Human Limits to Cognitive Information Fusion in a Military Decision-Making Task

摘要

One purported benefit of Network Enabled Operations is the increased availability of shared information, which can contribute to improved situational awareness, decision-making and overall mission effectiveness. Using a simulated Mission Command task focused on capturing high value targets (HVTs), we investigated how varying levels of available information affects human decision-making. The information presented to participants consisted of reports of possible HVT locations. Some reports indicated the correct location of a target; incorrect reports indicated a variable location immediately adjacent. As compared to a single report, it was possible to pinpoint the true location of a target by integrating information from multiple reports. However, results showed that participants did not locate HVTs faster with multiple intelligence reports compared to a single report. To determine if this was due to cognitive limits, human performance was compared to an Ideal Observer Model which had perfect information integration but the same task timing constraints. The model demonstrated a considerable improvement in performance with increased volume of information. These findings raise questions about human capabilities for information fusion given the high volume of information in military networks. Furthermore, results suggest decision support systems may enhance human capabilities for fusing and disambiguating information.