Chung, Delacruz, de Vries, Bewley, and Baker (2006) describe five dimensions that underlie marksmanship performance—perceptual-motor, cognitive, affective, equipment, and environment. While prior research on the impact of cognitive (Harllee, 1916), affective (Tierney, Cartner, & Thompson, 1979; Sade, Bar-Eli, Bresler, & Tenenbaum, 1990; Chung, O'Neil, Delacruz, & Bewley, 2005), equipment, and environment (Osborne, Morey, & Smith, 1980) variables abound, studies on the perceptual-motor dimension are relatively incomplete. A possible reason for this lack of attention is the inherent difficulty in measuring the fine-motor movements involved in the practice of marksmanship. In training environments, a marksmanship instructor typically relies on two methods to evaluate shooter performance, visual observations of the shooter and inspection of shot distribution on the target, also known as shot placement analysis. Both methods, while commonplace, are subjective and dependent on a human observer, making it difficult to consistently assemble accurate and reliable information about a shooter's skill performance. This in turn leads to inefficiencies in diagnosis and feedback during the training process. A reliable and objective method is needed to help make proper diagnosis possible so that appropriate instruction is made available.

This paper presents work on sensor-based measures of rifle marksmanship skill performance. Sensors were developed to collect information on four skill areas, three related to breath control and one for trigger control. Data collected from experts (n = 9) and novices (n = 30) were used to fit and test a model of skill performance. Using sensing information alone, a two-class model differentiating between expert and novice skill performance achieved an accuracy of >90%. These preliminary results suggest that sensor-based skill assessment is a viable option as a reliable and objective measure to discriminate levels of skill performance in rifle marksmanship.