Assessing the specificity of the relationship between brain alpha oscillations and tonic pain

Slowing in Peak-Alpha Frequency Recorded After Experimentally-Induced Muscle Pain is not Significantly Different Between High and Low Pain-Sensitive Subjects

“…Several studies have investigated PAF and subjective perception of pain, reporting contrasting results. A previous study described a positive correlation between PAF and pain NRS score 41 as confirmed by the current research, while others found a negative correlation 21,22 or no correlation 55 . In addition to several differences between the studies (i.e., different pain modalities, a diverse range of self-reported pain, reliability of self-reported pain across days), EEG data processing and alpha wave characteristics may help explain these partially divergent findings.…”

“…The PAF reduction during pain in the current study agrees with some of the previous experimental pain studies, which mostly showed decreased amplitude or peak frequency slowing of alpha oscillations during cutaneous tonic pain 13,19,21,40,43,49 . However, a few studies failed to show any alpha band changes 12,41,55 , or they found an increased power 2 , probably due to methodological differences, such as EEG data recordings or data processing.…”

“…However, most of the previous findings were inconclusive, especially concerning alpha brain oscillations. Recently, depending on the analytical approach and the pain model, several researchers have shown that PAF was found to reflect 21,22,41 or not reflect 55 pain sensitivity across individuals. Moreover, both positive 41 and negative 21,22 correlations between PAF and pain intensity have been reported.…”

Classification of Tonic Pain Experience based on Phase Connectivity in the Alpha Frequency Band of the Electroencephalogram using Convolutional Neural Networks

“…The average temperature of H condition was 44.5 °C, and the temperature of W was 6 lower, the temperature was adjusted according to the individual's unpleasantness ratings delivered through visual-analogue scale (VAS). For more details about the methodology and procedure see [6].…”

“…The first step in building a machine learning model is to investigate the optimal feature to be used as input. The quest for a specific EEG signature of pain has been mostly unsuccessful over the years because most of the brain responses are similar across sensory modalities and affective states [4] However, recent research indicates the peak of alpha (8-12 Hz) frequency (PAF) as a predictive index of pain sensitivity [5], [6]. Similarly, functional brain connectivity, the representation of integration among brain regions, has also been proposed as a metric of pain intensity in the brain [7].…”