The public's inability to gain direct personal experience or information$\backslash$nabout American military operations means that individuals must rely on$\backslash$ncues to form opinions about war. But in an environment filled will$\backslash$npotential cues, which ones do Americans tend to rely on when deciding$\backslash$nwhether to support an ongoing military operation? This experimental study$\backslash$nuses two distinct cues within the context of a newspaper story about the$\backslash$nIraq War to test four theoretical models of the American public's reliance$\backslash$non cues. The results provide fairly consistent support for the ‰ÛÏsurprising$\backslash$nevents- model of opinion formation, which suggests that individuals will$\backslash$nattend to news events that conflict with their expectations in an effort$\backslash$nto update their attitudes toward the war. These results also provide$\backslash$nsupport for the cost/benefit perspective on the formation of public$\backslash$nopinion toward war that underpins much of the literature on casualty$\backslash$ntolerance during military conflicts.