Validity of Low-Dimensional Representation of Whole-Body Movements
Research on impression evaluation from whole-body movements has traditionally focused on the relationships between emotional perception or aesthetic evaluation and movement patterns characterized by local kinematic features and inter-limb coordination. However, because observers perceive the entire body simultaneously when evaluating emotions or aesthetics in real-world viewing situations,…
Traditional research on evaluating impressions from whole-body movements has primarily examined the connections between emotional perception or aesthetic judgments and movement patterns, which are characterized by local kinematic features and coordination between limbs. However, when individuals assess emotions or aesthetics in real-world scenarios, they typically observe the entire body simultaneously.
This necessitates the investigation of a low-dimensional representation capable of encapsulating whole-body information in an easily grasped format. In this study, the researchers sought to determine if the state space generated by multidimensional scaling (MDS) serves as a valid representation for summarizing whole-body movements.
To do so, they used exercises from Radio Calisthenics No. 1 and investigated whether the distances between points in the MDS state space primarily reflected the exercise category instead of individual differences among performers. The findings revealed that the distances in the state space were more strongly associated with exercise category than with individual identity, indicating that MDS effectively captures similarities among postures within the same exercise.
Nevertheless, the researchers observed that the patterns of distances between conditions differed across various exercise categories. These results suggest that MDS-based state-space representations provide a valid and interpretable method for summarizing whole-body movements. However, further validation is necessary using a broader range of whole-body movements and a more diverse participant group.
Written by urgent.news from bioRxiv's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.