Anh T Nguyen, Hao Zhang, Tian Tan, Mohammad Abrar, Yuxin Zhang, Michelle J Johnson
{"title":"Integration of a Gripper-Equipped Humanoid Social Robot for EEG-Monitored Action Observation Paradigms.","authors":"Anh T Nguyen, Hao Zhang, Tian Tan, Mohammad Abrar, Yuxin Zhang, Michelle J Johnson","doi":"10.1109/ICORR66766.2025.11062928","DOIUrl":null,"url":null,"abstract":"<p><p>Action Observation (AO) therapy leverages the mirror neuron system (MNS) and may support motor recovery in neurorehabilitation. In this study, we integrated Flo v2, a humanoid robot equipped with grippers and object detection system, into an AO therapy paradigm with electroencephalography (EEG) monitoring. Flo v2's enhanced design enables the execution of upper-limb actions, either transitive (involving object interaction such as grasping a cup) or intransitive (gesture-based without object manipulation such as waving). The robot control is synchronized with EEG recording to facilitate the investigation of cortical responses during AO tasks. We also conducted a case study to assess of the upgraded robot system's feasibility. Three healthy participants observed and imitated robot-performed actions, where the robot actor was in person or on videos. Exploratory analyses of EEG signals examined sensorimotor mu event-related desynchronization (ERD) during video-based and in-person AO tasks. Results indicated stronger responses during bimanual and transitive AO in the in-person settings. However, individual variability in cortical responses was evident, with one subject showing less pronounced ERD patterns, and that comparisons of mu ERDs across different types of action in video-based AO settings were inconsistent among subjects. Flo v2's enhancements demonstrated its feasibility as a tool for robot-mediated AOE therapy and highlighted potential for further neurorehabilitation research.</p>","PeriodicalId":73276,"journal":{"name":"IEEE ... International Conference on Rehabilitation Robotics : [proceedings]","volume":"2025 ","pages":"1699-1705"},"PeriodicalIF":0.0000,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"IEEE ... International Conference on Rehabilitation Robotics : [proceedings]","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICORR66766.2025.11062928","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Action Observation (AO) therapy leverages the mirror neuron system (MNS) and may support motor recovery in neurorehabilitation. In this study, we integrated Flo v2, a humanoid robot equipped with grippers and object detection system, into an AO therapy paradigm with electroencephalography (EEG) monitoring. Flo v2's enhanced design enables the execution of upper-limb actions, either transitive (involving object interaction such as grasping a cup) or intransitive (gesture-based without object manipulation such as waving). The robot control is synchronized with EEG recording to facilitate the investigation of cortical responses during AO tasks. We also conducted a case study to assess of the upgraded robot system's feasibility. Three healthy participants observed and imitated robot-performed actions, where the robot actor was in person or on videos. Exploratory analyses of EEG signals examined sensorimotor mu event-related desynchronization (ERD) during video-based and in-person AO tasks. Results indicated stronger responses during bimanual and transitive AO in the in-person settings. However, individual variability in cortical responses was evident, with one subject showing less pronounced ERD patterns, and that comparisons of mu ERDs across different types of action in video-based AO settings were inconsistent among subjects. Flo v2's enhancements demonstrated its feasibility as a tool for robot-mediated AOE therapy and highlighted potential for further neurorehabilitation research.