May 2007
Volume 48, Issue 13
ARVO Annual Meeting Abstract  |   May 2007
Signal Detection Theory Analysis of Belmonte Esthesiometer Corneal Stimulation
Author Affiliations & Notes
  • J. Chen
    School of Optometry, University of Waterloo, Waterloo, Ontario, Canada
  • T. Simpson
    School of Optometry, University of Waterloo, Waterloo, Ontario, Canada
  • Footnotes
    Commercial Relationships J. Chen, None; T. Simpson, None.
  • Footnotes
    Support NSERC Canada
Investigative Ophthalmology & Visual Science May 2007, Vol.48, 396. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      J. Chen, T. Simpson; Signal Detection Theory Analysis of Belmonte Esthesiometer Corneal Stimulation. Invest. Ophthalmol. Vis. Sci. 2007;48(13):396.

      Download citation file:

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

  • Supplements

Signal detection theory (SDT) proposes that the response to sensory stimulus is affected by the subject’s perceptual sensitivity (d’) and response bias (criterion). The latter in particular has almost never been considered when measuring the effects of corneal stimulation and we therefore used SDT methods to measure the sensitivity and criterion effects when the cornea was stimulated using a pneumatic Belmonte esthesiometer.


12 healthy subjects (5 males and 7 females, 25 to 42 years) were used. Computer-controlled Belmonte esthesiometer was used to measure the response to pneumatic stimuli. Subjects’ d’ and criteria were estimated in 2 sessions one with 80% and another with 50% stimulus probability. Each session included 100 trials with the stimulus intensity set at the subject’s mechanical threshold determined using the ascending method of limits. Subjects reported their confidence of stimulus presence (via a computerized button box) using a 4-level rating scale (0- no stimulus to 3- certain stimulus). d’ and criteria were calculated using z-scores of hit and false alarm rates.


The figure shows the (well behaved) ROC curve of the pooled data. There was a linear relationships between z-scores of hit rates and false alarm rates for both stimulus probability data sets (R2= 0.995), suggesting that altering the criteria did not affect stimulus detectability. The slope of the bivariate regression is approximately 1.0 suggesting that a number of theoretical SDT conditions are present. The d’ for both lower and higher probability stimulus sets were approximately equal (1.4) while as anticipated, criteria differed.


Corneal mechanical sensitivity can be affected by both d’ and criterion. Using these newly applied methods, we can begin to examine the important role played by criterion in the processing of information by ocular surface sensory channels.  

Keywords: perception • innervation: sensation • cornea: clinical science 

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.