This paper reviews literature published on the psychophysiological effects of long-term human-animal interaction (i.e., pet ownership, pet adoption). A literature search was conducted using PsycInfo and Medline databases. Although the available evidence is far from being consistent, it can be concluded that, in some cases, long-term relationships with animals may moderate baseline physiological variables, particularly blood pressure. Results proved more coherent in studies where animals were adopted by owners as part of the procedure. This paper examines existing hypotheses seeking to account for these effects and the supporting evidence. Two major hypotheses have been suggested to explain the psychophysiological effects of long-term interaction, namely (1) stress-buffering effects of noncritical social support provided by pets; and (2) classical conditioning of relaxation. These mechanisms may partially account for the long-term health outcomes observed in a number of human-animal interaction studies.