Disparity thresholds for sinusoidal depth corrugations in 27 observers: retinal versus relative spatial frequency
Description
Disparity thresholds for detecting horizontal sinusoidal depth corrugations in random-dot stereograms, measured in 27 observers across 14 conditions in which retinal spatial frequency (0.3, 1.0 and 1.5 cycles per degree) and relative spatial frequency (1 to 15 cycles per object, set by square windows 1 to 10 degrees on a side) were varied independently. Each observer was measured three times in every condition, so the data are provided both as a 27 x 14 matrix of means and as a 27 x 42 matrix of individual trials. Thresholds are in arcseconds; analyses are run on log10 thresholds. The deposit also contains the complete analysis code. Every statistic and every figure in the associated article is reproduced by these scripts, including the principal component analysis with varimax and oblimin rotation, the component-retention analyses (Horn's parallel analysis, Velicer's MAP, bootstrap eigenvalue intervals), the loading-structure significance test (LSS test) with its Monte Carlo and permutation nulls, the reliability analyses across the three repeated measurements, and the sensitivity check excluding the observer with 1200 arcsec stereoacuity. Nothing in the article was transcribed by hand. See README.md for the condition table, the file inventory, and the procedure for reproducing the loading-structure significance test. Data were collected at Newcastle University under ethics approval 00625. All observers gave informed consent. Participant identifiers are anonymous codes and the deposit contains no demographic or identifying information.
Files
Institutions
- Fielding Graduate UniversityCalifornia, Santa Barbara