Adaptive learning under expected and unexpected uncertainty

Nature reviews. Neuroscience, Journal Year: 2019, Volume and Issue: 20(10), P. 635 - 644

Published: May 30, 2019

Language: Английский

---

Reinforcement learning in artificial and biological systems

Nature Machine Intelligence, Journal Year: 2019, Volume and Issue: 1(3), P. 133 - 143

Published: March 4, 2019

Language: Английский

---

Cross Talk: The Microbiota and Neurodevelopmental Disorders

Frontiers in Neuroscience, Journal Year: 2017, Volume and Issue: 11

Published: Sept. 15, 2017

Humans evolved within a microbial ecosystem resulting in an interlinked physiology. The gut microbiota can signal to the brain via immune system, vagus nerve or other host-microbe interactions facilitated by hormones, regulation of tryptophan metabolism and metabolites such as short chain fatty acids (SCFA), influence development, function behaviour. Emerging evidence suggests that may play role shaping cognitive networks encompassing emotional social domains neurodevelopmental disorders. Drawing upon preclinical clinical evidence, we review potential origins development related Autism spectrum disorders (ASD) Schizophrenia. Small preliminary studies have demonstrated alterations both ASD Schizophrenia compared healthy controls. However, await further mechanistic insights, together with large scale longitudinal trials, encompass systems level dimensional approach, investigate whether promising pre-clinical initial findings lead relevance.

Language: Английский

---

Reinforcement learning in robotic applications: a comprehensive survey

Artificial Intelligence Review, Journal Year: 2021, Volume and Issue: 55(2), P. 945 - 990

Published: April 20, 2021

Language: Английский

---

Multidimensional processing in the amygdala

Nature reviews. Neuroscience, Journal Year: 2020, Volume and Issue: 21(10), P. 565 - 575

Published: Aug. 24, 2020

Language: Английский

---

A model for learning based on the joint estimation of stochasticity and volatility

Nature Communications, Journal Year: 2021, Volume and Issue: 12(1)

Published: Nov. 15, 2021

Abstract Previous research has stressed the importance of uncertainty for controlling speed learning, and how such control depends on learner inferring noise properties environment, especially volatility: change. However, learning rates are jointly determined by comparison between volatility a second factor, moment-to-moment stochasticity. Yet much previous focused simplified cases corresponding to estimation either factor alone. Here, we introduce model, in which both factors learned simultaneously from experience, use model simulate human animal data across many seemingly disparate neuroscientific behavioral phenomena. By considering full problem joint estimation, highlight set previously unappreciated issues, arising mutual interdependence inference about This complicates enriches interpretation results, as pathological individuals with anxiety following amygdala damage.

Language: Английский

---

Goal-directed learning in adolescence: neurocognitive development and contextual influences

Nature reviews. Neuroscience, Journal Year: 2024, Volume and Issue: 25(3), P. 176 - 194

Published: Jan. 23, 2024

Language: Английский

---

Subcortical Substrates of Explore-Exploit Decisions in Primates

Neuron, Journal Year: 2019, Volume and Issue: 103(3), P. 533 - 545.e5

Published: June 10, 2019

Language: Английский

---

Prefrontal Cortex Predicts State Switches during Reversal Learning

Neuron, Journal Year: 2020, Volume and Issue: 106(6), P. 1044 - 1054.e4

Published: April 20, 2020

Language: Английский

---

Animal affect and decision-making

Neuroscience & Biobehavioral Reviews, Journal Year: 2020, Volume and Issue: 112, P. 144 - 163

Published: Jan. 25, 2020

The scientific study of animal affect (emotion) is an area growing interest. Whilst research on mechanism and causation has predominated, the function less advanced. This not due to a lack hypotheses; in both humans animals, affective states are frequently proposed play pivotal role coordinating adaptive responses decisions. However, exactly how they might do this (what processes implement function) often left rather vague. Here we propose framework for integrating decision-making that couched modern decision theory employs operational definition aligns with dimensional concepts core renders empirically tractable. We develop model affect, including short-term (emotion-like) longer-term (mood-like) states, influence via label options, predictions, outcomes which correspond similar schema links between human emotion decision-making. Our generalisable across species generates questions future research.

Language: Английский

---