Action observation for the prediction of outcomes and learning of motor skills

当我们观察别人的动作时，我们对他们行为的理解有多少来自我们自己的视觉运动体验？我们如何从观察他人的新奇行为中学习？在观察动作时，什么样的视觉信息支撑着运动系统的激活？反过来，我们对这些信息的处理又是如何受到我们自身经验的影响的，以及我们潜在地重新制定或模拟我们所看到的东西的能力？这些一般性问题是本研究提案的基础。当我们观察他人的行为时，我们也会在一个隐蔽的层面上贯穿这个行为，这个想法已经得到了大量的关注和经验的支持。然而，这种所谓的“模拟”假设在学习环境中很少被探索。此外，对于许多运动技能，如参加体育运动或进行手术，需要预期的决策，问题仍然是支持这些决策的过程以及驱动这些决策或由这些决策驱动的感知线索。我计划调查行动观察过程，因为它们涉及学习和预测他人的行动结果。我的建议的A部分是基于使用双任务方法来探测运动系统在动作预测过程中的参与。我将研究动作模拟如何受到视觉线索拾取的影响。通过对训练类型的操作（向地板或墙上的目标投掷飞镖，或使用重物），我将确定促进模拟的条件，通用运动表征的程度，以及人们在参与运动系统时的灵活性。一个奇怪的可能性是，当预测别人的行动时，自己的运动系统会更新（精神上）。在B部分，我将使用适应协议来研究促进短期和长期适应的条件。通过操纵观察的时间安排，我将研究观察何时以及如何影响以后的运动表现（隐式）更新的运动计划。我将使用一种相对新颖的技术（外周神经刺激），研究在保持间隔期间，行动观察如何影响与记忆巩固相关的过程。这项研究计划将增加我们对我们的视觉运动体验如何影响行动观察过程的基本理解，特别是关于对他人行动的潜在模拟以及控制这种情况的条件。这些知识将影响我们对如何获得运动技能的理解，我们如何有意和无意地从观看中“学习”，以及如何做出预测性决策。这可能对我们如何教授和评估技能以及我们当前的行为如何可能干扰我们正确评估他人行为的能力产生重大影响。

When we watch others move, how much of our understanding of their behaviour comes about from our own visual-motor experiences? How are we able to learn from watching others perform novel actions? What visual information underpins any motor-system activation during action observation and in turn, how is the processing of this information influenced by our own experiences and potentially our ability to covertly re-enact or simulate what we see? These general questions underpin this research proposal. The idea that when we watch other's acting, we too run through the action at a covert level, is an idea that has received significant attention and empirical support. This so called `simulation' hypothesis has, however, been little explored in a learning context. Moreover, for many motor skills, such as playing sports or performing surgery, where anticipatory decisions are required, questions remain about the processes underpinning these decisions and the perceptual cues that drive or are driven by these. I plan to investigate action observation processes, as they pertain to learning and the prediction of other's action outcomes. Part A of my proposal is based on the use of dual-task methods to probe motor system engagement during action-prediction. I will study how action simulation is influenced by (or influences) visual cue pick-up. Through manipulations to the type of training (throwing darts to targets on the floor or wall, or with a weight), I will determine conditions which promote simulation, how general purpose motor-representations are, and how flexible people are in engaging their motor system. One curious possibility is that one's own motor system is updated (detrimentally) when making predictions about another's action. In Part B, I will use an adaptation protocol to study conditions which promote short- and long-term adaptation from observation. Through manipulations to the scheduling of observation I will study when and how observation impacts later motor performance with respect to (implicit) updating of motor plans. I will study how processes related to memory consolidation are affected by action-observation during the retention interval, using a relatively novel technique (peripheral nerve stimulation). This program of research will add to our basic understanding of how our visual-motor experiences impact processes of action-observation, specifically with respect to potential simulation of another's actions and the conditions which govern this. This knowledge will impact our understanding of how we acquire motor skills, how we intentionally and unintentionally `learn' from watching and how predictive decisions are made. This could have significant implications for how we teach and assess skills and how our current actions could potentially interfere with our ability to correctly assess another's behaviour.