Elsevier

NeuroImage

Volume 35, Issue 4, 1 May 2007, Pages 1601-1612
NeuroImage

Neural pathways link social support to attenuated neuroendocrine stress responses

https://doi.org/10.1016/j.neuroimage.2007.01.038Get rights and content

Abstract

It is well established that a lack of social support constitutes a major risk factor for morbidity and mortality, comparable to risk factors such as smoking, obesity, and high blood pressure. Although it has been hypothesized that social support may benefit health by reducing physiological reactivity to stressors, the mechanisms underlying this process remain unclear. Moreover, to date, no studies have investigated the neurocognitive mechanisms that translate experiences of social support into the health outcomes that follow. To investigate these processes, thirty participants completed three tasks in which daily social support, neurocognitive reactivity to a social stressor, and neuroendocrine responses to a social stressor were assessed. Individuals who interacted regularly with supportive individuals across a 10-day period showed diminished cortisol reactivity to a social stressor. Moreover, greater social support and diminished cortisol responses were associated with diminished activity in the dorsal anterior cingulate cortex (dACC) and Brodmann’s area (BA) 8, regions previously associated with the distress of social separation. Lastly, individual differences in dACC and BA 8 reactivity mediated the relationship between high daily social support and low cortisol reactivity, such that supported individuals showed reduced neurocognitive reactivity to social stressors, which in turn was associated with reduced neuroendocrine stress responses. This study is the first to investigate the neural underpinnings of the social support-health relationship and provides evidence that social support may ultimately benefit health by diminishing neural and physiological reactivity to social stressors.

Introduction

Animal and human research has consistently demonstrated that supportive social contact enhances physical health while a lack thereof increases the risk of morbidity and mortality. Indeed, an absence of social relationships constitutes a major risk factor for mortality, comparable to other risk factors such as smoking, obesity, and high blood pressure (House et al., 1988). For example, in a study investigating the impact of social relationships on longevity, individuals with fewer social ties were 2–3 times more likely to die during a 9-year assessment period (Berkman and Syme, 1979).

Despite the robust relationship between social support and health, the mechanisms underlying this relationship remain unknown and the neurocognitive mechanisms, in particular, have been largely unexplored in humans. One hypothesis, which has garnered some support, is that social support reduces physiological stress reactivity to threatening situations, which, over time, can have deleterious health consequences (Uchino et al., 1996). The body responds to stressors, in part, by activating the hypothalamic–pituitary–adrenocortical (HPA) axis, which results in the release of cortisol, a neuroendocrine stress hormone that acts to mobilize energy resources. Although adaptive in the short-run, continuous activation of the HPA axis can be harmful to health (McEwen, 1998). Among other costs, cortisol has immunosuppressive effects, increasing vulnerability to infection and infectious diseases (Ader et al., 1991).

Although much of the work on stress reactivity has focused on the role of the HPA axis in inducing physiological stress responses, other neural regions are likely to play a role in modulating stress reactivity as a function of social support. As proposed by Cohen and Wills (1985), social support may modulate stress reactivity at two different point in the chain of events that lead from potential stressors to physiological stress responses. First, social support may alter the appraisal or perception of potentially threatening conditions such that they are no longer perceived as stressful. Thus, feeling supported and cared for may lead an individual to be less likely to appraise certain conditions as threatening, thus preventing the onset of physiological stress reactivity. To the extent that social support downregulates threat-related reactivity, social support may be associated with less activity in limbic structures that are typically involved in responding to negative or threatening experiences, such as the amygdala, insula, or anterior cingulate cortex (ACC).

The amygdala responds to dangerous or threatening stimuli (Davis, 2001) and has robust connections with the paraventricular nucleus (PVN) of the hypothalamus (Risold et al., 1997, Saphier and Feldman, 1986), which triggers the eventual release of cortisol, implicating the amygdala in the facilitation of physiological stress responses. The insula is thought to be involved in processing visceral sensation (Aziz et al., 2000, Cechetto and Saper, 1987) as well as negative affective states, such as disgust (Phan et al., 2004, Phillips et al., 1997). The insular cortex also has projections to the central nucleus of the amygdala and lateral hypothalamus and thus may play a role in the physiological responses that accompany emotional experience (Davidson and Irwin, 1999). Likewise, the ACC, depending on the location of activity, is involved in both affective experience and stress-related physiological responses. The dorsal portion of the ACC (dACC) has been shown to be involved in the distressing experience associated with physical pain (Rainville et al., 1997), social exclusion (Eisenberger et al., 2003), anxiety (Kimbrell et al., 1999, Wang et al., 2005), and perceived stress (Wang et al., 2005). The subgenual portion of the ACC (subACC) has been shown to be involved in the experience of sadness and is altered in individuals with depression (Mayberg et al., 1999, Phan et al., 2002). Although the subACC is more densely interconnected with regions of the PVN (Ongur et al., 1998), electrical stimulation to either the dACC or subACC (but not the pregenual ACC (pACC), the portion of the ACC anterior to the genu of the corpus callosum; Vogt, 2005) results in cortisol increases (Frankel et al., 1978, Setekleiv et al., 1961) and thus implicates these regions in stress-related physiological responding as well. Moreover, the ACC may be particularly important for the stress-protective effects of social support as this region is activated in response to maternal–infant separation in non-human mammals (Kirzinger and Jurgens, 1982, MacLean and Newman, 1988, Smith, 1945) and thus greater activity in this region may signify an absence of supportive others.

The second point at which social support may reduce physiological stress reactivity is after an event has been appraised as stressful but prior to the onset of prolonged physiological stress responses. Thus, individuals with greater social support may be better able to cope with or regulate negative stressful experiences, leading to reduced physiological stress responses through reappraisal or regulatory processes. To the extent that social support is important for regulating negative responses to stressors, social support may relate to increased activity in regions that are typically involved in regulating negative affect, such as ventrolateral and medial prefrontal cortex (VLPFC; MPFC; Ochsner and Gross, 2005).

VLPFC has been shown to play a role in regulating the distress of physical pain and negative affect (Hariri et al., 2000, Lieberman et al., 2004; Lieberman, et al., in press; Ochsner et al., 2004, Petrovic and Ingvar, 2002). MPFC appears to play a similar role, showing greater activation when individuals regulate, reappraise, or suppress negative emotional states (Phan et al., 2005, Urry et al., 2006). In addition, some have suggested that the ACC may also play a role in emotion regulatory processes (Ochsner et al., 2002, Phillips et al., 2003). Several of these studies seem to implicate the pACC as involved in regulatory processes (Das et al., 2005, Most et al., 2006), although some implicate regions of the dACC (Ochsner et al., 2002) and subACC (Kim et al., 2003, Phelps et al., 2004) as well. Thus, to the extent that social support relates to reduced physiological stress reactivity through enhanced regulatory capacity, social support should recruit greater activity in regions of the VLFPC, MPFC, and possibly ACC.

To investigate the types of neural processes that underlie the stress-protective effects of social support, we investigated how daily levels of social support related to both neurocognitive and cortisol reactivity to a social stressor. To assess daily levels of social support, participants completed a signal-contingent daily experience-sampling task, in which they were loaned a PalmPilot device and, for 10 days, were signaled at random times during the day to report on the degree to which their most recent interaction partner was someone they perceived to be generally supportive. To assess neural reactivity to social rejection, participants completed the Cyberball task within the fMRI scanner (Eisenberger et al., 2003), a virtual ball-tossing game in which participants were excluded partway through the game. To assess cortisol reactivity to a social stressor, all participants completed the Trier Social Stress Task (TSST; Kirschbaum et al., 1993), a task that requires participants to deliver an impromptu speech and perform mental arithmetic aloud in front of a non-responsive, rejecting panel, and, in a meta-analysis, has been shown to reliably elicit cortisol responses (Dickerson and Kemeny, 2004).1

We hypothesized that greater daily social support would be associated with reduced cortisol responses to the TSST, consistent with the notion that social support reduces physiological stress reactivity. We then examined the neural processes related to this social support-cortisol relationship. To the extent that social support relates to attenuated cortisol reactivity through reduced emotional reactivity to stressors, we expected that greater social support and the associated reductions in cortisol responses would be related to reduced activity in limbic regions such as the amygdala, insula, and ACC during social exclusion. Conversely, to the extent that social support relates to attenuated cortisol reactivity through enhanced emotion regulatory processes, we expected that greater social support and the associated reductions in cortisol responses would be related to greater activity in PFC regions involved in regulation, such as VLPFC and MPFC, and possibly regions of the ACC during social exclusion. Last, based on a large body of knowledge relating the hypothalamus to cortisol responses, we also examined the relationships between hypothalamic activity and both social support and cortisol reactivity.

Section snippets

Participants

Prospective participants responded to an ad posted around the UCLA campus, offering $60 in return for participating in the stress challenge component of the study (the TSST). Prospective participants completed a structured phone interview in which they were asked questions about their mental and physical health (e.g., whether a mental health professional had ever diagnosed them with an emotional disorder, whether they were seeing a mental health counselor for any disorder, whether they were

Self-reported responses to the TSST

On average, participants reported feeling stressed by the TSST. Participants reported feeling a significant level of general negative affect in response to the TSST (M = 3.48, SD = 0.98 on a scale ranging from 0 — “did not feel even the slightest bit” to 8—“most you have ever felt in your life;” t(31) = 20.06, p < 0.001 when comparing mean scores with zero, as affect measures were not assessed prior to the TSST). With regard to the more specific measure of anxiety and irritability (“anxious,”

Discussion

The present study investigated the neural correlates underlying the stress-protective effects of social support by examining how neural activity to a social stressor related to both daily levels of social support and cortisol responses to a social stressor. In summary, only two regions of the brain were significantly correlated with both daily levels of social support and cortisol reactivity and are thus strong candidates to play a role in the stress-protective effects of social support in the

Acknowledgments

We would like to thank the staff of the UCLA Brain Mapping Center for their assistance. This research was funded by a predoctoral research fellowship from the University of California, Los Angeles Graduate Division and a postdoctoral research fellowship from the National Institutes of Mental Health to N.I. Eisenberger (T32 MH-019925) and by National Institutes of Mental Health Grants to M.D. Lieberman (R21MH66709, R21MH071521) and S.E. Taylor (R01MH56880).

References (77)

  • L. Naccache et al.

    Effortless control: executive attention and conscious feeling of mental effort are dissociable

    Neuropsychologia

    (2005)
  • J.B. Nitschke et al.

    Functional neuroanatomy of aversion and its anticipation

    NeuroImage

    (2006)
  • K.N. Ochsner et al.

    The cognitive control of emotion

    Trends Cogn. Sci.

    (2005)
  • K.N. Ochsner et al.

    Deficits in visual cognition and attention following bilateral anterior cingulotomy

    Neuropsychologia

    (2001)
  • K.N. Ochsner et al.

    For better or for worse: the neural systems supporting the cognitive down- and up-regulation of negative emotion

    NeuroImage

    (2004)
  • P. Petrovic et al.

    Imaging cognitive modulation of pain processing

    Pain

    (2002)
  • K.L. Phan et al.

    Functional neuroanatomy of emotion: a meta-analysis of emotion activation studies in PET and fMRI

    NeuroImage

    (2002)
  • K.L. Phan et al.

    Neural substrates for voluntary suppression of negative affect: a functional magnetic resonance imaging study

    Biol. Psychiatry

    (2005)
  • E.A. Phelps et al.

    Extinction learning in humans: role of the amygdala and vmPFC

    Neuron

    (2004)
  • M.L. Phillips et al.

    Neurobiology of emotion perception I: the neural basis of normal emotion perception

    Biol. Psychiatry

    (2003)
  • G. Poeggel et al.

    Maternal separation and early social deprivation in Octodon degus: quantitative changes of nicotinamide adenine dinucleotide phosphate-diaphorase-reactive neurons in the prefrontal cortex and nucleus accumbens

    Neuroscience

    (1999)
  • J.K. Rilling et al.

    Neural correlates of maternal separation in rhesus monkeys

    Biol. Psychiatry

    (2001)
  • P.Y. Risold et al.

    The structural organization of connections between hypothalamus and cerebral cortex

    Brain. Res. Rev.

    (1997)
  • D. Saphier et al.

    Electrophysiology of limbic forebrain and paraventricular nucleus connections

    Brain Res. Bull.

    (1986)
  • L.R. Stroud et al.

    Sex differences in stress responses: social rejection vs. achievement stress

    Biol. Psychiatry

    (2002)
  • B.A. Vogt et al.

    Localization of Mu and delta opioid receptors to anterior cingulate afferents and projection neurons and input/output model of Mu regulation

    Exp. Neurobiol.

    (1995)
  • I. Ziabreva et al.

    Mother’s voice "buffers" separation-induced receptor changes in the prefrontal cortex of Octodon degus

    Neuroscience

    (2003)
  • R. Ader et al.

    Psychoneuroimmunology

    (1991)
  • Q. Aziz et al.

    Functional neuroimaging of visceral sensation

    J. Clin. Neurophysiol.

    (2000)
  • L. Baer et al.

    Cingulotomy for intractable obsessive–compulsive disorder

    Arch. Gen. Psychiatry

    (1995)
  • H.T. Ballantine et al.

    Stereotaxic anterior cingulotomy for neuropsychiatric illness and intractable pain

    J. Neurosurg.

    (1967)
  • L.F. Barrett et al.

    An introduction to computerized experience sampling in psychology

    Soc. Sci. Comput. Rev.

    (2001)
  • L.F. Berkman et al.

    Social networks, host resistance, and mortality: a nine-year follow-up study of Alameda county residents

    Am. J. Epidemiol.

    (1979)
  • K. Braun et al.

    Maternal separation followed by early social deprivation affects the development of monoaminergic fiber systems in the medial prefrontal cortex of Octodon degus

    Neuroscience

    (2000)
  • A. Bystritsky et al.

    Functional MRI changes during panic anticipation and imagery exposure

    NeuroReport

    (2001)
  • D.F. Cechetto et al.

    Evidence for a viscerotopic sensory representation in the cortex and thalamus in the rat

    J. Comp. Neurol.

    (1987)
  • S. Cohen

    Social relationships and health

    Am. Psychol.

    (2004)
  • S. Cohen et al.

    Stress, social support, and the buffering hypothesis

    Psychol. Bull.

    (1985)
  • Cited by (372)

    View all citing articles on Scopus
    View full text