A recent research interest led us to compute the expected proportion of correct responses when the signal is one of M orthogonal signals added to Gaussian noise. This is formally the same as the probability of a correct response PM(c) in an M-alternative forced-choice task, and therefore our efforts duplicate some earlier calculations published by Hacker and Ratcliff (1979). Our calculated values are almost precisely the same as theirs. We suggest a simple approximation that yields the correct value of PM(c) to within one quarter of a percent for several values of M.
For the newest resources, visit Wolfram Repositories and Archives »
