In an extensive series of experiments Sternberg, Monsell, Knoll, and Wright (1978) reported that simple RT increased as a linear function of the number of items to be pronounced or typed. The present experiments replicate a portion of these results, but show that the effect is less general than may have been supposed. Since the effect does not occur in every case in which a response programming interpretation would predict it, this interpretation must be rejected. This conclusion is consistent with the viewpoint that response programming should be investigated in a choice- rather than simple-RT paradigm. In this view, motivated subjects can program responses in advance of the simple-RT interval because the particular response to be made has been precued. Effects of response parameters which are observed for motivated subjects in the simple-RT paradigm, such as those reported by Sternberg et al. (1978), should be attributed to processes other than programming motor responses.