Fibromyalgia (FM) and other chronic pain syndromes are associated with cognitive dysfunction and attentional deficits, but the neural basis of such alterations is poorly understood. Dyscognition may be related to high levels of neural noise, understood as increased random electrical fluctuations that impair neural communication; however, this hypothesis has not yet been tested in any chronic pain condition. Here we compared electroencephalographic activity (EEG) in 18 FM patients -with high self-reported levels of cognitive dysfunction- and 22 controls during a cognitive control task. We considered the slope of the Power Spectrum Density (PSD) as an indicator of neural noise. As the PSD slope is flatter in noisier systems, we expected to see shallower slopes in the EEG of FM patients. Higher levels of neural noise should be accompanied by reduced power modulation and reduced synchronization between distant brain locations after stimulus presentation. As expected, FM patients showed flatter PSD slopes. After applying a Laplacian spatial filter, we found reduced theta and alpha power modulation and reduced midfrontal-posterior theta phase synchronization. Results suggest higher neural noise and impaired local and distant neural coordination in the patients and support the neural noise hypothesis to explain dyscognition in FM.
Many chronic pain patients present impaired cognitive functioning, as indicated by poorer performance and slower reaction times in neuropsychological tasks. These alterations occur more frequently in patients with generalized pain such as fibromyalgia (FM), in which cognitive dysfunction has recently been recognized as one of its core symptoms. Although impaired cognition may have a strong impact on quality of life, the causes of dyscognition in FM and other chronic pain pathologies are poorly understood.
Cognition depends on dynamic interactions between local and large scale brain networks. Disruption of these interactions by noisy and unreliable information transfer can thus cause a decline in cognitive performance. The slope of the power spectral density (PSD) plot has been proposed as a useful indicator of the level of neural noise. In the human electroencephalogram (EEG), a reduction in power is observed as frequency increases, yielding a negative slope in the PSD plot. Synchronized neural activity is related to steeper PSD slopes, while desynchronized activity yields flatter slopes (as an extreme, a random signal such as white noise gives a flat line). This relationship between PSD slope and correlated/decorrelated neural activity has also been observed using computer simulations. Therefore, the slope of the linear fit of the PSD can be used to estimate the degree of noise in a system, being flatter when there is more uncoordinated activity in large neural populations.
Stimulus processing in less coordinated and noisier environments may be blurred. Therefore an increase in neural noise should theoretically be accompanied by degraded time-frequency power modulation after stimulus presentation. Cognitive control and visual attention requirements are related to power modulations in midfrontal theta and posterior alpha. In this sense, higher noise should impair synchronization of local neural populations, reflected by reduced theta and alpha power modulation after stimulus presentation.
Cognitive dysfunction may also be related to impaired interactions between distant brain locations. Thus, the predicted increase in neural noise should also be accompanied by degraded neural communication in distant-range networks. Such distant brain areas (e.g. fronto-posterior regions) are believed to be synchronized by theta or other low frequency oscillations.
Although the presence of heightened neural noise may be a plausible explanation for the cognitive dysfunction observed in chronic pain patients, as far as we are aware this possibility has not yet been investigated. In the present study, we analysed the slope of the PSD plot, posterior alpha and midfrontal theta power, and fronto-posterior functional connectivity (theta phase synchronization) derived from EEGs recorded while study participants performed the Multi-Source Interference Task (MSIT). The MSIT is designed to assess the integrity of cognitive/attentional networks. We hypothesize that PSD slopes would be flatter for patients than for healthy controls. As convergent evidence, we also expected to find reduced modulation of midfrontal theta and posterior alpha power, along with a lower level of fronto-posterior theta phase synchronization in patients.