Publication Details
Overview
 
 
Fahimeh Akbarian
 

Thesis

Abstract 

Multiple sclerosis (MS) is a chronic disease affecting the central nervous system,characterized by neuroinflammation, demyelination, and neurodegeneration.Many people with MS (pwMS) experience cognitive difficulties, especially with working memory, processing speed, and attention. One proposed mechanism is a disruption of the balance between “excitation” (signals that drive activity)and “inhibition” (signals that suppress activity) in the brain, partly due to the loss of inhibitory synapses.Recent studies suggest that the 1/f spectral slope of the aperiodic component of neural activity provides a non-invasive marker of the excitation-inhibition(E/I) balance. A steeper 1/f slope reflects increased inhibition and/or reduced excitation. In this thesis, we used magnetoencephalography (MEG) data, an electrophysiological technique that captures neuronal activity with millisecond temporal resolution, to explore this spectral marker in MS.In three studies, we analysed MEG data from healthy controls (HCs) and people with MS (pwMS) to investigate the 1/f spectral slope during resting state, visuo-verbal n-back task (a memory task), and auditory oddball task (a sound detection task). Data were preprocessed, source reconstructed, and summarized into 42 cortical regions. Periodic and aperiodic spectral components were separated using the specparam (formerly FOOOF) algorithm. Neuropsychological tests were conducted to assess information processing speed, verbal learning, verbal fluency, and visuospatial memory.During resting-state eyes closed, pwMS being administered benzodiazepines showed steeper 1/f slopes than pwMS not treated with benzodiazepines; these effect remained after adjusting for periodic beta power, supporting the slope as an oscillation-independent measure. In pwMS not treated with benzodiazepines, cognitively impaired pwMS showed steeper slopes compared to cognitively preserved ones and HCs, suggesting a compensatory mechanism of over-inhibition.During the visuo-verbal n-back task, pre- and post-stimulus windows were analysed to evaluate changes in spectral components induced by the stimulus. Across all subjects, a consistent post-stimulus 1/f steepening (indicating increased inhibition) was observed. However, compared to HCs, pwMS exhibited a flatter post-distractor slope (indicating less inhibition), which suggests impaired phasic inhibitory control when irrelevant input must be suppressed.Furthermore, task-induced steepening was associated with offline visuospatial memory: in pwMS not treated with benzodiazepines, greater steepening predicted better performance, whereas the opposite relationship was observed in healthy controls.During the auditory oddball task, post-stimulus slope steepening remained significant even after correcting for event-related fields. Salient (non-standard)stimuli evoked larger steepening, and trials that required responses showed enhanced steepening over the sensorimotor cortex, consistent with response locked inhibitory gating. Cross-paradigm analyses indicated that individual differences in slope modulation appeared consistent across auditory oddball and n-back tasks for HCs and pwMS not treated with benzodiazepines. This finding suggests that the 1/f slope can be considered as a trait-like, paradigm independent index of cognitive control.Together, these findings indicate that the aperiodic 1/f slope effectively captures both the tonic (ongoing) and the phasic (transient) inhibitory dynamics altered in MS. It differentiates between medication effects, correlates with behaviour and cognitive scores, and generalizes across experimental paradigms.Consequently, the 1/f slope is proposed as a potential interpretable biomarker for assessing cognitive dysfunction in MS, motivating further longitudinal and multimodal validation, such as integration with PET and/or MR Spectroscopy.

Reference