When making multi-alternative decisions (i.e., decisions with more than two choice options), people do not seem to evaluate the utility of choice options independently. Accordingly, studying decisions in multi-alternative environments can improve our understanding of how decisions are actually made and of how different contexts determine different encoding of utility in the brain. I will present two studies in which we examined multi-alternative decision making and reward valuation using a combined approach of cognitive modeling, fMRI, and eye-tracking analyses. Our results challenge the view that utility is a context-independent property and that it could be measured with behavioral or neuroimaging methods in a context-free manner.