An integrative model of auditory phantom perception: tinnitus as a unified percept of interacting separable subnetworks.

Summary

Tinnitus is framed as an auditory phantom: a persistent conscious percept of a salient memory trace, externally attributed, with no sound source. It is experienced as one unified percept yet binds separable characteristics (loudness, sidedness, type, distress). This framework integrates deafferentation-based neurophysiological models with the dysfunctional noise-cancelling model, and proposes a 'tinnitus core', the minimal set of brain areas that must be jointly active for tinnitus to be consciously perceived, stripped of its affective components. The unified percept is cast as an emergent property of multiple, parallel, dynamically changing and partially overlapping subnetworks, each with its own oscillatory pattern and connectivity signature, communicating at hubs at discrete frequencies.

Key findings

• Proposes a 'tinnitus core' subnetwork: the minimal jointly-active areas needed for the sound to reach consciousness, separate from distress.
• The unified percept is an emergent property of multiple parallel, partially overlapping subnetworks.
• Each subnetwork carries its own oscillatory signature; they communicate at shared hubs at discrete frequencies.
• Explicitly integrates deafferentation-based models with the dysfunctional noise-cancelling model.