Dataset Information

Fast adaptation to rule switching using neuronal surprise.

ABSTRACT: In humans and animals, surprise is a physiological reaction to an unexpected event, but how surprise can be linked to plausible models of neuronal activity is an open problem. We propose a self-supervised spiking neural network model where a surprise signal is extracted from an increase in neural activity after an imbalance of excitation and inhibition. The surprise signal modulates synaptic plasticity via a three-factor learning rule which increases plasticity at moments of surprise. The surprise signal remains small when transitions between sensory events follow a previously learned rule but increases immediately after rule switching. In a spiking network with several modules, previously learned rules are protected against overwriting, as long as the number of modules is larger than the total number of rules-making a step towards solving the stability-plasticity dilemma in neuroscience. Our model relates the subjective notion of surprise to specific predictions on the circuit level.

SUBMITTER: Barry MLLR

PROVIDER: S-EPMC10906910 | biostudies-literature | 2024 Feb

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Fast adaptation to rule switching using neuronal surprise.

Barry Martin L L R MLLR Gerstner Wulfram W

PLoS computational biology 20240220 2

In humans and animals, surprise is a physiological reaction to an unexpected event, but how surprise can be linked to plausible models of neuronal activity is an open problem. We propose a self-supervised spiking neural network model where a surprise signal is extracted from an increase in neural activity after an imbalance of excitation and inhibition. The surprise signal modulates synaptic plasticity via a three-factor learning rule which increases plasticity at moments of surprise. The surpri ...[more]

PMID: 38377112

Similar Datasets

Project description:Auditory repetition suppression and omission activation are opposite neural phenomena and manifestations of principles of predictive processing. Repetition suppression describes the temporal decrease in neural activity when a stimulus is constant or repeated in an expected temporal fashion; omission activity is the transient increase in neural activity when a stimulus is temporarily and unexpectedly absent. The temporal, repetitive nature of musical rhythms is ideal for investigating these phenomena. During an fMRI session, 10 healthy participants underwent scanning while listening to musical rhythms with two levels of metric complexity, and with beat omissions with different positional complexity. Participants first listened to 16-s-long presentations of continuous rhythms, before listening to a longer continuous presentation with beat omissions quasi-randomly introduced. We found deactivation in bilateral superior temporal gyri during the repeated presentation of the normal, unaltered rhythmic stimulus, with more suppression of activity in the left hemisphere. Omission activation of bilateral middle temporal gyri was right lateralized. Persistent activity was found in areas including the supplementary motor area, caudate nucleus, anterior insula, frontal areas, and middle and posterior cingulate cortex, not overlapping with either listening, suppression, or omission activation. This suggests that the areas are perhaps specialized for working memory maintenance. We found no effect of metric complexity for either the normal presentation or omissions, but we found evidence for a small effect of omission position-at an uncorrected threshold-where omissions in the more metrical salient position, i.e., the first position in the bar, showed higher activation in anterior cingulate/medial superior frontal gyrus, compared to omissions in the less salient position, in line with the role of the anterior cingulate cortex for saliency detection. The results are consistent with findings in our previous studies on Parkinson's disease, but are put into a bigger theoretical frameset.

Dataset Information

Fast adaptation to rule switching using neuronal surprise.

Publications

Fast adaptation to rule switching using neuronal surprise.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets