ABSTRACT
Posner's method of using differences in Reaction time (RT) for physical and name matches to estimate the time constant of visual short-term memory (STM) is criticized as confounding the decay of the visual trace with the development of a name code. Posner and his co-workers have recently devised an ingenious technique by which RT measures may be used to study visual encoding, and from which they have drawn interesting conclusions about short-term visual memory. Closer examination of the rationale behind the Posner result suggests that it may be based upon a logical error. It assumes that the discrepancy between the RT to physically identical terms and terms with the same name but different case simply reflects the strength of the physical trace, with the two RTs becoming equal when the trace becomes indistinguishable from the visual background noise. Both accuracy and RTs give mutually consistent results, which suggest that RT may indeed provide a useful indicator of short-term forgetting.
