In the last decades of the twentieth century, interest in effects of consciousness on health and illness generated several lines of investigation into effects on cancer. Animal studies showed sensitivity of some cancers to hormonal and stressful influences. However, those findings did not translate into effects on humans, nor did they lead to advances in understanding of human cancer. The proposal that emotional state or stress, mediated through psycho-neuro-immunologic mechanisms would affect cancer generation or growth, resulted in conflicting information. Major surveys found no relationship. The proposal of a cancer personality (Type C) also was not confirmed. Initial observations that depression and stress affected human cancer seem to have best been explained by misinterpretations of cause and effect. By the mid 1990s, a remaining thesis--effect of psychosocial support on longevity and the course of cancer--was yet to be resolved. Initial positive results, especially findings in two popularly quoted studies, were not confirmed; they seem to have been due to inadequate numbers (chance) or to artifacts in study design or implementation. Psychosocial support may result in better adjustment and quality of life, but it does not directly affect the evolution of human cancer. Copyright 2002, Elsevier Science (USA). All rights reserved.