Psychological Review2000, Vol. 107, No. 3, 411-429Copyright 2000 by the American Psychological Association, Inc.0033-295X/0№00 DOI: 10.1037//0033-295X.107.3.411Biobehavioral Responses to Stress in Females:Tend-and-Befriend, Not Fight-or-FlightShelley E. Taylor, Laura Cousino Klein, Brian P. Lewis, Tara L. Gruenewald,Regan A. R. Gurung, and John A. UpdegraffUniversity of California, Los AngelesThe human stress response has been characterized, both physiologically and behaviorally, as "fight-or-flight." Although fight-or-flight may characterize the primary physiological responses to stress for bothmales and females, we propose that, behaviorally, females' responses are more marked by a pattern of"tend-and-befriend." Tending involves nurturant activities designed to protect the self and offspring thatpromote safety and reduce distress; befriending is the creation and maintenance of social networks thatmay aid in this process. The biobehavioral mechanism that underlies the tend-and-befriend patternappears to draw on the attachment-caregiving system, and neuroendocrine evidence from animal andhuman studies suggests that oxytocin, in conjunction with female reproductive hormones and endogenousopioid peptide mechanisms, may be at its core. This previously unexplored stress regulatory system hasmanifold implications for the study of stress.Survival depends on the ability to mount a successful responseto threat. The human stress response has been characterized asfight-or-flight (Cannon, 1932) and has been represented as anessential mechanism in the survival process. We propose thathuman female responses to stress (as well as those of some animalspecies) are not well characterized by fight-or-flight, as researchhas implicitly assumed, but rather are more typically characterizedby a pattern we term "tend-and-befriend." Specifically, we suggestthat, by virtue of differential parental investment, female stressresponses have selectively evolved to maximize the survival of selfand offspring. We suggest that females respond to stress by nur-turing offspring, exhibiting behaviors that protect them from harmand reduce neuroendocrine responses that may compromise off-spring health (the tending pattern), and by befriending, namely,Shelley E. Taylor, Laura Cousino Klein, Brian P. Lewis. Tara L.Gruenewald, Regan A. R. Gurung, and John A. Updegraff, Department ofPsychology, University of California, Los Angeles.Laura Cousino Klein is now in the Department of Biobehavioral Health,Pennsylvania State University; Brian P. Lewis is now in the Department ofPsychology, Syracuse University; and Regan A. R. Gurung is now in theDepartment of Psychology, University of Wisconsin—Green Bay.Support for preparation of diis article was provided by National ScienceFoundation Grant SBR 9905157, National Institute of Mental Health GrantMH 056880, and the MacArthur Foundation's SES and Health Network.All of the authors except Shelley E. Taylor were supported by a NationalInstitute of Mental Health Training Grant MH 15750 in health psychologyat various points throughout the preparation of this article. We are gratefulto Nancy Adler, David A. Armor, Lisa Aspinwall, John Cacioppo, ElissaEpel, Alan Fiske, Gregg Gold, Melissa Hines, Margaret Kemeny, JenniferLerner, Sonja Lyubomirsky, Karen Matthews, Bruce McEwen, L. AnnePeplau, Lien Pham, Inna Rivkin, Joan Silk, Robert Trivers, and RosemaryVeniegas for their comments on previous versions of this article.Correspondence concerning this article should be addressed to ShelleyE. Taylor, Department of Psychology, 1283 Franz Hall, University ofCalifornia, Los Angeles, Los Angeles, California 90095-1563. Electronicmail may be sent to [email protected] with social groups to reduce risk. We hypothesize andconsider evidence from humans and other species to suggest thatfemales create, maintain, and utilize these social groups, especiallyrelations with other females, to manage stressful conditions. Wesuggest that female responses to stress may build on attachment-caregiving processes that downregulate sympathetic and hypo-thalamic-pituitary-adrenocortical (HPA) responses to stress. Insupport of this biobehavioral theory, we consider a large animaland human literature on neuroendocrine responses to stress, sug-gesting that the tend-and-befriend pattern may be oxytocin medi-ated and moderated by sex hormones and endogenous opioidpeptide mechanisms.BackgroundThe fight-or-flight response is generally regarded as the proto-typic human response to sttess. First described by Walter Cannonin 1932, the fight-or-flight response is characterized physiologi-cally by sympathetic nervous system activation that innervates theadrenal medulla, producing a hormonal cascade that results in thesecretion of catecholamines, especially norepinephrine and epi-nephrine, into the bloodstream. In addition to its physiologicalconcomitants, fight-or-flight has been adopted as a metaphor forhuman behavioral responses to stress, and whether a human (or ananimal) fights or flees hi response to sympathetic arousal isthought to depend on the nature of the stressor. If the organismsizes up a threat or predator and determines that it has a realisticchance of overcoming the predator, then attack is likely. In cir-cumstances in which the threat is perceived to be more formidable,flight is more probable.A coordinated biobehavioral stress response is believed to be atthe core of reactions to threats of all kinds, including attacks bypredators; assaults by members of the same species; dangerousconditions such as fire, earthquake, tornado, or flooding; and otherthreatening events. As such, an appropriate and modulated stressresponse is at the core of survival. Through principles of natural411412TAYLOR ET AL.selection, an organism whose response to stress was successfulwould likely pass that response on to subsequent generations, andthe fight-or-flight response is thought to be such an evolvedresponse.A little-known fact about the fight-or-flight response is that thepreponderance of research exploring its parameters has been con-ducted on males, especially on male rats. Until recently, the genderdistribution in the human literature was inequitable as well. Priorto 1995, women constituted about 17% of participants in labora-tory studies of physiological and neuroendocrine responses tostress. In the past 5 years, the gender balance has been somewhatredressed. We identified 200 studies of physiological and