A NEW WINDOW ON TINNITUS: EXPLORING FUNCTIONAL CONNECTIVITY IN THE CLASSICAL AND NON-CLASSICAL AUDITORY PATHWAYS

Dave Langers1, Robert Levine1, Jennifer Melcher1
1 Eaton-Peabody Laboratory, Massachusetts

Eye and Ear Infirmary, Boston Functional magnetic resonance imaging (fMRI) has proved to be useful for the investigation of the neural mechanisms that underlie auditory perception.

However, the application of fMRI in tinnitus research is hampered by the fact that the tinnitus percept is inherently subjective. The present study begins to bring a promising new fMRI approach to bear on tinnitus.

Standard fMRI techniques involve measuring brain activity correlated with an external stimulus such as sound. The approach examined here instead detects temporal correlations of activity between brain areas. This so-called functional connectivity approach does not necessarily require an external stimulus and has the capability of revealing functional networks in the brain during “resting state”.

In tinnitus patients, this method may be able to reveal functional abnormalities that are directly related to the presence of tinnitus: for instance, connections in the non-classical auditory pathway between auditory and emotional centers may be abnormally strong.

Our current goal is to assess whether it is feasible to assess functional connectivity in the central auditory system. We report preliminary outcomes of a study that has been carried out on 13 control subjects with normal hearing; in addition 4 tinnitus patients have currently been included. Data were acquired using a sparse, cardiac-gated fMRI sequence. Subjects were engaged in a simple visual task while passively listening to 32-s broadband noise fragments.

Independent component analysis was performed on the measured fMRI signals; this method extracts relevant functional networks without prior information about the stimulus paradigm. In addition, functional connectivity analysis was carried out to identify brain regions with similar functional characteristics, by correlating the fMRI signals in the entire brain with those of predefined regions of interest.

In all subjects, components could reliably be identified that comprised the bilateral auditory cortices, thalamus (medial geniculate body), midbrain (inferior colliculus) and brainstem (cochlear nucleus). In addition, correlated activity of limbic structures in the thalamus (medial dorsal nucleus) and cortex (cingulate gyrus) was observed. These structures were found to be strongly functionally connected.

Our results indicate that functional connectivity measurements are feasible in tinnitus patients and controls, without relying on prior knowledge about stimulus perception. Such measurements can reveal the classical pathway, including sub-cortical processing centers. In addition, we showed that correlated activity in the nonclassical pathway can be detected, which is very difficult to achieve by conventional fMRI techniques.

Therefore, our methods provide a promising alternative to complement standard fMRI analyses in the assessment of tinnitus related abnormalities. Additional measurements will be carried out to further validate our methods, and to assess differences between tinnitus subjects and controls.