) scale. The average and standard deviation (SD) of amplitudes of the EEG data were used to calculate amplitude differences among groups and walking conditions.

2.5. Statistical Analysis

All statistical tests were performed in MATLAB. To assess the normality of the data, Shapiro–Wilk tests were initially applied, revealing an abnormal distribution of the data. Subsequently, statistical non-parametric mapping (SnPM) analysis was employed for all time-series EEG electrode data. We were primarily interested in comparing group x condition interaction effects, specifically the within-group interaction effects (OSC vs. OsSC and TSC vs. TsSC) to compare walking with and without an auditory cue on EEG brain activity across the gait cycle. Moreover, we were interested in comparing the between-group effects (OSC vs. TSC) to examine the impact of the walking surface on EEG brain activity when employing the auditory cue. Therefore, a two-way between-subject (group: O, T) analysis of variance (ANOVA) with repeated measures (condition: sSC and SC) was conducted to explore differences in EEG amplitude within each frequency band as part of the SnPM analysis. Where significant main interaction effects were identified, a post hoc test with a Bonferroni correction was applied. Only the results demonstrating statistical significance or important interactions are presented in this study. All data are reported as means ± standard deviation (SD) unless otherwise stated. The significance level was set at p < 0.05.

3. Results

SnPM analysis revealed differences in EEG time series at several brain locations in the amplitude in the EEG delta frequency band (AmEEGDel; 0.5–4 Hz). A significant interaction effect was found in EEG time-series data at the Fz position (p = 0.01, F* = 9.953), with a significant increase in EEG magnitude between TsSC and OsSC (p = 0.039; Figure 3a) at 3.7% to 7.2% of the gait cycle, corresponding to the loading response phase. A significant interaction was also found at the FC1 position (p = 0.01, F* = 9.135), with a significant increase in EEG magnitude between TsSC and OsSC (p = 0.039; Figure 3b) at 28.2% to 30.8% of the gait cycle, corresponding to the mid stance phase. However, no within-group differences in EEG time-series activity were found across brain locations between OSC and OsSC (all p > 0.05) or TSC and TsSC (all p > 0.05) conditions during the gait cycle. Similarly, there was no significant between-group difference in EEG time-series activity observed across brain regions in the OSC and TSC conditions across the gait cycle (all p > 0.05). The number of steps performed over the 3 min walking task within each group and condition are displayed in Table 1.

4. Discussion

In this study, we examined how an auditory cue influences brain activity during overground and treadmill walking, using EEG time-series data. Our main findings are as follows: (1) no significant differences were observed across any brain region between the O and T groups in the SC condition throughout the gait cycle, (2) there were no significant differences across brain regions between the sSC and SC conditions within the O and T groups throughout the gait cycle, and (3) a significant difference was found in the sSC conditions between the O and T groups. Specifically, there was a significant difference in AmEEGDel at the central frontal position during the loading response phase and the frontocentral position during the mid stance phase of the gait cycle. These findings may have implications for rehabilitation practitioners, especially when tailoring interventions for patients undergoing gait rehabilitation on different walking surfaces.

4.1. Between-Group Differences in EEG Activity across the Gait Cycle

In our study, we observed similar brain activity in the SC condition between the O and T groups across various brain frequency bands during the gait cycle. Previous research on the impact of auditory cueing on cortical activity during walking on different surfaces is limited. However, contrary to our observations, Roeder et al. [23] reported a significant increase in the magnitude of cortical activity in the theta frequency band during the double-support (mid stance) phase when walking overground compared to on a treadmill at participants’ preferred speed (0.92–1.33 m/s or 3.3–4.8 km/h). Without the influence of an auditory cue, Roeder et al.’s [23] findings align with the expectation that differences in kinematics, kinetics and electromyographic (EMG) activity between these two surfaces [37] independently modulate brain activity. In our study, it is challenging to explain the lack of a marked difference in cortical activity in the SC condition between overground (auditory cue) and treadmill walking (non-auditory cue). However, it may be speculated that the task of stepping overground in synchronization with the auditory cue at the speed used in our study (0.55 m/s, 2.88 km/h) provided too little mental engagement to lead to a noticeable increase in cortical activity. Alternatively, increased cortical activity may have been present with the auditory cue, but potentially masked by participants also having increased cortical activity when walking on a treadmill.
In contrast to the similar brain activity observed in the SC condition between groups, we found a marked difference in brain activity in the sSC condition. Specifically, the central frontal brain area in the delta frequency band exhibited a significantly increased activity in TsSC compared with OsSC during the loading response phase (3.7–7.2%; Figure 3a) of the gait cycle. Moreover, the frontocentral position in the delta frequency band was significantly increased in TsSC compared with OsSC at the mid stance phase (28.2–30.8%; Figure 3b) of the gait cycle. It is understood that the frontocentral area plays an important role in brain processing during walking intention, particularly in gait initiation and the generation of movement-related cortical potentials during stepping [38]. This has previously been demonstrated during a walking task using a brain–computer interface (BCI) requiring participants to control an avatar’s walking [39]. The central frontal cortex is interconnected with cortical motor areas in humans [40] and plays a pivotal role in movement-related motor cortex and visuomotor processing [41]. Taken together, the increased brain activation in both the frontocentral and central frontal cortex in the delta frequency band suggests that treadmill walking results in a greater cognitive load compared with overground walking at similar self-selected speeds. In TsSC, brain function in the frontal area may require precise motor planning and timing adjustments to synchronize with the belt’s motion. It has previously been proposed that walking at a fixed speed on a treadmill is monotonous and requires little mental engagement for motor learning [25]. However, our current data suggest that treadmill walking actually requires greater mental engagement than overground walking (at similar speed) during the gait phase. Nevertheless, this finding may simply be a result of participant unfamiliarity with walking on a treadmill.

4.2. Within-Group Differences in EEG Activity across the Gait Cycle

Another key finding from our study indicated consistent EEG activity across various brain regions during the gait cycle, irrespective of whether participants were in the sSC or SC condition. In the O group, our results suggest that the introduction of an auditory cue during overground walking does not alter brain activity across the gait cycle compared to walking at a similar self-selected walking speed (for the same duration). This finding contradicts previous research suggesting heightened brain activity when processing auditory and sensorimotor-related information [28,42]. For instance, past studies have linked the left temporal cortex to responses to auditory cues, such as synchronizing finger-tapping with a metronome [43]. This brain area typically plays a role in receiving and interpreting auditory cues related to a desired walking speed. Earlier investigations also noted changes in amplitude in the parieto-occipital region during overground walking, particularly when participants simultaneously engaged in visual discrimination tasks using virtual reality [44] or precision stepping [28]. However, our findings do not align with the idea of increased utilization of visual and sensory processing resources in the central parieto-occipital area, which is conventionally thought to be actively involved in processing visual information and responding to external stimuli [45]. Given the absence of differences between the OsSC and OSC conditions, our results challenge the notion that the convergence of visual, tactile and auditory information occurs in the parietal regions with auditory cueing during overground walking. This convergence is commonly recognized as a key sensorimotor interface facilitating multisensory integration during ongoing movements [46,47]. The discrepancy between our findings and previous research questions the extent to which auditory cues significantly impact brain activity during overground walking.
Similarly, we found no significant difference in brain activity across the gait cycle between the sSC and SC conditions in the T group. In the SC condition, participants walked on the treadmill at a fixed speed (0.55 m/s, 2.88 km/h), which matched the speed of overground walking accompanied by an auditory cue. This allowed for a direct comparison between groups in the SC condition. In contrast, in the sSC condition, participants walked at a self-selected speed not exceeding 0.8 m/s (2.88 km/h). Despite the same walking duration (3 min) and brain activity in both the TSC and TsSC conditions (and in the O group), a discrepancy in step frequency was noted (Table 1). This could potentially complicate the interpretation of brain activity derived from EEG time-series data when making comparisons between conditions. Although this may be considered a study limitation, we argue that permitting participants to select their walking speed offers a more ecologically valid comparison, especially considering the limited alternatives for controlling walking speed without the use of auditory or visual cues.

4.3. Methodological Considerations

In this study, a significant methodological consideration concerns the potential influence of movement artifacts in the EEG recording. To mitigate this effect, we controlled walking speed at 0.55 m/s (1.98 km/h) in the SC condition, consistent with previous studies [29,30]. Additionally, participants were instructed not to exceed a walking speed of 0.8 m/s (2.88 km/h) in the sSC condition, as previous research has indicated that movement artifacts become more pronounced beyond this threshold [28]. Furthermore, ICA was employed to remove artifacts from EEG data obtained during movement. ICA is a well-established and widely utilized artifact removal method for EEG data during walking and movement tasks [20,21,33,34,48,49,50,51,52,53,54,55]. The fast and highly efficient fixed-point fastICA algorithm was specifically utilized in this study to separate original signals from artifacts present in contaminated EEG signals. The FastICA algorithm effectively addresses ocular originated artifacts and glossokinetic artifacts, respectively [33,34,52]. However, it is important to acknowledge that ICA typically requires a large amount of data for accurate component estimation. Consequently, the optimization process can be computationally intensive and time-consuming. This limitation warrants consideration, particularly in the context of clinical applications and the utilization of BCI.
Several methodological considerations could potentially offer additional insights into our data. While this study employed slow walking speeds (0.55 m/s and <0.8 m/s) to minimize movement artifacts in EEG recordings, it is acknowledged that these walking speeds may not fully represent muscle activity or natural gait speeds. Future studies should explore cortical activity during more natural gait speeds, employing EEG recordings less susceptible to movement artifacts and utilizing more sophisticated artifact removal techniques.

5. Conclusions

Our EEG data suggest a minimal impact of employing an auditory cue during overground walking to increase cortical processing, as we did not observe an increase in brain activity compared to walking without an auditory cue. Furthermore, treadmill walking at a self-selected speed appears to elicit heightened brain activity compared with walking overground at a similar speed, suggesting that treadmill walking requires enhanced motor coordination and synchronization under these conditions. These findings provide insights with potential implications ranging from walking rehabilitation and the development of volitional control systems for neuroprostheses to enhancing complex movements in sports performance.

Author Contributions

K.T. conceptualized and designed the study, submitted the project to the ethics committee, contributed to data collection, performed formal data analysis, interpreted the findings, and drafted and finalized the manuscript; A.N. conceptualized and designed the study, recruited participants, contributed to data collection, performed data analysis, interpreted the findings, and approved the final version of the manuscript. W.L. and M.P. participated in the study design and contributed to the final version of the manuscript. All authors have read and agreed to the published version of the manuscript.

Funding

This research (Grant No. RGNS 64-142) was funded by the Office of the Permanent Secretary, Ministry of Higher Education, Science, Research and Innovation (OPS MHESI), Thailand Science Research and Innovation (TSRI).

Institutional Review Board Statement

The study was conducted in accordance with the Declaration of Helsinki, and approved by the research ethics committee of Mahidol University—Central Institutional Review Board (protocol code COA No. MU-CIRB 2022/060.2005 and approved on 20 May 2022).

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study.

Data Availability Statement

The data that support the findings of this study are available at https://osf.io/2R9BA/ (accessed on 9 January 2024).

Acknowledgments

We thank Christopher Mawhinney and Frederic Stucky for their suggestions in the preparation of this manuscript.

Conflicts of Interest

The authors have no conflict of interests to report.

References

  1. Hamacher, D.; Herold, F.; Wiegel, P.; Hamacher, D.; Schega, L. Brain activity during walking: A systematic review. Neurosci. Biobehav. Rev. 2015, 57, 310–327. [Google Scholar] [CrossRef]
  2. Leisman, G.; Moustafa, A.A.; Shafir, T. Thinking, Walking, Talking: Integratory Motor and Cognitive Brain Function. Front. Public Health 2016, 4, 94. [Google Scholar] [CrossRef] [PubMed]
  3. Yang, J.F.; Gorassini, M. Spinal and brain control of human walking: Implications for retraining of walking. Neuroscientist 2006, 12, 379–389. [Google Scholar] [CrossRef] [PubMed]
  4. Drew, T.; Prentice, S.; Schepens, B. Cortical and brainstem control of locomotion. Prog. Brain Res. 2004, 143, 251–261. [Google Scholar] [CrossRef] [PubMed]
  5. Fortier-Lebel, N.; Nakajima, T.; Yahiaoui, N.; Drew, T. Microstimulation of the Premotor Cortex of the Cat Produces Phase-Dependent Changes in Locomotor Activity. Cereb. Cortex 2021, 31, 5411–5434. [Google Scholar] [CrossRef] [PubMed]
  6. Nishimaru, H.; Matsumoto, J.; Setogawa, T.; Nishijo, H. Neuronal structures controlling locomotor behavior during active and inactive motor states. Neurosci. Res. 2023, 189, 83–93. [Google Scholar] [CrossRef]
  7. Vitorio, R.; Stuart, S.; Gobbi, L.T.B.; Rochester, L.; Alcock, L.; Pantall, A. Reduced Gait Variability and Enhanced Brain Activity in Older Adults with Auditory Cues: A Functional Near-Infrared Spectroscopy Study. Neurorehabil. Neural Repair 2018, 32, 976–987. [Google Scholar] [CrossRef]
  8. Ghai, S.; Ghai, I. Effects of Rhythmic Auditory Cueing in Gait Rehabilitation for Multiple Sclerosis: A Mini Systematic Review and Meta-Analysis. Front. Neurol. 2018, 9, 386. [Google Scholar] [CrossRef] [PubMed]
  9. Almarwani, M.; Van Swearingen, J.M.; Perera, S.; Sparto, P.J.; Brach, J.S. The Effect of Auditory Cueing on the Spatial and Temporal Gait Coordination in Healthy Adults. J. Mot. Behav. 2019, 51, 25–31. [Google Scholar] [CrossRef]
  10. Lim, S.B.; Yang, C.L.; Peters, S.; Liu-Ambrose, T.; Boyd, L.A.; Eng, J.J. Phase-dependent Brain Activation of the Frontal and Parietal Regions during Walking After Stroke—An fNIRS Study. Front. Neurol. 2022, 13, 904722. [Google Scholar] [CrossRef]
  11. Schaefer, R.S. Auditory rhythmic cueing in movement rehabilitation: Findings and possible mechanisms. Philos. Trans. R. Soc. Lond. B Biol. Sci. 2014, 369, 20130402. [Google Scholar] [CrossRef]
  12. Travis, F. Temporal and spatial characteristics of meditation EEG. Psychol. Trauma Theory Res. Pract. Policy 2020, 12, 111–115. [Google Scholar] [CrossRef]
  13. Oliveira, A.S.; Arguissain, F.G.; Andersen, O.K. Cognitive processing for step precision increases beta and gamma band modulation during overground walking. Brain Topogr. 2018, 31, 661–671. [Google Scholar] [CrossRef]
  14. Nakanishi, Y.; Yanagisawa, T.; Shin, D.; Kambara, H.; Yoshimura, N.; Tanaka, M.; Fukuma, R.; Kishima, H.; Hirata, M.; Koike, Y. Mapping ECoG channel contributions to trajectory and muscle activity prediction in human sensorimotor cortex. Sci. Rep. 2017, 7, 45486. [Google Scholar] [CrossRef]
  15. Presacco, A.; Forrester, L.W.; Contreras-Vidal, J.L. Decoding intra-limb and inter-limb kinematics during treadmill walking from scalp electroencephalographic (EEG) signals. IEEE Trans. Neural Syst. Rehabil. Eng. 2012, 20, 212–219. [Google Scholar] [CrossRef]
  16. Presacco, A.; Goodman, R.; Forrester, L.; Contreras-Vidal, J.L. Neural decoding of treadmill walking from noninvasive electroencephalographic signals. J. Neurophysiol. 2011, 106, 1875–1887. [Google Scholar] [CrossRef] [PubMed]
  17. Yokoyama, H.; Kaneko, N.; Ogawa, T.; Kawashima, N.; Watanabe, K.; Nakazawa, K. Cortical correlates of locomotor muscle synergy activation in humans: An electroencephalographic decoding study. iScience 2019, 15, 623–639. [Google Scholar] [CrossRef] [PubMed]
  18. Bulea, T.C.; Prasad, S.; Kilicarslan, A.; Contreras-Vidal, J.L. Sitting and standing intention can be decoded from scalp EEG recorded prior to movement execution. Front. Neurosci. 2014, 8, 376. [Google Scholar] [CrossRef] [PubMed]
  19. Contreras-Vidal, J.L.; Bortole, M.; Zhu, F.; Nathan, K.; Venkatakrishnan, A.; Francisco, G.E.; Soto, R.; Pons, J.L. Neural decoding of robot-assisted gait during rehabilitation after stroke. Am. J. Phys. Med. Rehabil. 2018, 97, 541–550. [Google Scholar] [CrossRef] [PubMed]
  20. Gwin, J.T.; Gramann, K.; Makeig, S.; Ferris, D.P. Removal of movement artifact from high-density EEG recorded during walking and running. J. Neurophysiol. 2010, 103, 3526–3534. [Google Scholar] [CrossRef]
  21. Gwin, J.T.; Gramann, K.; Makeig, S.; Ferris, D.P. Electrocortical activity is coupled to gait cycle phase during treadmill walking. Neuroimage 2011, 54, 1289–1296. [Google Scholar] [CrossRef]
  22. Seeber, M.; Scherer, R.; Wagner, J.; Solis-Escalante, T.; Müller-Putz, G.R. High and low gamma EEG oscillations in central sensorimotor areas are conversely modulated during the human gait cycle. Neuroimage 2015, 112, 318–326. [Google Scholar] [CrossRef] [PubMed]
  23. Roeder, L.; Boonstra, T.W.; Smith, S.S.; Kerr, G.K. Dynamics of corticospinal motor control during overground and treadmill walking in humans. J. Neurophysiol. 2018, 120, 1017–1031. [Google Scholar] [CrossRef]
  24. Mate, K.K. Effects of Visual, Auditory, and Combined Cues on Human Movement and Brain Regions Involved in Perception Action. McGill J. Med. 2022, 20, 38. [Google Scholar] [CrossRef]
  25. Bulea, T.C.; Kim, J.; Damiano, D.L.; Stanley, C.J.; Park, H.-S. User-driven control increases cortical activity during treadmill walking: An EEG study. In Proceedings of the 2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Chicago, IL, USA, 26–30 August 2014; pp. 2111–2114. [Google Scholar]
  26. Michel, C.M. Chapter 12—High-resolution EEG. In Handbook of Clinical Neurology; Levin, K.H., Chauvel, P., Eds.; Elsevier: Amsterdam, The Netherlands, 2019; Volume 160, pp. 185–201. [Google Scholar]
  27. Knutson, G.A. Anatomic and functional leg-length inequality: A review and recommendation for clinical decision-making. Part I, anatomic leg-length inequality: Prevalence, magnitude, effects and clinical significance. Chiropr. Osteopat. 2005, 13, 11. [Google Scholar] [CrossRef]
  28. Yokoyama, H.; Kaneko, N.; Masugi, Y.; Ogawa, T.; Watanabe, K.; Nakazawa, K. Gait-phase-dependent and gait-phase-independent cortical activity across multiple regions involved in voluntary gait modifications in humans. Eur. J. Neurosci. 2021, 54, 8092–8105. [Google Scholar] [CrossRef] [PubMed]
  29. Kline, J.E.; Huang, H.J.; Snyder, K.L.; Ferris, D.P. Isolating gait-related movement artifacts in electroencephalography during human walking. J. Neural Eng. 2015, 12, 046022. [Google Scholar] [CrossRef]
  30. Nathan, K.; Contreras-Vidal, J.L. Negligible motion artifacts in scalp electroencephalography (EEG) during treadmill walking. Front. Hum. Neurosci. 2016, 9, 708. [Google Scholar] [CrossRef]
  31. Sabatini, A.M. Quaternion-based extended Kalman filter for determining orientation by inertial and magnetic sensing. IEEE Trans. Biomed. Eng. 2006, 53, 1346–1356. [Google Scholar] [CrossRef]
  32. Frølich, L.; Dowding, I. Removal of muscular artifacts in EEG signals: A comparison of linear decomposition methods. Brain Inform. 2018, 5, 13–22. [Google Scholar] [CrossRef]
  33. Gorjan, D.; Gramann, K.; De Pauw, K.; Marusic, U. Removal of movement-induced EEG artifacts: Current state of the art and guidelines. J. Neural Eng. 2022, 19, 011004. [Google Scholar] [CrossRef] [PubMed]
  34. Bist, K. Artifact Removal in EEG Using Fast ICA; Mondragon Unibertsitatea: Mondragón, Spain, 2015. [Google Scholar] [CrossRef]
  35. Jacquelin Perry, M. Gait Analysis: Normal and Pathological Function; SLACK: Haddonfield, NJ, USA, 2010. [Google Scholar]
  36. Abhayasinghe, N.; Murray, I. Human gait phagse recognition based on thigh movement computed using IMUs. In Proceedings of the 2014 IEEE Ninth International Conference on Intelligent Sensors, Sensor Networks and Information Processing (ISSNIP), Singapore, 21–24 April 2014; pp. 1–4. [Google Scholar]
  37. Semaan, M.B.; Wallard, L.; Ruiz, V.; Gillet, C.; Leteneur, S.; Simoneau-Buessinger, E. Is treadmill walking biomechanically comparable to overground walking? A systematic review. Gait Posture 2022, 92, 249–257. [Google Scholar] [CrossRef]
  38. do Nascimento, O.F.; Nielsen, K.D.; Voigt, M. Influence of directional orientations during gait initiation and stepping on movement-related cortical potentials. Behav. Brain Res. 2005, 161, 141–154. [Google Scholar] [CrossRef] [PubMed]
  39. Luu, T.P.; Nakagome, S.; He, Y.; Contreras-Vidal, J.L. Real-time EEG-based brain-computer interface to a virtual avatar enhances cortical involvement in human treadmill walking. Sci. Rep. 2017, 7, 8895. [Google Scholar] [CrossRef]
  40. Teitti, S.; Määttä, S.; Säisänen, L.; Könönen, M.; Vanninen, R.; Hannula, H.; Mervaala, E.; Karhu, J. Non-primary motor areas in the human frontal lobe are connected directly to hand muscles. Neuroimage 2008, 40, 1243–1250. [Google Scholar] [CrossRef]
  41. Praamstra, P.; Oostenveld, R. Attention and movement-related motor cortex activation: A high-density EEG study of spatial stimulus–response compatibility. Cogn. Brain Res. 2003, 16, 309–322. [Google Scholar] [CrossRef]
  42. Tamás, G.; Chirumamilla, V.C.; Anwar, A.R.; Raethjen, J.; Deuschl, G.; Groppa, S.; Muthuraman, M. Primary sensorimotor cortex drives the common cortical network for gamma synchronization in voluntary hand movements. Front. Hum. Neurosci. 2018, 12, 130. [Google Scholar] [CrossRef]
  43. Rosso, M.; Leman, M.; Moumdjian, L. Neural entrainment meets behavior: The stability index as a neural outcome measure of auditory-motor coupling. Front. Human. Neurosci. 2021, 15, 668918. [Google Scholar] [CrossRef] [PubMed]
  44. Nenna, F.; Do, C.-T.; Protzak, J.; Gramann, K. Alteration of brain dynamics during natural dual-task walking. bioRxiv 2020. [Google Scholar] [CrossRef]
  45. Morand, S.M.; Harvey, M.; Grosbras, M.-H. Parieto-occipital cortex shows early target selection to faces in a reflexive orienting task. Cereb. Cortex 2014, 24, 898–907. [Google Scholar] [CrossRef]
  46. Pasalar, S.; Ro, T.; Beauchamp, M.S. TMS of posterior parietal cortex disrupts visual tactile multisensory integration. Eur. J. Neurosci. 2010, 31, 1783–1790. [Google Scholar] [CrossRef]
  47. Vandenberghe, R.; Gillebert, C.R. Parcellation of parietal cortex: Convergence between lesion-symptom mapping and mapping of the intact functioning brain. Behav. Brain Res. 2009, 199, 171–182. [Google Scholar] [CrossRef]
  48. Bradford, J.C.; Lukos, J.R.; Ferris, D.P. Electrocortical activity distinguishes between uphill and level walking in humans. J. Neurophysiol. 2016, 115, 958–966. [Google Scholar] [CrossRef]
  49. Bulea, T.C.; Kim, J.; Damiano, D.L.; Stanley, C.J.; Park, H.-S. Prefrontal, posterior parietal and sensorimotor network activity underlying speed control during walking. Front. Hum. Neurosci. 2015, 9, 247. [Google Scholar] [CrossRef]
  50. Gramann, K.; Gwin, J.T.; Bigdely-Shamlo, N.; Ferris, D.P.; Makeig, S. Visual evoked responses during standing and walking. Front. Hum. Neurosci. 2010, 4, 202. [Google Scholar] [CrossRef] [PubMed]
  51. Oliveira, A.S.; Schlink, B.R.; Hairston, W.D.; König, P.; Ferris, D.P. Restricted vision increases sensorimotor cortex involvement in human walking. J. Neurophysiol. 2017, 118, 1943–1951. [Google Scholar] [CrossRef] [PubMed]
  52. Sburlea, A.I.; Montesano, L.; de la Cuerda, R.C.; Alguacil Diego, I.M.; Miangolarra-Page, J.C.; Minguez, J. Detecting intention to walk in stroke patients from pre-movement EEG correlates. J. Neuroeng. Rehabil. 2015, 12, 113. [Google Scholar] [CrossRef] [PubMed]
  53. Snyder, K.L.; Kline, J.E.; Huang, H.J.; Ferris, D.P. Independent component analysis of gait-related movement artifact recorded using EEG electrodes during treadmill walking. Front. Hum. Neurosci. 2015, 9, 639. [Google Scholar] [CrossRef] [PubMed]
  54. Wagner, J.; Solis-Escalante, T.; Grieshofer, P.; Neuper, C.; Müller-Putz, G.; Scherer, R. Level of participation in robotic-assisted treadmill walking modulates midline sensorimotor EEG rhythms in able-bodied subjects. Neuroimage 2012, 63, 1203–1211. [Google Scholar] [CrossRef] [PubMed]
  55. Weersink, J.B.; Maurits, N.M.; de Jong, B.M. EEG time-frequency analysis provides arguments for arm swing support in human gait control. Gait Posture 2019, 70, 71–78. [Google Scholar] [CrossRef] [PubMed]
Sensors 24 01548 g001
Figure 1. Montage of the EEG system in the experiment, 32 electrodes.
Figure 1. Montage of the EEG system in the experiment, 32 electrodes.
Sensors 24 01548 g001
Sensors 24 01548 g002
Figure 2. (a) The average of 105 gait cycles from a participant walking in the TsSC condition over 3 min; (b) a typical example of raw IMU data from a participant walking in the TsSC condition. The red mark indicates the peak of the IMU data, representing the initial contact point of the gait cycle.
Figure 2. (a) The average of 105 gait cycles from a participant walking in the TsSC condition over 3 min; (b) a typical example of raw IMU data from a participant walking in the TsSC condition. The red mark indicates the peak of the IMU data, representing the initial contact point of the gait cycle.
Sensors 24 01548 g002
Sensors 24 01548 g003aSensors 24 01548 g003b
Figure 3. A colormap showing the average (blue line), and SD (gray shade) brain activity of the participants in each walking task: TsSC, TSC, OsSC, OSC. The colormap represents gait-phase-related brain activity changes in arbitrary units (a.u.) in the AmEEGDel (0.5–4 Hz). The average and SD represent gait-phase-related brain activity changes in microvolts (μV). Brain activities across the phases of the gait cycle showing EEG magnitude and direction at (a) Fz and (b) FC1 positions. The red vertical line demonstrates the percentage phase of the gait cycle where a significant interaction effect between groups (O, T) and conditions (sSC, SC) was identified.
Figure 3. A colormap showing the average (blue line), and SD (gray shade) brain activity of the participants in each walking task: TsSC, TSC, OsSC, OSC. The colormap represents gait-phase-related brain activity changes in arbitrary units (a.u.) in the AmEEGDel (0.5–4 Hz). The average and SD represent gait-phase-related brain activity changes in microvolts (μV). Brain activities across the phases of the gait cycle showing EEG magnitude and direction at (a) Fz and (b) FC1 positions. The red vertical line demonstrates the percentage phase of the gait cycle where a significant interaction effect between groups (O, T) and conditions (sSC, SC) was identified.
Sensors 24 01548 g003aSensors 24 01548 g003b
Table 1. The number of steps in 3 min (stride cycle) in each walking condition.
Table 1. The number of steps in 3 min (stride cycle) in each walking condition.
Subject IDGroupSpeed ConditionSubject IDGroupSpeed Condition
sSCSCsSCSC
N1T10568N13O7269
N2T11968N14O9368
N3T12969N15O8967
N4T11368N16O10669
N5T7668N17O9169
N6T10368N18O8169
N7T11067N19O9968
N8T12469N20O8467
N9T13467N21O7967
N10T11969N22O6568
N11T13768N23O7868
N12T13167N24O12468
Average steps (SD)T116.7 (17.0)68 (0.7)Average stepsO88.4 (16.0)68 (0.8)
T: treadmill walking, O: overground walking; sSC: walking with self-selected speed control, SC: walking with speed control.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Tharawadeepimuk, K.; Limroongreungrat, W.; Pilanthananond, M.; Nanbancha, A. Auditory Cue Effects on Gait-Phase-Dependent Electroencephalogram (EEG) Modulations during Overground and Treadmill Walking. Sensors 2024, 24, 1548. https://doi.org/10.3390/s24051548

AMA Style

Tharawadeepimuk K, Limroongreungrat W, Pilanthananond M, Nanbancha A. Auditory Cue Effects on Gait-Phase-Dependent Electroencephalogram (EEG) Modulations during Overground and Treadmill Walking. Sensors. 2024; 24(5):1548. https://doi.org/10.3390/s24051548

Chicago/Turabian Style

Tharawadeepimuk, Kittichai, Weerawat Limroongreungrat, Metaneeya Pilanthananond, and Ampika Nanbancha. 2024. "Auditory Cue Effects on Gait-Phase-Dependent Electroencephalogram (EEG) Modulations during Overground and Treadmill Walking" Sensors 24, no. 5: 1548. https://doi.org/10.3390/s24051548

APA Style

Tharawadeepimuk, K., Limroongreungrat, W., Pilanthananond, M., & Nanbancha, A. (2024). Auditory Cue Effects on Gait-Phase-Dependent Electroencephalogram (EEG) Modulations during Overground and Treadmill Walking. Sensors, 24(5), 1548. https://doi.org/10.3390/s24051548

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop