In the face of ample experimental evidence on the importance of working memory capacity for everyday life, there is a growing need for measures suited for clinical assessment of working memory. For this purpose, the Adaptive Digit Ordering Test (DOT-A), a new version of a digit ordering test introduced by Cooper et al. [Brain 114 (1991) 2095], was developed in analogy to the Digit spans. In Study 1, we investigated DOT-A performance in patients with Parkinson's disease (PD) and patients with frontal lobe damage, as these groups often exhibit working memory impairments within the framework of executive dysfunctions. In comparison with matched controls, both patient groups showed reduced performance in DOT-A but not in Digit span performance. This pattern was found to be particularly sensitive for patients with PD. In Study 2, DOT-A performance was assessed in 50 healthy subjects carefully selected according to demographic criteria in order to ensure representativity. Parallel test and split-half correlations indicated sufficient reliability of the DOT-A. Concurrent validity was confirmed by significant correlation with a well-established working memory test (two-back task). We conclude that DOT-A is a promising diagnostic instrument, and its economy and direct comparability to the Wechsler Digit spans and high sensitivity for patient populations make it especially well-suited for assessment in clinical practice.