Search bioRxiv⌕ Search

Biology subjects

Ellwood, I. T.

Publications and source records attributed to Ellwood, I. T..

2 recordsLinked to original sources

Prefrontal cortex dopamine responds to the total valence of stimuli

The prefrontal cortex (PFC) dopamine system plays an essential role in cognitive flexibility, working memory and psychiatric disease, but determining the conditions under which dopamine in the PFC is released remains an open problem. Both rewarding and aversive stimuli have been found to trigger release, but studies have disagreed on whether the valence of a stimulus or other variables like novelty and salience are the most important. Here we report on recordings of dopamine-dependent fluorescence using a high-sensitivity dopamine indicator. We deliver an array of rewarding, aversive and mixed valence stimuli, as well as stimuli without any obvious valence. We observe that stimuli without valence, as well as the omission of expected stimuli, do not lead to large changes in fluorescence, even when these stimuli and omissions are both novel and engaging. In contrast, both rewarding and aversive stimuli lead to increases in fluorescence, with the most rewarding and most aversive stimuli leading to the largest increases. We test the effect of adding an aversive component to a rewarding stimulus and find that the increases in fluorescence are consistent with a summation of the rewarding and aversive components. We propose that dopamine release in the PFC responds to the total valence of a stimulus, in contrast with the traditional view of basal ganglia dopamine release that depends on the net valence.

neuroscience↗

Short-term Hebbian learning can implementtransformer-like attention

Transformers have revolutionized machine learning models of language and vision, but their connection with neuroscience remains tenuous. Built from attention layers, they require a mass comparison of queries and keys that is difficult to perform using traditional neural circuits. Here, we show that neurons can implement attention-like computations using short-term, Hebbian synaptic potentiation. We call our mechanism the match-and-control principle and it proposes that when activity in an axon is synchronous, or matched, with the somatic activity of a neuron that it synapses onto, the synapse can be briefly strongly potentiated, allowing the axon to take over, or control, the activity of the downstream neuron for a short time. In our scheme, the keys and queries are represented as spike trains and comparisons between the two are performed in individual spines allowing for hundreds of key comparisons per query and roughly as many keys and queries as there are neurons in the network.

neuroscience↗