Brain-Rhythm-based Inference (BRyBI) for time-scale invariant speech processing

Rhythms stretching across multiple interacting frequencies and spatial scales are ubiquitous in brain activity during complex cognitive tasks. Yet their functional significance is hotly debated between a structural epiphenomenon of brain activity and mechanisms implementing computations. Speech processing, with its temporal cadence and multi-scale of syntactic invariants (syllables, words), is a paradigmatic example where rhythms have been proposed to play a key role [1], with a speech-modulated hierarchical structure of intercoupled cortical oscillations correlating with successful comprehension. Experiments show that speech recognition remains largely intact when compressed up to a certain temporal factor [2] and the re-spacing chunks of incomprehensible compressed speech with silences recovers comprehension. Previous models proposed that theta-gamma interactions enable syllable parsing [3,4]. However, this did not resolve questions about word recognition and, notably, the observed role of delta rhythm in top-down information flow. To address the above questions, we propose an inference model (BRyBI) that incorporates a wide range of brain-rhythm data mechanistically and accounts for time-invariant word recognition. In this model, the hierarchically arranged interacting rhythms actively maintain top-down and bottom-up information flow during the inference process: theta-gamma interactions predict and parse phonemes/syllable sequences, while the delta-rhythm adaptively generates the inferred word context. We show that word recognition degrades when the speed of words and syllables is compressed beyond the delta and theta rhythms, respectively. The top-down contextual delta-implemented context allows us to explain why re-spacing compressed speech recovers comprehension. Our model further predicts that delta-implemented word context allows for syllable parsing without a precise locking of the theta-rhythmic activity. In general, we propose a potential resolution to the debated role of brain oscillations as mechanisms to control bottom-up and top-down flows of information to contextualize sensory inference processes.