Picture naming and picture fragment identification: An investigation using a priming paradigm
The process of identifying an object is thought to involve at least three stages; the analysis of the object's visual features, access to its meaning and access to its name. When an object is identified repeatedly, large improvements in performance are observed. The effect of a repeated presentation may be due to a facilitation of one or more of the stages involved. The aim of the experiments reported here was to evaluate the stages and try to determine which are likely candidates for the repetition priming effect.
A short-term priming paradigm was used to investigate priming of speeded picture naming and picture fragment identification. Experiments 1, 2, 5, and 6 used primes that were designed to selectively affect each of the hypothetical processing stages--a visual prime, a semantic prime, and a lexical prime. A common assumption is that the most difficult stage should benefit most from a relevant prime and it was hypothesized that the most difficult stages would differ between the two tasks. As predicted, speeded picture naming was most affected by the visual and lexical primes whereas fragmented picture identification was most affected by the visual and semantic primes.
Experiment 8 investigated the effects of the same primes on picture clarification, a task which combined the response demands of picture naming with the task demands of picture fragment identification. The pattern of results resembled that obtained for speeded picture naming. Experiments 3 and 4 addressed the question of whether the semantic prime, the target's category name, provided visual as well as semantic information. The results supported our assumption that only the semantic stage was being affected.