A pilot study investigated the quality of diagnostic decisions that are made on the basis of a precise model of reading and learning to read. The study was carried out by creating a computer simulated reading clinician that embodied a recent model of the reading process and by observing the behavior of this simulated clinician as it diagnosed reading case records. Specific areas investigated were: (1) how well the simulation embodied the model, (2) how consistent and reliable the diagnostic decisions directed by the model were, and (3) how valid the diagnostic decisions directed by the model were. Findings indicated that the simulation was possible in as much as diagnostic test information provided answers similar to the items in the model, the simulation seemed to be reliable when comparing its diagnoses to those of reading clinicians, and the validity of the decisions was difficult to determine because case study information was inadequate. (MKM)