May 2008
Volume 49, Issue 13
Free
ARVO Annual Meeting Abstract  |   May 2008
Meander Mazes: Using Simulations of Extremely Limited Prosthetic Vision to Guide Hand Movement
Author Affiliations & Notes
  • L. A. Ostrin
    Wilmer Eye Institute, Johns Hopkins University, Baltimore, Maryland
  • D. C. Duval
    Wilmer Eye Institute, Johns Hopkins University, Baltimore, Maryland
  • M. Barry
    Wilmer Eye Institute, Johns Hopkins University, Baltimore, Maryland
  • G. Dagnelie
    Wilmer Eye Institute, Johns Hopkins University, Baltimore, Maryland
  • Footnotes
    Commercial Relationships  L.A. Ostrin, None; D.C. Duval, None; M. Barry, None; G. Dagnelie, None.
  • Footnotes
    Support  NIH 5 T32 EY07143
Investigative Ophthalmology & Visual Science May 2008, Vol.49, 3029. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      L. A. Ostrin, D. C. Duval, M. Barry, G. Dagnelie; Meander Mazes: Using Simulations of Extremely Limited Prosthetic Vision to Guide Hand Movement. Invest. Ophthalmol. Vis. Sci. 2008;49(13):3029.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose: : Degenerative eye diseases such as macular degeneration and retinitis pigmentosa cause irreversible damage to the photoreceptors. Retinal prostheses are currently being implanted in retinitis pigmentosa patients in the US and Germany; recipients most likely will require extensive training to learn to interpret this type of vision. We tested eye-hand coordination (EHC) under simulated retinal implant conditions and evaluated how different phosphene settings affect EHC.

Methods: : EHC was tested in 3 normally and 1 partially sighted subjects using a maze tracing task on a tablet computer in two experiments. In each, five conditions of a 60 electrode simulated retinal implant were studied; the 6x10 phosphene grid was locked to the subjects’ gaze. In exp. 1, high and low contrast were used, with cumulative addition of phosphene dropout, dynamic noise, and grid distortion; in exp. 2, the final condition of exp. 1 was presented with two levels of spatial and three levels of temporal "smearing." The most complex condition was equivalent to VA<20/3200, delay > 1 s. Testing was conducted in 10 sessions per experiment, resulting in each subject tracing 240 and 160 mazes in exp. 1 and 2, respectively. EHC was assessed through two time measures (finding starting point and tracing to end point) and one error measure (area subtended by trace outside boundaries) as a function of test parameters and practice.

Results: : Subjects were quite different in their ability to accurately trace mazes in a timely manner. In general, traicing time and error area increased as contrast decreased and drop out, noise and distortion increased. Trace time was not necessarily correlated to error area. In exp 1, trace time and trace error for the high contrast condition were 34±25 sec and 995±674 mm2. This increased to 50±24 sec and 2464±2742 mm2 for the most complex viewing condition. Learning effects improved tracing time for high and low contrast conditions by 66±14%, and trace error decreased by 73±66%. Learning effects were limited to the initial 50-100 trials. With 70% of data in exp. 2 collected, we find that trace time and error were lowest for the most simple condition, 19±17 sec and 12±19 pixels2, resectively, and increased up to 28±47 sec and 21±24 pixels2 for more complex conditions.

Conclusions: : The results show that subjects were able to perform the tracing task under impaired visual conditions. However, performance is related to subject motivation and training. The integration of prosthetic vision simulation and maze-tracing provides a novel approach to low vision rehabilitation and performance evaluation for retinal implant wearers.

Keywords: low vision • retinal degenerations: hereditary • degenerations/dystrophies 
×
×

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.

×