In working memory (WM) span tests, participants maintain memory items while performing processing tasks. In this study, we examined the impact of task processing requirements on memory-storage activities, looking at the stimulus order effect and the impact of storage requirements on processing activities, testing the processing time effect in WM span tests. The stimulus order effect is a phenomenon whereby span scores are higher when span lists end with a shorter task than when they end with a longer task. The processing time effect refers to the phenomenon whereby the processing speed slows for the later processing positions in a list in WM span tests when participants have more memory items to remember. Two experiments demonstrated the domain-specific nature of the stimulus order effect, which was observed only when the processing and storage materials were in the same domain (either verbal or visuospatial domains), and the domain-general nature of the processing time effect, which was observed with any combination of processing and storage materials. Although the domain specificity of the stimulus order effect is compatible with the representation-based interference view proposed by Saito and Miyake [Saito, S., & Miyake, A. (2004). On the nature of forgetting and the processing-storage relationship in reading span performance. Journal of Memory and Language, 50, 425-443.], the domain generality of the processing time effect is inconsistent with their account and indicates the involvement of a central bottleneck in WM span performance. (C) 2006 Elsevier Inc. All rights reserved.