%0 Journal Article
%A Serrano Pedraza, Ignacio
%A Vancleef, Kathleen
%A Herbert, William
%A Goodship, Nicola
%A Woodhouse, Maeve
%A Read, Jenny C. A.
%T Efficient estimation of stereo thresholds: What slope should be assumed for the psychometric function?
%D 2020
%U https://hdl.handle.net/20.500.14352/116044
%X Bayesian staircases are widely used in psychophysics to estimate detection thresholds.Simulations have revealed the importance of the parameters selected for the assumed subject’spsychometric function in enabling thresholds to be estimated with small bias and highprecision. One important parameter is the slope of the psychometric function, or equivalentlyits spread. This is often held fixed, rather than estimated for individual subjects, becausemuch larger numbers of trials are required to estimate the spread as well as the threshold.However, if this fixed value is wrong, the threshold estimate can be biased. Here we determinethe optimal slope to minimize bias and maximize precision when measuring stereoacuitywith Bayesian staircases. We performed 2- and 4AFC disparity detection stereoexperiments in order to measure the spread of the disparity psychometric function in humanobservers assuming a Logistic function. We found a wide range, between 0.03 and 3.5 log10arcsec, with little change with age. We then ran simulations to examine the optimal spreadusing the empirical data. From our simulations and for three different experiments, we recommendselecting assumed spread values between the percentiles 60–80% of the populationdistribution of spreads (these percentiles can be extended to other type of thresholds).For stereo thresholds, we recommend a spread around the value σ = 1.7 log10 arcsec for2AFC (slope β = 4.3 /log10 arcsec), and around σ = 1.5 log10 arcsec for 4AFC (β = 4.9 /log10arcsec). Finally, we compared a Bayesian procedure (ZEST using the optimal σ) with fiveBayesian procedures that are versions of ZEST-2D, Psi, and Psi-marginal. In general, forthe conditions tested, ZEST optimal σ showed the lowest threshold bias and highestprecision.
%~