Is there an accepted way to interpret d' (d-prime) for evidence of detection

Question

I have run a learning experiment, with a yes-no familiarity test at the end, and computed d' across various conditions.

Is there some rule of thumb (perhaps dependent on sample size) as to how d' measures can be interpreted? Is there some level at which we could say there is "strong evidence of detection"?

Answer 1

The best way I found to do this was simply to simulate the data with 10k iterations of a binomial distribution for a chance-level learner, and calculate the d'.

I then took the 95th percentile as a cut off rate, with all d' above the 95th percentile being taken as statistically unlikely to be from a non-learner.

Very simple to run in R, especially with the dprime command from the neuropsychology package. I provide the code below, this for a study with 8 'yes' trials and 8 'no trials':

library("neuropsychology")

hit <- rbinom(10000,8,0.5)
miss <- 8-hit
fa <- rbinom(10000,8,0.5)
cr <- 8-fa
dprime.10k <- dprime(hit, miss, fa, cr)
quantile(dprime.10k$dprime,probs = (0.95))

This gave me a value of 0.8706082, although obviously you'd get a different value for a different number of trials, or if you wanted to go for a more conservative cutoff rate (I took the 5% rate under the assumption that I would not get any "anti-learners" who identified noise over signal).