individuals reported poorer sleep quality, longer sleep latency, longer perceived sleep duration, and greater daytime dysfunction due to sleepiness than socially embedded individuals (Buysse et al., 1989). Other data confirmed that lonely individuals slept less efficiently, took slightly longer to fall asleep, evidenced longer REM latency, and had more frequent microawakenings during the night than embedded and normal individuals (Cacioppo et al., in press b). Thus, the restorative act of sleep was more efficient and effective in socially embedded than in lonely individuals.

The most extensive research on how social ties influence health, however, pertains to the underlying physiological routes (e.g., Uchino et al., 1996; Seeman, 1996; Seeman and McEwen, 1996; Cohen and Herbert, 1996; Kang et al., 1998; Kiecolt-Glaser et al., 1994). Meta-analyses of the experimental literature support the hypothesis that perceived social isolation is associated with physiological adjustments, with the most reliable effects found for blood pressure, catecholamines, and aspects of both cellular and humoral immune function (Uchino et al., 1996; see also Seeman and McEwen, 1996). In a study of carotid atherosclerosis in middle-aged men, higher intima media thickness of the carotid artery was found in those living alone than in cohabitating counterparts even after controlling for age, health status, education level, saturated-fat consumption, and smoking (Helminen et al., 1995). Although individual differences in personality play a role, the biological effects of loneliness are evident even after controlling for common individual differences (e.g., extroversion, neuroticism) and in intervention studies designed to reduce social isolation and improve physiological functioning (Cacioppo et al., in press b; Uchino et al., 1996). People 's beliefs, attitudes, and values pertaining to others appear to be especially important, because subjective indices of social isolation have been found to be more powerful predictors of stress and health than objective indices (e.g., Seeman et al., 1997; Uchino et al., 1996).

Social relationship researchers looking at attachments in early and later life and at close personal relationships, including marital and family ties, have described some features of deep, meaningful, loving human connections (Ryff and Singer, 2000). Numerous investigations have examined the nature of affect in intimate relationships, its developmental course over time, and related expressions of emotion during marital interaction (e.g., Carstensen et al., 1995, 1996; Gottman, 1994; Gottman and Levenson, 1992). Collectively, research on interpersonal flourishing gives greater attention to the emotional upside of significant social relationships and their consequences for improved health (see Ryff and Singer, 1998, 2000; Taylor et al., 2000). Of particular importance are studies that track the cumulative long-term features of social relational experience and its biological sequelae. Individuals on positive relationship pathways (positive ties with parents in childhood, intimate ties with spouse in adulthood) are

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement