Articulatory preparation in the delayed naming task

The assumptions that acoustic onset must follow articulatory onset by a fixed delay and that response execution level processes are always effectively isolated in the delayed naming task were investigated with respect to the issue of articulatory preparation in three experiments. The results of these experiments showed that for the delayed naming task with the standard instructions (1) there was differential preparation across the different delays, with articulatory preparation being initiated earlier relative to the signal to respond in the longer delays (in some cases, even before the signal to respond was given); (2) the articulatory onset to acoustic onset interval (AAI) was variable across the different delays as a direct consequence of differential preparation; and (3) preparation included the building up of pressure for word-initial plosives when it was possible. However, when explicit instructions were given to minimize articulatory preparation in advance, an acoustic latency difference between plosives and nasals was found even with long delays. Moreover, comparison of the delayed naming task under different instructions and the standard naming task showed that execution of a response, in terms of acoustic latency and segment duration, can vary systematically across instructions and tasks. Methodological and theoretical implications of these results are discussed.