Studies testing cognitive theory of depression (Beck, 1963, 1987) and defining depression as a clinical syndrome are reviewed. Many aspects of the theory's descriptive claims about depressive thinking have been substantiated empirically, including (a) increased negativity of cognitions about the self, (b) increased hopelessness, (c) specificity of themes of loss to depressive syndromes rather than psychopathology in general, and (d) mood-congruent recall. Evidence that depressive thinking is especially inaccurate or illogical, however, is weak. Fewer studies have tested the theory's causal (diathesis-stress) hypotheses, and there is no strong evidence supporting them.