The shift from first-generation antipsychotic medications to second-generation antipsychotic medications initially caused a wave of excitement about the potential for improved and broader efficacy of these medications concurrent with an improved side-effect profile. Recent data from high-quality research analyses have subsequently raised significant questions about these claims. This research evidence has, however, not altered prescribing behavior in a way that would be expected from fully rational evaluation of the evidence. Prescribing decisions represent poorly understood, complex behaviors influenced by a number of external and internal forces, some of which may be elucidated by advances in social and cognitive psychology. In this article, the decision to prescribe first- versus second-generation antipsychotic medications is examined, and specific social psychological biases and individual cognitive biases are hypothesized to be significant influences on clinicians. These biases may perpetuate disparity between research evidence and clinical practice.