Decision making under stressful circumstances may involve strong emotions and requires adequate prediction and valuation capabilities. In a social context contagion from others plays an important role as well. Moreover, agents adapt their decision making based on their experiences over time. Knowledge of principles from neuroscience provides an important source of inspiration to model such processes. In this paper an adaptive agent-based computational model is proposed to address the above-mentioned aspects in an integrative manner. As an application adaptive decision making of an agent in an emergency evacuation scenario is explored. By means of formal analysis and simulation, the model has been explored and evaluated. © 2013 Elsevier B.V.