Predicting misuse and disuse of combat identification systems

Two combat identification systems have been designed to reduce fratricide by providing soldiers with the ability to "interrogate" a potential target by sending a microwave or laser signal that, if returned, identifies the target as a "friend." Ideally, gunners will appropriately rely on these automated aids, which will reduce fratricide rates. However, past research has found that human operators underutilize (disuse) and overly rely on (misuse) automated systems (cf. Parasuraman & Riley, 1997). The purpose of this laboratory study was to simultaneously examine misuse and disuse of an automated decision-making aid at varying levels of reliability. With or without the aid of an automated system that is correct about 90%, 75%, or 60% of the time, 91 college students viewed 226 slides of Fort Sill terrain and indicated the presence or absence of camouflaged soldiers. Regardless of the reliability of the automated aid, misuse was more prevalent than disuse, F(1, 65) = 31.43, p < .01; p = .27 for misuse, p = .13 for disuse. Results are interpreted within a general framework of automation use (Dzindolet, Beck, Pierce, & Dawe, 2001).