A novel theoretical framework that describes a possible rationale for the regularity in how we move, how we learn, and how our brain predicts events. In
Biological Learning and Control, Reza Shadmehr and Sandro Mussa-Ivaldi present a theoretical framework for understanding the regularity of the brain's perceptions, its reactions to sensory stimuli, and its control of movements. They offer an account of perception as the combination of prediction and observation: the brain builds internal models that describe what should happen and then combines this prediction with reports from the sensory system to form a belief.
Considering the brain's control of movements, and variations despite biomechanical similarities among old and young, healthy and unhealthy, and humans and other animals, Shadmehr and Mussa-Ivaldi review evidence suggesting that motor commands reflect an economic decision made by our brain weighing reward and effort. This evidence also suggests that the brain prefers to receive a reward sooner than later, devaluing or discounting reward with the passage of time; then as the value of the expected reward changes in the brain with the passing of time (because of development, disease, or evolution), the shape of our movements will also change.
The internal models formed by the brain provide the brain with an essential survival skill: the ability to predict based on past observations. The formal concepts presented by Shadmehr and Mussa-Ivaldi offer a way to describe how representations are formed, what structure they have, and how the theoretical concepts can be tested.
一個新的理論框架,描述了我們移動、學習以及大腦預測事件的規律性可能的理由。
在《Biological Learning and Control》中,Reza Shadmehr 和 Sandro Mussa-Ivaldi 提出了理解大腦感知規律性、對感官刺激的反應以及運動控制的理論框架。他們將感知描述為預測與觀察的結合:大腦建立內部模型,描述應該發生的事情,然後將這一預測與感官系統的報告結合,形成信念。
考慮到大腦對運動的控制,以及儘管在生物力學上存在相似性,老年人和年輕人、健康人和不健康人,以及人類和其他動物之間的變異,Shadmehr 和 Mussa-Ivaldi 回顧了證據,表明運動指令反映了我們的大腦在權衡獎勵和努力後所做出的經濟決策。這些證據還表明,大腦更傾向於儘早獲得獎勵,而不是延遲,隨著時間的推移,獎勵的價值會被貶值或折扣;隨著時間的推移(因為發展、疾病或進化),預期獎勵的價值在大腦中變化,我們的運動形狀也會隨之改變。
大腦形成的內部模型為大腦提供了一項基本的生存技能:根據過去的觀察進行預測的能力。Shadmehr 和 Mussa-Ivaldi 提出的正式概念提供了一種描述表徵如何形成、它們具有何種結構以及如何測試這些理論概念的方法。
Reza Shadmehr is Professor of Bioengineering and Professor of Neuroscience at the Johns Hopkins University School of Medicine. He is the coauthor of
The Computational Neurobiology of Reaching and Pointing and
Biological Learning and Control (both published by the MIT Press).
Sandro Mussa-Ivaldi is Professor of Physiology in the Medical School at Northwestern University, with joint appointments in Physical Medicine and Rehabilitation and Biomedical Engineering. He is also Founder and Director of the Robotics Laboratory at the Rehabilitation Institute of Chicago.
Reza Shadmehr 是約翰霍普金斯大學醫學院的生物工程教授和神經科學教授。他是《The Computational Neurobiology of Reaching and Pointing》和《Biological Learning and Control》(均由麻省理工學院出版社出版)的共同作者。
Sandro Mussa-Ivaldi 是西北大學醫學院的生理學教授,並在物理醫學與復健以及生物醫學工程領域擔任聯合職位。他也是芝加哥復健研究所機器人實驗室的創始人和主任。