October 2003
Volume 3, Issue 9
Vision Sciences Society Annual Meeting Abstract  |   October 2003
Learning Optimal Gaze Decomposition
Author Affiliations
  • Eric Wiewiora
    UC San Diego, USA
  • David Berg
    UC San Diego, USA
  • Jochen Triesch
    UC San Diego, USA
  • Tomonori Hashiyama
    City University, Nagoya, Japan
Journal of Vision October 2003, Vol.3, 437. doi:https://doi.org/10.1167/3.9.437
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Eric Wiewiora, David Berg, Jochen Triesch, Tomonori Hashiyama; Learning Optimal Gaze Decomposition. Journal of Vision 2003;3(9):437. https://doi.org/10.1167/3.9.437.

      Download citation file:

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

  • Supplements

When the head is free to move, subjects frequently engage in coordinated head and eye movements to bring a target object to the fovea. Freedman and Sparks (1997) found that the relative contribution of head and eye movements to the total gaze shift are a non-linear function of initial eye position and total gaze displacement. Freedman (2001) and Wang et.al. (2002) have recently proposed descriptive mathematical models for the decomposition of total gaze shift into head and eye movements. It is however an open question a) if and how this decomposition can be seen as resulting from an optimality principle, b) if this decomposition strategy is learned during development and c) if so, what learning mechanisms are responsible for its aquisition. We propose a model for the simultaneous learning of the calibration of goal directed head/eye movements and the optimal gaze shift decomposition based on a reinforcment learning mechanism (Schultz et.al., 1997). We show that the rather complex behaviorally observed gaze decomposition can be understood as the result of optimizing a simple cost function. In our model, the cerebellum plays a key role in learning a gaze shift decomposition that accurately brings the desired target to the fovea while at the same time minimizing this cost function. Our model is consistent with the known anatomy and physiology of oculomotor control systems. The model efficiently learns gaze shift decompositions observed experimentally and makes a number of testable predictions. The model is also implemented and tested in an anthropomorphic robot head that autonomously learns the optimal gaze decomposition.


FreedmanEGSparksDL(1997). J. Neurophysiol., 77:2328–2348, FreedmanEG(2001). Biol. Cybern., 84:453–462, SchultzWDayanPMontaguePR(1997). Science, 275:1593–1599

WangXJinJJabriM(2002). Neural Networks, 15:811–832

Wiewiora, E., Berg, D., Triesch, J., Hashiyama, T.(2003). Learning Optimal Gaze Decomposition [Abstract]. Journal of Vision, 3( 9): 437, 437a, http://journalofvision.org/3/9/437/, doi:10.1167/3.9.437. [CrossRef]

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.