Extracting the fingerprints of sequences of random rhythmic auditory stimuli from electrophysiological data

It has been classically conjectured that the brain assigns probabilistic models to sequences of stimuli. An important issue associated with this conjecture is the identification of the classes of models used by the brain to perform this task. We address this issue by using a new clustering procedure for sets of electroencephalographic (EEG) data recorded from participants exposed to a sequence of auditory stimuli generated by a stochastic chain. This clustering procedure indicates that the brain uses the recurrent occurrences of a regular auditory stimulus in order to build a model.Author summary: A classical conjecture is that the brain is constantly estimating regularities from sequences of events to be able to properly act upon the environment. We assume that, by doing statistics, the brain chooses a model from a class of possible models. Which class of models is used by the brain to encode sequences of events? We used an algorithm to generate a sequence of hand claps step by step reproducing a samba-like rhythm. These sequences were generated with stochasticity, where some auditory events were omitted with small probability. We retrieved the regularities of these random sequences of stimuli from EEG data recorded as the participants listened to the samba rhythm. To extract the information encoded in the EEG data we introduced a novel procedure for clustering sets of functional data by their relevant statistical features. The clusters obtained from the experimental data show that the strong beat of the rhythmic structure is used by the brain to encode the sequence. The strong beat has a remarkable property of separating the sequence into smaller independent blocks. This leads to a natural and economical explanation on how the brain organises the sequence in order to estimate the next event.