This study investigated the effects of assessment context on state anxiety and attention according to the Mirsky (1996) model of attention. Context varied in the physical testing environment, demeanor of the assessor, and explanation of the purpose of testing. A relaxed condition (RC) and structured medical condition (SMC) distinction was made prior to data collection and the two contexts were designed to reflect contrasting practices of neuropsychologists. Elements of attention evaluated included Encoding (Digit Span), Focusing/Executing (Visual Search and Attention Test), Shifting (Wisconsin Card Sorting Test: Computerized Version 2), Sustaining, and Stabilizing (Continuous Performance Test-Identical Pairs). Eighty healthy adult females participated in the study. The findings suggest that the SMC caused higher levels of anxiety and lower valence than the RC, which in turn caused poorer sustained attention and superior shifting attention for this condition. Such interpretations are consistent with several theories on the effects of anxiety on attention. It should be noted, however, that differences observed in attention were limited to select measures. Factor analysis also indicates that the encode, shift, and sustain elements of attention were largely consistent with the factor solution proposed by Mirsky, while findings on the focus/execute and stabilize elements bring into question the construct validity of these aspects of the model. Findings from the study are considered relevant to those interested in attention theory and particularly researchers and clinicians involved in the administration of neuropsychological testing.