Christopher Schow and Dr. Scott C. Steffensen Ph.D., Department of Psychology

Introduction:

Biofeedback is a process by which a person may learn to control physiological functions of their body they would otherwise not be aware of e.g. heart rate. This is accomplished by relaying real-time information back to the user. The user then changes their physiological activity in order to achieve set goals. Neurofeedback is one form of biofeedback that utilizes EEG to collect a person’s electrical brainwave activity and displays it to them, usually in the form of a video game.

Recent research indicates that neurofeedback may be effective in improving physical balance. Hammond (2005) used a neurofeedback procedure, which places electrodes just beneath electrode sites O1 and O2 (areas lateral to midline on back of the head) to treat patients with balance disorders. The treatment was used in four cases and Hammond observed improvements in balance after two to three sessions of neurofeedback training. Hammond (2005) learned the protocol from Ayers (p. 27), who also used neurofeedback training to treat stroke victims. Ayers reports significant improvements in physical balance when using neurofeedback for this purpose (Ayers, 1999). Azarpaikan, Torbati, Sohrabi (2014) used a similar protocol to the one described in Ayers (1999) and Hammond (2005) to study the efficacy of neurofeedback in improving physical balance in Parkinson’s Disease (PD) patients. Data analysis of the treatment group demonstrated significant improvement between initial and final balance assessments. Additionally, the treatment group was shown to improve in balance significantly more than a sham control (p. 179).

The present study is designed to act as a proof of concept for the application of neurofeedback to improve physical balance in general and as a precursor to specific balance disorder studies in the future. To accomplish this, a protocol similar to the ones utilized by Ayers (1999), Hammond (2005), and Azarpaikan et al. (2014) was implemented.

Methodology:

Participants for the study were students recruited from a class in the neuroscience major at Brigham Young University. All participants were randomly assigned to either the neurofeedback group (NF) (n = 9) or the sham control group (SC) (n = 6). Additionally, a negative control group (C) (n = 9) was recruited using flyers placed around Brigham Young University’s campus. Criteria for exclusion were individuals with a history of head injury, injury to the lower extremities, stroke, ataxia and other balance disorders. Also, females participated in the study; however, their data will be excluded from the final data analysis due to hormone fluctuations of the menstrual cycle and the possible effects of hormone metabolites on the GABAergic system in the brain. Approval for the study was given by Brigham Young Universities Institutional Review Board (protocol number X14242).

Balance was assessed using Nintendo (Kyoto, Kyoto Prefecture, Japan) Wii Balance board. Six measures were used: 1. Eyes opened feet together, 2. Eyes opened left foot balance, 3. Eyes opened right foot balance, 4. Eyes closed feet together, 5. Eyes closed left foot balance, and 6. Eyes closed right foot balance. Data from each measurement was plotted for 10 seconds and area under curve summated for indication of balance as a deviation from center. All three groups (NF, SC, C) underwent an initial and final balance assessment. Between initial and final balance assessments, six 20 minute neurofeedback or sham neurofeedback sessions were administered to NF and SC respectively. EEG data was collected from NF utilizing EEG headsets (EPOC Emotiv) developed by Emotiv Limited (Kwun Tong, Hong Kong). EEG data from O1 and O2 sensors of the headset was used as input into two video game programs we developed using LabVIEW software created by National Instruments (Austin, TX). The videogames functioned by taking a ratio of theta to beta brainwave activity and allowing the participant to accomplish tasks when the ratio fell beneath a specified value. The purpose of this was to train the participant to inhibit theta activity and increase beta activity. An example of this is our neurofeedback game “Rat Race.” This game displays two rats on the left side of the screen; one green rat and one blue rat. When the theta/beta ratio was high the green rat would move forward, when the ratio was low the blue rat would move forward. The goal of this game was to get the blue rat across the finish line first. SC was administered two alternate programs with similar display. However, EEG was not collected from SC but was realistic simulated EEG data that controlled the program.

Results:

All data has been collected, but the analysis of the large dataset is still being conducted. Our initial analysis between groups (ANOVA) demonstrated no significance between any of the groups and no improvement within any of the groups. Consequentially, EEG data will be analyzed to determine which participants were able to improve their theta/beta ratio over the course of the study. Once this is done, the corresponding balance results of those who improved their ratio compared to those who did not (whether in SC or NF) will be analyzed.

Discussion:

Other studies utilizing a similar protocol to the one utilized in this study have researched the efficacy of neurofeedback training as a treatment for specific balance disorders or diseases known to result in difficulty with physical balance such as PD. Our study differs in this respect because all the participants were healthy. This may have limited our ability to detect significance in our results; conversely, it also may have provided us with a proof of concept and validation of the protocol used. In addition, our study was limited by a small and specific sample. Due to the fact our sample was so specific; the extent to which our study can be generalized is limited. Also, because the participants in NF and SC were receiving class credit for participation, they may not have been fully motivated to give a full effort during their sessions. Another potential limitation of this study is the use of the EPOC Emotiv technology, a relatively new system for collecting EEG. Much of the research on the efficacy of the EPOC Emotiv system seems to indicate that it is functional, but is not on the level of medical-grade EEG systems (Duvinage et al., 2013, p. 10-12). Additionally, the results from our balance assessment appear to be highly variable. Future research needs to include larger sample sizes, recruitment methods that will provide samples that can be more generalized and adaptations to the balance assessment used for improved data clarity.

Conclusion:

Further data analysis is necessary before drawing any conclusions from this study. From our initial analysis, it appears that there was no significant difference between groups and no significant improvement within groups. If further analysis lends the same results then adaptations need to be made to our methodology and more research needs to be conducted to determine the efficacy of neurofeedback in improving physical balance.

References:

Ayers, M.E. (1999). Assessing and treating open head trauma, coma, and stoke using real-time digital EEG neurofeedback. In J. R. Evans & A. Abarbanel (Eds.), Introduction to Quantitative EEG and Neurofeedback. (pp. 203-222). New York, NY: Academic Press.

Azarpaikan, A., Torbati, H. T., & Sohrabi, M. (2014). Neurofeedback and physical balance in parkinson’s patients. Gait and Posture, 40, 177-181.

Duvinage, M., Castermans, T., Petieau, M., Hoellinger, T., Cheron, G., & Dutoit, T. (2013). Performance of the Emotiv Epoc headset for P300-based applications. Biomedical Engineering Online, 12(1), 1-15. doi:10.1186/1475-925X-12-56

Hammond, C. (2005). Neurofeedback to improve physical balance, incontinence, and swallowing. Journal of Neurotherapy, 9(1), 27-36.