Abstract
Although binocular depth perception is usually better than monocular, responses may overlap greatly in some circumstances. Rather than supplying simply a measure of statistical significance, it is suggested that signal detection methodology be used to provide an index of separation between response distributions. Reanalyzing the data of Barrett and Williamson (1966) who used a paired-comparison technique plus confidence of response ratings, a d′ value of .5 was obtained. This supports their contention that distribution overlap existed but provides a better quantitative index of the extent of overlap than previously reported.
Get full access to this article
View all access options for this article.
