Controlled response selection benefits explicit, but not implicit sequence learning

In two experiments with the serial reaction time task the effect of response selection processes on sequence learning was examined by manipulating stimulus-response compatibility between training groups. In Experiment 1 participants were first trained with either compatible or incompatible stimulus-response mapping. Then, to dissociate effects on sequence learning versus sequence performance, transfer across stimulus-response compatibilities was measured in order to allow comparison of sequence learning under similar conditions. Surprisingly, the data from the training phase showed that sequence learning was better with compatible than incompatible stimulus-response mapping. The divergent nature of this finding from those observed in previous studies (e.g., Deroost & Soetens, 2006b; Koch, 2007) was hypothesized to indicate that explicit but not implicit sequence learning is affected by stimulus-response compatibility. Experiment 2 supported this notion as stimulus-response compatibility did not affect sequence learning while employing a complex probabilistic sequence, known to produce very limited explicit sequence knowledge.