Skip to content
Computational Neurology Lab
Nuffield Department of Clinical Neurosciences, Oxford
  • Home
  • Research
  • People
  • News
  • Data / Code
  • Contact

Month: May 2021

Hunger increases reinforcement learning rate

May 26, 2021 admin Leave a comment

Sometimes we plan ahead, thinking about future consequences of our actions. Other times, we select actions based only on their…

Continue Reading →

Recent Posts

  • Neurons that rapidly change their selectivity February 28, 2023
  • What does the claustrum do? February 9, 2022
  • Hunger increases reinforcement learning rate May 26, 2021
  • Frontal brain damage makes decisions less biased February 13, 2021
  • Drugs affect subtypes of Parkinson’s disease differently December 26, 2020
  • Parkinson’s medication has opposite effects on two kinds of motivation November 29, 2020
  • Are working memory and visual search windows into the same neural process? October 1, 2020
  • Pupil size betrays contents of working memory October 25, 2019
  • Dopamine can make our working memory more or less flexible September 19, 2019
  • Neural model of working memory July 29, 2019

Recent Comments

    Archives

    • February 2023
    • February 2022
    • May 2021
    • February 2021
    • December 2020
    • November 2020
    • October 2020
    • October 2019
    • September 2019
    • July 2019
    • April 2019

    Recent work

    • Neurons that rapidly change their selectivity
      The activity of neurons carries information because they become active only in particular situations. This is called ‘selectivity’, and it allows a group of neurons to signal the state of the world. We may remember things either by keeping neurons active, or by changing their selectivities. However, we haven’t really worked out how changing a neuron’s selectivity works, to help us perform tasks. Here in Bocincova et al. PNAS (2022) (free) we find that neurons in […]
    • What does the claustrum do?
      Here with colleagues from physiology, we review the clinical effects of damage to the claustrum in Atilgan et al. Brain (2022). The claustrum is a thin sheet of neurons in the frontal lobe. Very few reported cases have isolated claustrum damage. In those who did, the findings don’t clearly reflect what you might expect, given the known fMRI activations and connections of the claustrum. […]
    • Hunger increases reinforcement learning rate
      Sometimes we plan ahead, thinking about future consequences of our actions. Other times, we select actions based only on their immediate reward associations. Planning ahead is crucial to staying healthy, but might be affected by motivation. We asked whether hunger affects planned vs directly reinforced action (van Swieten, Bogacz & Manohar Cogn. Aff. Beh. Neurosci. 2021). We found that people learned quicker about action values when they were hungry. Hunger didn’t affect planning. […]
    • Frontal brain damage makes decisions less biased
      Learning from reinforcement is a classic way to study how brain areas contribute to adaptive behaviour. The most frontal parts of the brain probably contribute at a very high, abstract level. In this study (Manohar et al. Cortex 2021), we asked how confident people are in what they have learned. Underside of a human brain showing a part of the ventromedial prefrontal cortex. Healthy people are biased by previous choices, and by competing information, when making these […]
    • Drugs affect subtypes of Parkinson’s disease differently
      In this commentary, I discuss the implications of new work from Hanneke Den Ouden’s lab. The authors subdivided Parkinson’s disease patients into those with and without tremor, and found that learning was affected by dopamine in opposite directions in the two groups! […]
    • Parkinson’s medication has opposite effects on two kinds of motivation
      Patients with Parkinson’s disease lack the brain chemical dopamine. Dopamine is thought to signal upcoming rewards, and this might explain why patients on treatment can develop impulse control disorders. My lab is studying two different ways to motivate people. One way is to reward or punish them based on how well they do — like performance-related pay. The other way is to promise a guaranteed reward, which also tends to keep people motivated (even though they don’t have to). […]
    • Are working memory and visual search windows into the same neural process?
      When you recall an item from memory, a prompt usually brings associated parts of that item into mind. Could this process be the same thing that occurs when you search for a visual target? We tested a neural model designed to perform working memory tasks, to see if it could also perform visual search. The model retrieves information when a partial cue re-activates a pattern of neurons by associative pattern completion. This same process could occur when we look for an item that we have in mind: […]