How the brain detects patterns in sound sequences

This project employs state of the art neuroimaging (EEG, MEG and fMRI) and behavioural experimentation (psychophysics) to reveal how human listeners discover patterns and statistical regularities in rapidly unfolding sound sequences. Accumulating experimental evidence suggests that the brain is sensitive to statistical regularities in sensory input, at multiple time scales, and that this sensitivity plays a key role in our ability to understand, efficiently interact with, and survive in the environment. However, a key question - the process through which patterns are detected in the first place - has largely eluded investigation. My proposal focuses on this crucial missing link. I propose a paradigm, recently developed and validated in my laboratory, that allows us to observe these processes, with great detail, within the auditory system. The method, based on measuring behavioural and time-locked brain responses to rapid tone-pip sequences governed by specifically controlled rules aims to uncover (1) how the brain discovers patterns in sound sequences, (2) which neural mechanisms are involved, (3) to what degree the process is automatic or susceptible to attentional state and behavioural goals of the listener. The results will reveal an important missing link in understanding audition, and perception more generally, and inform the current debate in systems neuroscience surrounding sensitivity to input statistics and predictive coding. The project draws upon, and will extend, my expertise in psychophysics, electrophysiology, and brain imaging of auditory function. It is backed by the strong multidisciplinary scientific and clinical environment of the UCL Ear Institute and the Wellcome Trust Centre for Neuroimaging.

This project employs state of the art neuroimaging (EEG, MEG and fMRI) and behavioural experimentation (psychophysics) to reveal how human listeners discover patterns and statistical regularities in rapidly unfolding sound sequences. Accumulating experimental evidence suggests that the brain is sensitive to statistical regularities in sensory input, at multiple time scales, and that this sensitivity plays a key role in our ability to understand, efficiently interact with, and survive in the environment. However, a key question - the process through which patterns are detected in the first place - has largely eluded investigation. My proposal focuses on this crucial missing link. I propose a paradigm, recently developed and validated in my laboratory, that allows us to observe these processes, with great detail, within the auditory system. The method, based on measuring behavioural and time-locked brain responses to rapid tone-pip sequences governed by specifically controlled rules aims to uncover (1) how the brain discovers patterns in sound sequences, (2) which neural mechanisms are involved, (3) to what degree the process is automatic or susceptible to attentional state and behavioural goals of the listener. The results will reveal an important missing link in understanding audition, and perception more generally, and inform the current debate in systems neuroscience surrounding sensitivity to input statistics and predictive coding. The project draws upon, and will extend, my expertise in psychophysics, electrophysiology, and brain imaging of auditory function. It is backed by the strong multidisciplinary scientific and clinical environment of the UCL Ear Institute and the Wellcome Trust Centre for Neuroimaging.

The present project is classified as 'basic research' with the goal of understanding how normal hearing listeners detect the emergence of patterns and statistical regularities within rapid, stochastic sound sequences. The major implications of this work are in the academic domain, however it also has the potential of generating impact in the clinical, industry and arts spheres.


(1) Sensitivity to patterning is a fundamental aspect of listening. Understanding the range of sound patterns to which listeners are particularly sensitive, and the neural signature corresponding to their detection, will inform work efforts in all aspects of auditory-related technologies. This including the design of hearing aids and of brain-computer interfaces and any technology that produces sound for the purpose of conveying information, providing feedback or creating ambience. Understanding sensitivity to sound patterns could inform the design of these systems for example by programming them so that they produce events that listeners are particularly sensitive to. This could benefit individuals and the commercial private sector.

(2) Our results could also find an interesting outlet in the arts field including sound-scape design and composition.


(3) Our experimental paradigms, designed to study acoustic pattern processing in the normal (healthy) brain may be used to understand and characterize deficits in clinical populations (e.g. 'Auditory processing disorder', Schizophrenia) which exhibit deficits in complex sensory processing. This will benefit individuals, health professionals as well as NHS policy makers.


(4) Analysis tools developed in the course of this project might be of use to brain-machine applications thus benefiting government agencies, and the commercial private sector.