Using EEG Alpha States to Understand Learning During Alpha Neurofeedback Training for Chronic Pain

Abstract

ObjectiveAlpha-neurofeedback (α-NFB) is a novel therapy which trains individuals to volitionally increase their alpha power to improve pain. Learning during NFB is commonly measured using static parameters such as mean alpha power. Considering the biphasic nature of alpha rhythm (high and low alpha), dynamic parameters describing the time spent by individuals in high alpha state and the pattern of transitioning between states might be more useful. Here, we quantify the changes during α-NFB for chronic pain in terms of dynamic changes in alpha states.MethodsFour chronic pain and four healthy participants received five NFB sessions designed to increase frontal alpha power. Changes in pain resilience were measured using visual analogue scale (VAS) during repeated cold-pressor tests (CPT). Changes in alpha state static and dynamic parameters such as fractional occupancy (time in high alpha state), dwell time (length of high alpha state) and transition probability (probability of moving from low to high alpha state) were analyzed using Friedman’s Test and correlated with changes in pain scores using Pearson’s correlation.ResultsThere was no significant change in mean frontal alpha power during NFB. There was a trend of an increase in fractional occupancy, mean dwell duration and transition probability of high alpha state over the five sessions in chronic pain patients only. Significant correlations were observed between change in pain scores and fractional occupancy (r = −0.45, p = 0.03), mean dwell time (r = -0.48, p = 0.04) and transition probability from a low to high state (r = -0.47, p = 0.03) in chronic pain patients but not in healthy participants.ConclusionThere is a differential effect between patients and healthy participants in terms of correlation between change in pain scores and alpha state parameters. Parameters providing a more precise description of the alpha power dynamics than the mean may help understand the therapeutic effect of neurofeedback on chronic pain.

Neurofeedback Therapy for Sensory Over-Responsiveness—A Feasibility Study

“…Yet, the pick alpha frequency-A well-recognized marker for pain sensitivity [92,93] utilized as a target for alleviating pain, e.g., [44]-was found to be linked to self-reported daily pain sensitivity in this study. This finding supports previously reported coupling between alpha oscillations and pain sensitivity [44,46,47]. However, recently, it was suggested that alpha power may not measure the success of learning underlying pain reduction via neurofeedback.…”

“…Specifically, alpha oscillation is characterized by two modes of high and low amplitude [52], suggesting that extracting the mean power may be misleading [47]. Indeed, parameters measuring the dynamics in alpha state, not reflected in the mean power, were found as more sensitive to capture learning [47]. However, we found a very large effect size, indicating meaningful changes in delta and theta bands.…”

“…However, recently, it was suggested that alpha power may not measure the success of learning underlying pain reduction via neurofeedback. Specifically, alpha oscillation is characterized by two modes of high and low amplitude [52], suggesting that extracting the mean power may be misleading [47]. Indeed, parameters measuring the dynamics in alpha state, not reflected in the mean power, were found as more sensitive to capture learning [47].…”

“…Alpha is considered an inhibitory oscillation [39], and in pain conditions shows reduced power [40][41][42][43]. Like pain patients [44][45][46][47], individuals with SOR during EEG resting-state recording demonstrated an overall reduction in cortical activity, most prominently in the alpha band power [48]. Interestingly, alpha activity within the sensory cortex is linked to environmental stimulation intake, and lower power is suggested to accompany a state of higher stress or arousal, both characterizing individuals with SOR [14,16,48].…”

Resting-state electroencephalography and magnetoencephalography as biomarkers of chronic pain: a systematic review

“…Finally, 76 studies were included. 1,5–8,10,11,15,16,18–27,30–33,35,37–49,51,56,59,60,63,66–69,74–76,79,80,82,89–91,95–98,100–105,109,110,112–115,120,122,124–126 Figure 1 depicts the PRISMA flow diagram of study selection along with reasons for study exclusion at every stage.…”