In serial recall from short-term memory, repeated items are recalled w
ell when close together (repetition facilitation), but not when far ap
art (repetition inhibition; the Ranschburg effect). These effects were
re-examined with a new scoring scheme that addresses the possibility
that repetitions are distinct tokens in memory. Repetition facilitatio
n and repetition inhibition proved robust, and were shown to interact
with the temporal grouping of items (Experiment 1), which affected the
probability of detecting repetition (Experiments 2A and 2B). It is ar
gued that detection of a repetition is necessary for repetition facili
tation, attributable to the tagging of immediate repetition, whereas t
he failure to detect or remember a repetition results in repetition in
hibition, attributable to an automatic suppression of previous respons
es and a bias against guessing repeated items (Experiment 3). The find
ings are discussed in relation to models of short-term memory and the
phenomenon of repetition blindness.