In models of optimal decision making, assumptions about managerial behavior are often made with the hope that the prescriptions offered by these models will be effective in practice, even if actual behavior occasionally strays from these assumptions. However, recent revenue management (RM) research has demonstrated what appear to be systematic deviations from normative models of decision making. These deviations can even be observed in relatively simple RM contexts. We suggest that technical errors in capacity allocation decisions are linked to issues such as arousal and stress associated with state conditions of RM tasks. Our study goes beyond existing findings by considering behavioral phenomena in concurrent task settings, where the decision maker is faced with managing decisions for more than one product or service. Physiological measures of eye dilation and blink rate are used as markers of arousal and stress in subjects engaged in RM tasks. Our analysis shows that physiological responses are indeed associated with both the state conditions of RM tasks and the number of capacity blocks managed concurrently by an individual. Deviations from modeled decision making appear to be significantly dependent upon these physiological responses. We conclude with a discussion of implications for further research and practice.