The puzzle is complicated: when should working memory be related to implicit sequence learning, and when should it not? (Response to Martini et al.).

Martini, Sachse, Furtner, and Gaschler (in press) commentary highlights two points related to the relationship between workingmemory (WM) and sequence learning (SL): 1) how the functional distinction of WM can affect the results when investigating this relationship, and 2) how implicit WM might have a stronger influence on implicit SL than the traditional construct of WM does. We agree with Martini et al. (in press) that different aspects of WM can have different relations to SL. We briefly addressed this question in our paper, suggesting, for example, that a stronger relationship might be found within-domain (e.g., between visual short-term memory/ WM and visuospatial SL) than between-domains (e.g., between verbal short-term memory/WM and visuospatial SL) (Janacsek & Nemeth, 2013). In addition, it is also important to consider whether SL is implicit or explicit. Studies so far have shown a stronger relationship between WM and explicit rather than implicit SL (e.g., Frensch & Miner, 1994; Unsworth & Engle, 2005; Weitz, O'Shea, Zook, & Needham, 2011). Supporting this claim, Martini, Furtner, and Sachse (2013) have demonstrated the role of relational integration and WM updating in SL under specific conditions (with longer, 300 msec response-to-stimulus intervals, RSIs). As they reported, under this condition participants gained more explicit knowledge about the hidden sequence structure than in the 0 msec RSI condition. WM capacity correlated with SL as well as with the level of explicit knowledge under the longer RSI condition, suggesting that these participants were able to use more cognitive control during SL. This result is in line with our conclusion that WM capacity is more related to explicit than to implicit SL.