The Application of Electro- and Magneto-Encephalography in Tinnitus Research – Methods and Interpretations

Abstract

In recent years, there has been a significant increase in the use of electroencephalography (EEG) and magnetoencephalography (MEG) to investigate changes in oscillatory brain activity associated with tinnitus with many conflicting results. Current view of the underlying mechanism of tinnitus is that it results from changes in brain activity in various structures of the brain as a consequence of sensory deprivation. This in turn gives rise to increased spontaneous activity and/or synchrony in the auditory centers but also involves modulation from non-auditory processes from structures of the limbic and paralimbic system. Some of the neural changes associated with tinnitus may be assessed non-invasively in human beings with MEG and EEG (M/EEG) in ways, which are superior to animal studies and other non-invasive imaging techniques. However, both MEG and EEG have their limitations and research results can be misinterpreted without appropriate consideration of these limitations. In this article, I intend to provide a brief review of these techniques, describe what the recorded signals reflect in terms of the underlying neural activity, and their strengths and limitations. I also discuss some pertinent methodological issues involved in tinnitus-related studies and conclude with suggestions to minimize possible discrepancies between results. The overall message is that while MEG and EEG are extremely useful techniques, the interpretation of results from tinnitus studies requires much caution given the individual variability in oscillatory activity and the limits of these techniques.

EEG oscillatory power dissociates between distress- and depression-related psychopathology in subjective tinnitus

“…We are well-aware, that not all of these suggestions have the same gravity in regards to source estimation and diverging results across studies, yet are convinced that also minor differences may influence results as in tinnitus research weak effects, nil findings and irreconcilable results are omnipresent. Finally, while not directly addressing the methodological comparison of the two studies but general validity of the applied source estimation, limitations of the method should be considered and properly discussed (Adjamian, 2014). Of central interest here, we identify the fact the standardized boundary element method volume conductor model behind the LORETA algorithm (Fuchs et al, 2002) is based on a standardized MNI brain template derived from a young, healthy cohort in their mid-twenties, which generally contrasts the mean age of the samples used in tinnitus studies (mean=42.78 years, in our study).…”

“…However, we favor neurophysiological M/EEG techniques in the context of tinnitus research because these approaches are not accompanied by bothersome gradient noise and hence are more comfortable for TI, who often show symptoms of hyperacusis. Further, comprehensive leaps have been made to improve the spatial resolution of M/EEG through the development of innovative source estimation approaches (Adjamian, 2014).…”

“…In the realm of M/EEG, spectral power analysis of spontaneous oscillations during the resting-state has been identified as an advantageous tool, because TI have demonstrated deviations from the normal EEG pattern in the activity strength of various frequency bands (for an overview see Adjamian 2014;Elgoyhen et al 2015). Resting-state neurophysiological measurements seem ideally suited to investigate the neural correlates of tinnitus because, during data acquisition, participants are in a relaxed and quiet setting without the distractions of external stimulation.…”

Accounting for Heterogeneity: Mixed-Effects Models in Resting-State EEG Data in a Sample of Tinnitus Sufferers

“…Indeed, a clear distinction of similarities and dissimilarities between the tinnitus and non-tinnitus brain has yet to be identified, as findings have not been consistent across studies (Elgoyhen et al 2015;Güntensperger et al 2017). Potential reasons for the discrepancies in oscillatory findings are manifold, and may include differences in study design and sample selection; technique-and analysis-specific aspects (Adjamian 2014;Gross et al 2013;Meyer et al 2017); biological factors such as age (Schlee et al 2012), tinnitus duration (Schlee et al 2009), and degree of hearing loss (Adjamian et al 2012); comorbidities such as hyperacusis, sleep disorders, headache, and concentration problems (Zirke et al 2010); and associated psychopathological symptoms such as tinnitus-related distress and depression, which contribute their own levels of oscillatory correlates. Along these lines, Meyer et al conducted a study that was explicitly designed to re-enact a former EEG resting-state experiment (Joos et al 2012;Meyer et al 2017).…”

Characteristics of tinnitus in adolescents and association with psychoemotional factors

“…Diagnosing tinnitus is difficult, because the condition may not be detected using objective methods. Attempts have been made to address this problem using brain imaging modalities, such as functional magnetic resonance imaging and electroencephalometry . These modalities have demonstrated that tinnitus patients show modulated brain activity in nonauditory centers, such as limbic, paralimbic, and even frontal areas, as well as in the auditory centers .…”