Published: Jan. 1, 2025
Language: Английский
Nature Neuroscience, Journal Year: 2022, Volume and Issue: 25(10), P. 1257 - 1272
Published: Sept. 26, 2022
Language: Английский
Citations
143
Trends in Cognitive Sciences, Journal Year: 2021, Volume and Issue: 25(9), P. 730 - 743
Published: June 16, 2021
Language: Английский
Citations
134
Artificial Intelligence in Medicine, Journal Year: 2024, Volume and Issue: 151, P. 102861 - 102861
Published: March 30, 2024
Language: Английский
Citations
87
Cell, Journal Year: 2023, Volume and Issue: 186(22), P. 4885 - 4897.e14
Published: Oct. 1, 2023
Human reasoning depends on reusing pieces of information by putting them together in new ways. However, very little is known about how compositional computation implemented the brain. Here, we ask participants to solve a series problems that each require constructing whole from set elements. With fMRI, find representations novel constructed objects frontal cortex and hippocampus are relational compositional. MEG, replay assembles elements into compounds, with sequence constituting hypothesis possible configuration The content sequences evolves as puzzle, progressing predictable uncertain gradually converging correct configuration. Together, these results suggest computational bridge between apparently distinct functions hippocampal-prefrontal circuitry role for generative inference testing.
Language: Английский
Citations
52
Biological Psychology, Journal Year: 2024, Volume and Issue: 186, P. 108741 - 108741
Published: Jan. 4, 2024
This review paper offers an overview of the history and future active inference—a unifying perspective on action perception. Active inference is based upon idea that sentient behavior depends our brains' implicit use internal models to predict, infer, direct action. Our focus conceptual roots development this theory (basic) sentience does not follow a rigid chronological narrative. We trace evolution from Helmholtzian ideas unconscious inference, through contemporary understanding In doing so, we touch related perspectives, neural underpinnings opportunities for development. Key steps in include formulation predictive coding theories neuronal message passing, sequential planning policy optimization, importance hierarchical (temporally) deep (i.e., generative or world) models. has been used account aspects anatomy neurophysiology, offer psychopathology terms aberrant precision control, unify extant psychological theories. anticipate further all these areas note exciting early work applying beyond neuroscience. suggests just biology, but robotics, machine learning, artificial intelligence.
Language: Английский
Citations
31
Science, Journal Year: 2025, Volume and Issue: 387(6733), P. 541 - 548
Published: Jan. 30, 2025
The mechanisms by which the brain replays neural activity sequences remain unknown. Recording from large ensembles of hippocampal place cells in freely behaving rats, we observed that replay content is strictly organized over multiple timescales and governed self-avoidance. After movement cessation, avoided animal’s previous path for 3 seconds. Chains self-repetition a shorter timescale. We used continuous attractor model to demonstrate neuronal fatigue both generates produces self-avoidance timescales. In addition, past experience became predominant later into stopping period, manner requiring cortical input. These results indicate mechanism generation unexpectedly constrains can be produced across time.
Language: Английский
Citations
3
Philosophical Transactions of the Royal Society B Biological Sciences, Journal Year: 2021, Volume and Issue: 377(1844)
Published: Dec. 27, 2021
This article considers the evolution of brain architectures for predictive processing. We argue that mechanisms perception and action are not late evolutionary additions advanced creatures like us. Rather, they emerged gradually from simpler loops (e.g. autonomic motor reflexes) were a legacy our earlier ancestors—and key to solving their fundamental problems adaptive regulation. characterize simpler-to-more-complex brains formally, in terms generative models include increasing hierarchical breadth depth. These may start simple homeostatic motif be elaborated during four main ways: these multimodal expansion control into an allostatic loop; its duplication form multiple sensorimotor expand animal's behavioural repertoire; gradual endowment with depth (to deal aspects world unfold at different spatial scales) temporal select plans future-oriented manner). In turn, elaborations underwrite solution biological regulation faced by increasingly sophisticated animals. Our proposal aligns neuroscientific theorising—about processing—with comparative data on animal species. is part theme issue ‘Systems neuroscience through lens theory’.
Language: Английский
Citations
61
Trends in Cognitive Sciences, Journal Year: 2023, Volume and Issue: 28(2), P. 97 - 112
Published: Nov. 15, 2023
Prominent accounts of sentient behavior depict brains as generative models organismic interaction with the world, evincing intriguing similarities current advances in artificial intelligence (AI). However, because they contend control purposive, life-sustaining sensorimotor interactions, living organisms are inextricably anchored to body and world. Unlike passive learned by AI systems, must capture sensory consequences action. This allows embodied agents intervene upon their worlds ways that constantly put best test, thus providing a solid bedrock is – we argue essential development genuine understanding. We review resulting implications consider future directions for AI.
Language: Английский
Citations
41
Advanced Robotics, Journal Year: 2023, Volume and Issue: 37(13), P. 780 - 806
Published: June 26, 2023
Creating autonomous robots that can actively explore the environment, acquire knowledge and learn skills continuously is ultimate achievement envisioned in cognitive developmental robotics. Importantly, if aim to create develop through interactions with their learning processes should be based on physical social world manner of human development. Based this context, paper, we focus two concepts models predictive coding. Recently, have attracted renewed attention as a topic considerable interest artificial intelligence. Cognitive systems better predict future sensory observations optimize policies, i.e. controllers. Alternatively, neuroscience, coding proposes brain predicts its inputs adapts model own dynamics control behavior environment. Both ideas may considered underpinning development humans capable continual or lifelong learning. Although many studies been conducted robotics neurorobotics, relationship between model-based approaches AI has rarely discussed. Therefore, clarify definitions, relationships, status current research these topics, well missing pieces conjunction crucially related such free-energy principle active inference context Furthermore, outline frontiers challenges involved toward further integration robotics, creation real capabilities future.
Language: Английский
Citations
38
Journal of Neurophysiology, Journal Year: 2023, Volume and Issue: 129(3), P. 552 - 580
Published: Feb. 8, 2023
Memory reactivations and replay, widely reported in the hippocampus cortex across species, have been implicated memory consolidation, planning, spatial skill learning. Technological advances electrophysiology, calcium imaging, human neuroimaging techniques enabled neuroscientists to measure large-scale neural activity with increasing spatiotemporal resolution provided opportunities for developing robust analytic methods identify replay. In this article, we first review a large body of historically important representative replay studies from animal literature. We then discuss our current understanding functions learning, consolidation further progress computational modeling that has contributed these improvements. Next, past present analyses their limitations challenges. Finally, looking ahead, some promising detecting nonstereotypical, behaviorally nondecodable structures recordings. argue seamless integration multisite recordings, real-time decoding, closed-loop manipulation experiments will be essential delineating role wide range cognitive motor functions.
Language: Английский
Citations
26