Loss aversion, risk aversion, and the probability weighting function (PWF) are three central concepts in explaining decisionmaking under risk. I examine interlinkages between these concepts in a model of decisionmaking that allows for loss averse/tolerant stochastic reference dependence and optimism/pessimism over probability distributions. I give a preference interpretation to commonly observed shapes of PWF and to risk aversion. In particular, I establish a connection between loss aversion and both risk aversion and the inverse-S PWF: loss aversion is a necessary condition to observe each of these phenomena. The results extend to distinct PWFs in the gain and loss domains, as under prospect theory.