How 25 Years of Behavioral Neuroscience Shaped a New Digital Therapy for Tinnitus

Tinnitus, the perception of sound without an external source, affects hundreds of millions of people worldwide and remains one of the most stubborn problems in auditory medicine. There is no curative treatment, and existing therapies help some people some of the time. A new review published in Frontiers in Audiology and Otology by Grant D. Searchfield of the University of Auckland and TrueSilence Therapeutics lays out the behavioral neuroscience research used to develop the QCore™ Tinnitus Therapy Platform, a smartphone-delivered digital therapy built on an ecological model of tinnitus and adaptation-level theory. The review is unusual in its candor: it maps the scientific reasoning behind each therapy mode while openly acknowledging that much of the supporting evidence is indirect and that rigorous independent evaluation is still needed.

The starting point for the therapy is a recognition that tinnitus is not a single lesion but a distributed pattern of neuroplastic change. The review describes how tinnitus characteristics, emotional impact, and responses to sound vary enormously between individuals, and how the disorder arises from complex interactions between sensory and affective neurophysiology. Proposed mechanisms include central gain control, attentional capture, failed noise cancellation, and prediction errors. In this framework, hierarchically organized levels of auditory processing feed errors forward, producing sensitization and altered connectivity across multiple neural hubs. Once the tinnitus signal is detected, the absence of a real-world sound source forces the brain to allocate extra attention and cognitive load to clarify the perception, which is thought to sensitize a core tinnitus network spanning the ventromedial prefrontal cortex, inferior parietal cortex, parahippocampal cortex, and auditory cortex. Sensory sensitivity then propagates into non-sensory affective networks, producing what clinicians recognize as tinnitus disorder.

The theoretical heart of the QCore™ platform combines two ideas. The first is an ecological model of tinnitus, which treats the brain as parsing tinnitus as if it were an auditory object embedded in a real-world scene. Sounds perceived as near are prioritized for attention, but tinnitus is internally generated and lacks confirmation from other senses, an incongruence with learned expectations that makes the percept more likely to attract attention and be appraised as significant. Quiet environments, low task load, stress, fatigue, and personal meanings attached to the sound all bias the competition between tinnitus and external auditory objects in tinnitus’s favor. In predictive-coding terms, tinnitus behaves like a learned internal prediction that is given too much weight. The second idea is adaptation-level theory, which models how the brain judges tinnitus against a shifting baseline shaped by external sound, internal state, and attentional set. Monotonous soundscapes, high arousal, and low cognitive load are believed to lower the adaptation level, allowing central gain to rise and increase tinnitus detectability, while enriched sound environments, lowered arousal, and engaged attention raise the baseline and reduce the relative salience of the phantom sound.

From these principles the platform defines tinnitus magnitude as the product of loudness, annoyance, and awareness, and it targets the three levers through three therapy modes: Relief, Relax, and Retrain. The Relief mode is designed to reduce perceived loudness and detectability by providing continuous external sound that competes with the tinnitus signal. It employs both energetic masking, in which the masking sound physically overlaps the tinnitus in time and frequency, and informational masking, in which the tinnitus remains audible but becomes difficult to segregate or attend to because of perceptual and cognitive interference. A key tool is Threshold Adjusted Noise, a sound tailored to individual hearing thresholds and masking levels that provides additional energy near the tinnitus pitch. Informational masking options include spatially co-located external sound sources aligned with the perceived location of the tinnitus, which the review reports has been more efficient at reducing tinnitus than excitatory masking alone, because spatial disparity cues that would otherwise help the brain segregate the tinnitus as a distinct object are minimized.

The physiological evidence for masking-based relief is among the more striking parts of the review. Responders to masking paradigms typically show altered EEG activity in insular, inferior parietal, thalamic, temporal, and cingulo-frontal regions, consistent with a shift away from sustained, effortful frontal control toward more efficient posterior sensory integration. Spiking neural network models have demonstrated that baseline EEG features can distinguish masking responders from non-responders with high accuracy; responders are characterized by elevated occipitoparietal connectivity at baseline, which normalizes after therapy, whereas non-responders retain their pre-therapy dynamics. Residual inhibition studies add band-limited power increases, most consistently in the alpha range and sometimes gamma, during tinnitus suppression, indicating a regulated cortical state rather than mere distraction. The review also notes caveats: neural contrast control, in which the auditory system boosts gain for low-contrast signals, may counteract benefits in some users and could explain declining masking effectiveness over time, and sound class matters, with broadband noise producing larger short-term loudness reductions while natural sounds may offer affective benefits that support longer-horizon adaptation.

The Relax mode addresses the emotional and autonomic side of tinnitus. Through calming soundscapes, guided visualization, mindfulness, and structured relaxation, it aims to shift the nervous system from hypervigilance toward safety, reducing the emotional gain applied to the tinnitus signal and weakening the loop in which distress amplifies salience, salience captures attention, and attention further entrenches distress. The acoustic design is deliberate: slower tempos, predictable envelopes, and natural soundscapes are chosen to promote parasympathetic activity and perceived safety, and preferences for sounds perceived as receding rather than approaching underscore the role of predictability in lowering arousal. Experimental work cited in the review shows that negative mood or unpleasant emotional sounds can increase perceived tinnitus loudness, whereas positive or neutral states reduce it, consistent with adaptation-level theory’s emphasis on the emotional-autonomic component of magnitude estimation.

Neuroimaging evidence suggests that different sound classes act through complementary but distinct mechanisms. In one study, auditory categorization training that morphed a tinnitus avatar into a nature sound over ten minutes increased activity in perceptual and parietal regions while reducing activity in the dorsolateral prefrontal cortex and gamma-band activity in auditory cortex, changes associated with tinnitus distress and intensity. White noise, by contrast, produced more diffuse cortical activation with recurrent involvement of the angular gyrus and parahippocampal cortex, possibly reflecting attentional diversion and modulation of auditory memory networks. Sleep research reinforces the Relax rationale: in a randomized crossover study, a 30-minute pre-sleep positive visualization, with or without overnight natural-sound reinforcement, produced large reductions in sleep-onset latency and improved self-reported attentional control in people with tinnitus-related sleep disturbance.

The Retrain mode targets attention and perceptual organization. Because tinnitus lacks visual confirmation and violates expectations for real sounds, it recruits attention and conflicts with memory-based predictions, becoming an over-dominant auditory object. The mode uses structured listening activities, auditory object identification and localization tasks, and attention-based exercises to weaken habitual self-monitoring and reclassify tinnitus as a low-priority, low-salience perception. The evidence base includes a randomized controlled trial of a 20-day attention-training game that produced clinically significant reductions on the Tinnitus Functional Index, with gains correlating with better mixed auditory-visual attention and shorter N1 latencies to non-tinnitus targets. Multisensory attention training showed smaller but significant TFI reductions, accompanied by measurable changes in brain connectivity, including enhanced bilateral coupling between the left and right posterior intraparietal sulcus and strengthened coupling within a right-lateralized network linking the frontal eye field, temporoparietal junction, and superior temporal sulcus, alongside reduced coupling between the right middle frontal gyrus and primary auditory cortex, consistent with reduced executive control over early auditory processing.

Personalization is the platform’s other pillar. Assessment combines goal-anchored planning using the Client Oriented Scale of Improvement in Tinnitus, which elicits patient-defined priorities such as sleep and control, with mechanism-informed profiling that integrates hearing status, tinnitus psychoacoustics including spatial characteristics, and masking profiles. The system recommends a total of 60 minutes of therapy per day, allocated adaptively across the three modes, with early emphasis on Relief and Relax before progressively increasing Retrain. Counseling is embedded as short, theory-informed prompts delivered automatically rather than generated by AI or written individually by therapists. A controlled trial of the prototype, then labeled UpSilent, compared it against a popular sound-therapy app: participants using the prototype showed clinically meaningful reductions in TFI scores at both 6 and 12 weeks, with early improvements in annoyance and the ability to ignore tinnitus, followed by later reductions in tinnitus strength, a pattern the author interprets as early limbic and attentional modulation preceding slower sensory-gain recalibration.

The review is candid about its limitations. Much of the cited research was conducted by the developer and collaborators, who hold financial interests in the platform, and the author acknowledges the potential for investigator bias, publication bias, and placebo and expectancy effects, which are particularly potent in a condition measured largely by patient-reported outcomes. Not all individuals benefit equally, and factors such as tinnitus duration, psychological comorbidities, and hearing-loss profiles may drive non-response. The review concludes that convergent behavioral and neuroscientific evidence supports the plausibility of the multimodal approach, but that larger, independent, comparative trials with active controls, longitudinal follow-up, and mechanistic biomarkers are essential to establish efficacy, durability, and scalability. For now, the platform stands as a testable hypothesis rendered in software: a structured attempt to turn decades of behavioral neuroscience into a clinical pathway for one of medicine’s most persistent phantom perceptions.

Subject of Research: Behavioral neuroscience research underlying the development of the QCore digital tinnitus therapy platform

Article Title: A review of the applied behavioral neuroscience research used to develop the QCore™ tinnitus therapy platform

Article References: Searchfield, G. D. (2026). A review of the applied behavioral neuroscience research used to develop the QCore™ tinnitus therapy platform. Frontiers in Audiology and Otology, 4, Article 1860026. https://doi.org/10.3389/fauot.2026.1860026

Image Credits: AI Generated

DOI: 10.3389/fauot.2026.1860026

Keywords: tinnitus, behavioral neuroscience, neuroplasticity, sound therapy, adaptation-level theory, attention training, masking, digital therapeutics, EEG, counseling, auditory perception, clinical guidelines