TIDE Project: Biomarker Discovery for Chronic Tinnitus Diagnosis

The tinnitus detect (TIDE) consortium has been designed to identify and validate a biomarker for the presence and intensity of tinnitus. For this purpose, two test paradigms are combined which are derived from the most recent models of tinnitus and which have shown promise in pilot studies. A first approach is based on a method used in animals to objectively assess tinnitus by gap detection, namely Gap prepulse inhibition of the acoustic startle (GPIAS). The GPIAS paradigm is inspired from the combination of both gap detection and pre-pulse inhibition (PPI), used for the assessment of temporal sensory processing in both animals and humans. The basic principle of PPI relies on the ability of a weak lead stimulus (a prepulse or a silent gap presented in a carrier background) to inhibit a startling effect of a following, more intense, abrupt stimulus. These paradigms have been used to assess the automatic or preconscious inhibition of the motor reflex response that occurs in healthy subjects. Because GPIAS is regulated at the level of the auditory cortex, cortical responses may provide a more accurate measure of inhibition than motor reflex responses, such as the startle response in animals or the eye blinking in humans. The hypothesis is that individuals with tinnitus would have impaired inhibition of cortical evoked responses to sound pulses when preceded by a silent gap, in comparison to non-tinnitus individuals, due to the ongoing tinnitus percept. A second approach is based on the concept that tinnitus occurrence is related to altered processing of prediction errors. This concept is supported by empirical evidence demonstrating that people with tinnitus exhibit a more pronounced electrophysiological response to a mismatch between predictions (the expected sound based on the auditory training paradigm) and perceptions as compared to non-tinnitus individuals. This sensitivity to prediction errors (the difference between prediction and input) is associated with how loud patients perceive their tinnitus, independent of co-occurring hyperacusis and hearing loss. For example, the amplitude of the mismatch negativity (MMN) positively correlates with subjectively reported tinnitus loudness. Thus, the MMN might be a biomarker for how loud tinnitus patients perceive their tinnitus. That is, people with a more pronounced electrophysiological response to a prediction error perceive a louder phantom sound. Using another paradigm, a group, independent of us, confirmed the importance of the MMN for tinnitus detection. Furthermore, an increase in amplitude and a delay in latency for the late positive evoked brain response (P300) has been demonstrated in tinnitus patients using both auditory and visual oddball paradigms. Both the MMN and P300 are conceived as measures of prediction errors identified at respectively the sensory cortex and higher levels (i.e., frontal-parietal cortex). The P300 is thought to reflect processes involved in stimulus evaluation or categorization; whether the stimulus is behaviourally relevant or not. Therefore, the aim of the current approach is to validate that both the MMN and the P300, using an auditory oddball paradigm with omission, can be used as biomarkers for tinnitus loudness and presence, respectively, (i.e., cross-validation). We propose to investigate these two paradigms in a large international sample of tinnitus patients and controls in order to determine the sensitivity of GPIAS and the oddball paradigm in diagnosing tinnitus in humans.