Itzuli Aurreko ekintzak: Alejandro Tabas. Predictive processing as a synergistic cortico-subcortical computation

- BCBL zoom room 2 only (online talk)

What: Predictive processing as a synergistic cortico-subcortical computation

Where: BCBL Zoom Room # 2 (online talk) (If you would like to attend to this meeting reserve at

Who: Alejandro Tabas (PhD, Research Associate, Cambridge University, UK.

When:  Thursday,  Feb 16th at 12:00 PM noon.

Speech signals are noisy, fast-varying, and often ambiguous. However, humans are remarkably competent at communication in a wild variety of environments. Converging evidence seems to indicate that predictive processing, a theory that explains perception as inference, may be the main strategy the brain uses to excel at this task. Understanding the neural mechanisms underlying its implementation is crucial to understand perceptual function and dysfunction: dyslexia, autism and psychosis have all been linked to an impaired handling of predictions in the brain.

Predictive processing has been implicitly assumed to be exclusive of the cerebral cortex. This assumption was based on that only populations in the cerebral cortex, with long integration time constants, could construct robust and broad-viewed predictions on, for instance, the semantic meaning of the next sentence in a conversation. However, only subcortical nuclei, with their much shorter timescales, have the temporal resolution to test and generate predictions on short and fast-varying elements like phonemes. Motivated by this reasoning, we have recently tested and shown that the auditory midbrain and thalamus do partake on predictive processing.

In my talk I will first summarise these previous findings and argue that the inclusion of subcortical stations to the predictive processing network requires a profound and exciting reformulation of the theory. I will introduce my main hypotheses on what this theory should look like, and outline a hybrid experimental-computational research plan for its development. If successful, my research plan will result in a mechanistic understanding of how cortical and subcortical stages work together to proficiently process fast, complex, and ambiguous sensory inputs like speech, providing for for a cortical-subcortical integrative understanding of perceptual function and dysfunction.