Prediction and memory of sounds: Evidence from brain studies across species
The brain is thought to generate internal predictions, based on the memory of past stimulation, to optimise behaviour. However, it is unclear to what extent these predictions are modulated by other top-down factors such as attention and task demands, and whether predictions of different sensory features are mediated by the same neural mechanisms.
In this talk Dr. Ryszard Aukstulewicz will present results of several studies combining human and rodent electrophysiology with computational modelling to identify the neural mechanisms of sensory predictions and their interactions with other cognitive factors. First, in a series of non-invasive studies using MEG/EEG and direct recordings from humans using ECoG, analysis of behavioural and neural data showed that the effects of predictions are not automatic but are modulated by their contextual relevance. Second, computational modelling of the data suggested that these modulations could be linked to specific candidate mechanisms, including gain control in sensory regions. Finally, in a series of studies using invasive recordings in anaesthetised rodents, neural representations related to stimulus memory and predictions could be decoded from auditory cortical activity, shedding light on the evolutionary conservation of predictive processing across species.
About the speaker
Ryszard Auksztulewicz conducts empirical and theoretical work at the interface of cognitive, computational, and systems neuroscience, focusing primarily on the neural mechanisms of predictive coding and their modulation by cognitive factors. He is currently funded by the European Commission’s Marie Skłodowska-Curie Global Fellowship for his work on neural mechanisms of prediction signalling, and is hosted by the Neuroscience Department of the MPI for Empirical Aesthetics, Frankfurt am Main (working closely with Prof. Lucia Melloni and Prof. David Poeppel), and the Department of Neuroscience, City University of Hong Kong (Prof. Jan Schnupp). Previously, Ryszard has worked in the world’s leading cognitive and computational neuroscience groups at University College London (Prof. Karl Friston) and Oxford University (Prof. Kia Nobre), spearheading empirical and modelling studies of the neural mechanisms of prediction error signalling. Most of his previous and ongoing work is directly related to predictive coding and auditory mismatch signalling in the brain.