Observer-Agent Kinematic Similarity Facilitates Action Intention Decoding

Observer-Agent Kinematic Similarity Facilitates Action Intention Decoding
复制标题

DOI:
10.1038/s41598-020-59176-z
复制
发表时间:
2020-02-13
期刊:
影响因子:
4.6
通讯作者:
Fabbri-Destro, Maddalena
Fabbri-Destro, Maddalena
中科院分区:
综合性期刊3区
文献类型:
--
作者:
De Marco, Doriana;Scalona, Emilia;Fabbri-Destro, Maddalena

文献摘要

被引文献

相似文献

众所周知,动作的运动学是由潜在的运动意图调节的。反过来,运动学在动作观察期间也充当线索,提供关于观察动作意图的提示。然而,一个悬而未决的问题是,是否在他们的运动学基础上解码他人的意图仅仅取决于运动学在不同动作中的变化程度,或者更确切地说,它也受到其与观察者运动库的相似性的影响。在21名志愿者和一名演员的上肢运动中,根据目标大小和环境的不同,记录了他们的抓握和放置动作的执行情况。随后,志愿者们必须观察演员动作中唯一的“伸手到抓”阶段,并预测其潜在意图。评估了运动学参与者相似度对识别精度的潜在好处。在执行过程中,目标大小和环境调节了具体的运动学参数。更重要的是,尽管参与者在意图识别上的表现高于偶然,但运动模式的相似性与识别准确性呈正相关。总的来说,这些数据表明运动相似性在意图识别中发挥了促进作用,进一步支持了动作意图识别是基于运动共振的视觉-运动过程的观点。
It is well known that the kinematics of an action is modulated by the underlying motor intention. In turn, kinematics serves as a cue also during action observation, providing hints about the intention of the observed action. However, an open question is whether decoding others' intentions on the basis of their kinematics depends solely on how much the kinematics varies across different actions, or rather it is also influenced by its similarity with the observer motor repertoire. The execution of reach-to-grasp and place actions, differing for target size and context, was recorded in terms of upper-limb kinematics in 21 volunteers and in an actor. Volunteers had later to observe the sole reach-to-grasp phase of the actor's actions, and predict the underlying intention. The potential benefit of the kinematic actor-participant similarity for recognition accuracy was evaluated. In execution, both target size and context modulated specific kinematic parameters. More importantly, although participants performed above chance in intention recognition, the similarity of motor patterns positively correlated with recognition accuracy. Overall, these data indicate that kinematic similarity exerts a facilitative role in intention recognition, providing further support to the view of action intention recognition as a visuo-motor process grounded in motor resonance.