Previous research has shown that explicit and implicit knowledge of artificial grammars may decay at different rates (e.g., Tamayo & Frensch, 2007; Tunney, 2003). We extend these findings to sequential regularities embedded in Serial Reaction Time (SRT) tasks. We compared the forgetting patters of implicit and explicit knowledge after a retention interval of 7 days without rehearsal. Explicit knowledge decayed after 7 days whereas implicit knowledge was retained. These data were modeled according to the assumptions involved in the single-system model suggested by Shanks, Willkinson, and Channon (2003). The best fit for the model is obtained by modifying the parameters related to (a) the common knowledge-strength variable for implicit and explicit knowledge, and (b) the reliability for the explicit test. We interpret these dissociations as a boundary condition for single-system models that assume constant random noise to explain dissociations in the forgetting patterns of implicit and explicit sequential knowledge.