Evaluating how placebos produce change. Logical and causal traps and understanding cognitive explanatory mechanisms.

The renewed interest in placebo effects from psychological and methodological angles is manifested in two complementary facets: the causal relationship between placebo effects inside and outside of clinical trials and the continual effort to understand how the underlying mechanisms relate to notions of efficacy and effectiveness. The article challenges the premise that placebo effects can be measured precisely enough that variance attributed to "nonspecific" placebo effects can be hierarchically partitioned from variance attributed to other "more specific" elements of therapy and discusses some of the most important recent developments in the understanding of how placebos produce change through cognitive explanatory mechanisms, including efficacy, outcome, and response expectancies; reverse placebo effects; schemas; and biased-information processing. Although much is already known about the complexity of explanatory mechanisms in placebo and expectancy effects, there is still much to do before establishing causal relationships and developing valid treatments based on this knowledge.