December 2022
Volume 22, Issue 14
Open Access
Vision Sciences Society Annual Meeting Abstract  |   December 2022
Do subjective judgements of grasp movements reflect objective kinematic information?
Author Affiliations
  • Leah Ettensohn
    National Institute of Mental Health
  • Chris I Baker
  • Maryam Vaziri-Pashkam
Journal of Vision December 2022, Vol.22, 3753. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Leah Ettensohn, Chris I Baker, Maryam Vaziri-Pashkam; Do subjective judgements of grasp movements reflect objective kinematic information?. Journal of Vision 2022;22(14):3753.

      Download citation file:

      © ARVO (1962-2015); The Authors (2016-present)

  • Supplements

Beyond executing grasp movements, we also regularly observe others grasping objects. How well can we identify the targets of observed grasp movements? Are we sensitive to kinematic information in grasp movements when identifying the target of a grasp or judging grasp movement similarity? We gathered kinematic data of hand movements as participants grasped a set of 58 3D-printed objects. To determine if the kinematic data contained enough information for distinguishing different grasp movements, we measured accuracy of a linear classifier in discriminating grasp movements from the kinematic data. The accuracy of the classifier at the final grasp position ranged from 76% to 96% across objects. Next, using the kinematic data we created point-light videos to serve as stimuli in two behavioral experiments. In the first experiment, participants performed an odd-one-out task to report similarity between observed grasp movements. By aggregating responses, we obtained a similarity matrix with high reliability (r = 0.74). In the second experiment, participants performed an identification task and chose the target of a point-light grasp movement from two options. Accuracies across objects had high reliability (r = 0.9), but performance was highly variable across objects ranging from 40-90% (mean > 65%). To determine if participants rely on the same kinematic information used by the classifier to perform judgements on the behavioral tasks, we used the confusion matrix from the linear classifier as a measure of kinematic similarity and compared it with the similarity matrices from the odd-one-out task and the confusion matrix from the identification task. The classifier accuracies had low albeit significant correlations with similarity measures from our behavioral experiments. These results suggest that only a subset of the kinematic information is used by the participants to perform the behavioral judgements. This indicates a dissociation between objective grasp movements and subjective judgements of those movements.


This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.