University of Cambridge > Talks.cam > Computational Neuroscience > Dopamine as prediction error in active inference

Dopamine as prediction error in active inference

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact Yul Kang.

Dopaminergic neurons are known to encode reward prediction error, which according to the reinforcement learning theory, enables learning about results of actions. However, dopaminergic neurons also play an important role in action planning, as evident from movement difficulties in Parkinson’s disease. This talk will propose that a more complete description of dopamine function can be obtained by integrating reinforcement learning theory with ideas from active inference. It will suggest that a fraction of dopaminergic neurons encodes reward prediction error, but an expectation of the reward only arises from formulating a plan to obtain it. During action planning, these dopaminergic neurons provide feedback on whether the current motor plan is sufficient to obtain the available reward, and they facilitate action planning until a suitable plan is found. This framework accounts for a wide range of experimental data including diversity of dopaminergic responses, effects of dopamine depletion on behaviour, and it makes experimental predictions.

This talk is part of the Computational Neuroscience series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.

 

© 2006-2024 Talks.cam, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity