Social Support Hypothesis - an overview (2023)

3 Social Support and Health

Since Cassel hypothesized in 1974 that, under stressful conditions, the lack of social support will lead to a higher risk of becoming ill, research has shown that supportive interactions among people are protective against the health consequences of life stressors. It appears that social support can protect people in crises from a wide variety of pathological states, from low birthweight to death. However, recent research reviews of the empirical evidence of social relationship factors being protective against life stressors have yielded rather inconsistent results. One of the main reasons for such inconsistent findings might be the rich diversity of definitions and conceptualizations of social support and the complexity of the support process. A person’s ability to cope effectively depends on the stressor itself, the type of preferred coping resource, and the resources available to the individual. There are several major types of coping resources, and social support is only one of them.

Empirical evidence for both direct and buffering effects of social support has emerged. In 1986, Hobfoll reported that the direct and stress buffering effects occur simultaneously and that one type of effect is more prominent than the other in a given instance.

An overwhelming amount of evidence shows that social support has effects on both psychological and physical health. One remarkable result for physical health is that social support reduces the risk of dying from disease, suicide, or accidents by about the same percentage as does smoking. Moreover, increased social support from family, friends, and/or colleagues is related to decreased psychological distress. With help and support, people are psychologically better able to manage critical life events, such as unemployment and divorce, as well as everyday problems and hassles.

Social Support and Health

Since Cassel hypothesized in 1974 that, under stressful condition, the lack of social support will lead to a higher risk of becoming ill, research has shown that supportive interactions among people are protective against the health consequences of life stressors. It appears that social support can protect people in crisis form a wide variety of pathological states, from low birthweight to death. However, recent research reviews of the empirical evidence of social relationship factors being protective against life stress yielded rather inconsistent results. One of the main reasons for such inconsistent findings might be the rich diversity of definitions and conceptualizations of social support and the complexity of the support process. A person's ability to cope effectively depends on the stressor itself as well as on the type of preferred coping resource, and the resources available to the individual. There are several major types of coping resources, and social support is only one of them.

Empirical evidence for both, direct and buffering effects of social support has emerged. Hobfoll as one expert on social support says that both direct and stress-buffering effects occur simultaneously and that one type of effect is more prominent in a given instance than the other.

An overwhelming amount of evidence showed that social support has effects on both, psychological and physical health. One remarkable result for physical health is, for example, that social support decreases the risk of dying from disease, suicide, or accidents by about the same percentage as smoking does. Moreover, increased social support from family, friends or colleagues is related to decrease psychological distress. With help and support people are psychologically better able to manage critical live events as unemployment, divorce as well as everyday problems and daily hassles.

Affect in Human-Animal Interactions

It is well known that for many people, animal companions provide emotional support through mutual affectionate attachments (i.e., the human-animal bond, Archer, 1997; Beck & Katcher, 2003). In a recent review, Beetz, Uvnäs-Moberg, Julius, and Kotrschal (2012) summarized their findings on the effects of human-animal interaction on psychosocial and psychophysiological parameters. The authors conclude that human-animal interactions (HAI) have clear positive effects on a variety of psychological states and factors, (1) improvement in social attention, behavior, interpersonal interaction, and mood, (2) reduction in stress-related parameters such as cortisol, heart rate, and blood pressure, (3) reduction in self-reported fear and anxiety, as well as (4) improvement in mental and physical health, especially cardiovascular health. Additionally, limited evidence from a small number of studies suggests that HAI has positive effects on (1) stress-related parameters such as epinephrine and norepinephrine in patients with heart failure (Cole, Gawlinksi, Steers, & Kotlerman, 2007), (2) immune system functioning (Charnetski, Riggers, & Brennan, 2004), (3) pain management in nursing homes (Darrah, 1996), (4) trustworthiness and trust in other persons (Guéguen & Ciccotti, 2008), and in children leads to (5) reduced aggression (Hergovic, Monshi, Semmler, & Ziegelmayer, 2002) and (6) enhanced empathetic skills (Hergovic et al., 2002) as well as (7) improved learning (Gee, Christ, & Carr, 2010).

In view of these positive findings on the effects of HAI on emotions, it is important for emotional robotics to find out the reasons why and how animals affect the mental and physical health of humans. Although a unified framework on the question is not available, three theories that specifically address the emotional effects of human-animal interactions would appear to be important (also see O’Haire, 2010).

The biophilia hypothesis (Wilson, 1984) claims that humans have an instinctive bond and an innate propensity to attend to and be attracted by animals. From an evolutionary standpoint, attending to animals in the wild and understanding an animal’s behavior increases survival chances. In line with this reasoning, a number of studies have demonstrated that animals have calming and relaxing effects on humans (Gullone, 2000), while agitated and aggressive animals are purported to have the opposite effect. By testing different populations in different settings (DeMello, 1999; DeSchriver & Riddick, 1990; Friedman, 1995), it has also been shown that simply viewing calm and friendly animals, and even more so when interacting with them, can reduce anxiety (reduction in cardiovascular, behavioral, and psychological indicators and measures).

The social support hypothesis, in contrast, is based on the idea introduced by Cobb (1976) that social support alleviates the detrimental effects of adverse life events. A wealth of research indicates that social support is very important to well-being under stressful life circumstances and both helps prevent and alleviate depression and anxiety, and aid recovery from physical illnesses (see Thoits, 2011, for a review). In the context of HAI, the social support theory posits that humans and animals function as a source of social support in different ways (McNicholas & Collis, 2006). On the one hand, animals are a source of social support in and of themselves. They provide both emotional support in the form of concern, comfort, reassurance, sense of belongingness, and self-esteem, for example, by encouraging feelings of self-worth and confidence. According to a report by Bonas, McNicholas, and Collis (2000), animals score even more highly than humans on some aspects of social support, namely in terms of providing companionship, nurturance, and a reliable alliance. This finding is probably rooted in their constant availability, their nonjudgmental support, and their unconditional love (see Walsh, 2009). Unlike humans, animals are unable to criticize, retaliate, feel overwhelmed by or reject another’s feelings, and thus they provide a more consistent, demonstrative, and positive mirror that promotes a positive sense of self (see Brown, 2004, for a perspective on human-animal bonding and self-psychology). In addition, a lack of cognitive as well as linguistic abilities means an animal is unable to lie, verbally criticize, or betray trust (Serpell, 2002), which may also explain the positive effects of HAI.

Besides providing direct social support, animals also act indirectly as social facilitators or “social lubricants” (Kruger & Serpell, 2006; McNicholas & Collis, 2006). They provide either a neutral topic of conversation (Fine, 2006), or they affect the beliefs of others about their owner. For example, it has been shown that persons with animal companions are perceived as happier, friendlier, less threatening, more relaxed, and more desirable as acquaintances or friends, even when they belong to groups that are commonly stigmatized (Eddy, Hart, & Boltz, 2001). Woods, Giles-Corti, Bulsara, and Bosch (2007) report that dogs can play a significant role in mediating social relationships ranging from greetings by strangers to deep friendships, and the membership in broad social networks. Thus, animals can indirectly help humans to fulfill their social and emotional needs.

Social cognitive theory is also used to explain the positive effects of nurturing and caring for animals. This theoretical perspective suggests that interaction with an animal enhances self-efficacy (Bandura, 1977) and personal agency. Studies show that animals are effective in both aspects (Allen, Kellegrew, & Jaffe, 2000; Berget, Ekeberg, & Braastad, 2008). This, in turn, may foster a sense of personal achievement and self-esteem (see Catanzaro, 2003), all of which encourage positive affect. Furthermore, the dependence of animals on humans provides a reason for people to adopt health-promoting behaviors toward themselves, and can also help them to schedule the day, e.g., retirees (Dembicki & Anderson, 1996). It has also been demonstrated in clinical samples that caring for and looking after another live being has positive behavioral effects (Allen et al., 2000; Berget et al., 2008).

Last, but not least, human-animal bonding is not unidirectional but dyadic in nature (Odendaal, 2000; Serpell, 2002). Not only do animals fulfill various roles for humans such as that of friend, companion, or servant, but humans also ascribe different roles to their animals (e.g., that of caretaker or teacher). Due to their dependence on humans, animals are provided with nurturance and protection (O’Haire, 2010), and they reward these efforts by reacting positively.

Overall, empirical research on the effects of human-animal interactions has demonstrated that interacting with animals has a number of positive emotional effects on humans, such as reduced stress, anxiety, and fear, and that these can be assessed on a behavioral as well as a physiological level. Human-animal interaction is not only of evolutionary importance but also directly and indirectly supports the human need for social and emotional fulfillment.

1.2 Social reactions to interpersonal violence

According to the stress-buffering hypothesis, social support is especially beneficial to health under times of stress or adversity (S. Cohen & Wills, 1985; Taylor, 2011), and research has increasingly demonstrated the importance of reactions of others in recovery from trauma (S. B. Campbell & Renshaw, 2018; Wagner, Monson, & Hart, 2016). Expectancy violations may be more harmful after trauma, when the ability of social supporters to “step up” is put to the test. Of the forms of trauma, expectancy violations may be even more potent following SA and other forms of interpersonal victimization, given that they are highly-stigmatized forms of trauma (Kennedy & Prock, 2018). When survivors disclose, this creates an opportunity for their positive expectancies of their social supporters to be violated, given that social supporters may have problematic beliefs about victimization or limited knowledge of how to respond due to the stigmatized nature of these experiences (Banyard, Moynihan, Walsh, Cohn, & Ward, 2010). Although many survivors of violence report disclosing to others to obtain help, support, or understanding (Ahrens, Campbell, Ternier-Thames, Wasco, & Sefl, 2007) and may be distressed by a violation of these explicit expectations, any responses that violate basic expectations of caring or concern from social supporters in response to a crisis may be harmful. Indeed, research emphasized the harm of social reactions to victimization that involve betrayal (Freyd, 1994), which could be considered as a violation of an expectancy of protection. In addition, cognitive theories of the development of PTSD emphasize the importance of trauma-related beliefs (Ehlers & Clark, 2000; Resick, Monson, & Chard, 2016), which could include the development or confirmation of beliefs that others cannot or will not provide help or support when needed. In addition, social interactions after victimization may affect beliefs about oneself (e.g., self-blame) (Kennedy & Prock, 2018) as well as beliefs about appropriate coping strategies (e.g., approach versus avoidance) (S. B. Campbell, Renshaw, Kashdan, Curby, & Carter, 2017) that could affect recovery processes.

Following the broader literature on social support, social support offered in response to interpersonal victimization can be considered in terms of both received support and perceptions of that support. Received support in response to interpersonal violence, termed “social reactions,” includes objectively-observable interactions with others regarding victimization, both positive and negative. These social reactions are typically assessed using the Social Reactions Questionnaire (SRQ) (Ullman, 2000). The SRQ was developed by using qualitative data and existing literature to generate a set of reactions commonly received by SA survivors. Survivors are asked to indicate how often they received each reaction from others following victimization. Factor analyses of the SRQ have characterized social reactions both in terms of valence (i.e., positive or negative) and specific types of positive (e.g., emotional support) and negative (e.g., victim blame, distraction) reactions. Overall, the vast majority of survivors who disclose report receiving both positive social reactions (97%) and negative social reactions (98%) (Filipas & Ullman, 2001). Perceived support in response to interpersonal violence, in contrast, reflects survivors' appraisals of these social reactions (e.g., perceptions of the degree to which the social reactions they receive are helpful or satisfying). Unlike the general form of perceived social support described previously, which reflects perceptions of both anticipated and actual support provided, perceived support in response to interpersonal violence reflects perceptions of the actual social reactions received. Importantly, qualitative research has shown that survivors sometimes report positive appraisals of negative social reactions and negative appraisals of positive social reactions (Ahrens & Aldana, 2012; Dworkin, Newton, & Allen, 2018; Lorenz et al., 2018), indicating that perceptions do not necessarily correspond to the actual social reactions received.

Evidence suggests that the relationship of these various types of victimization-specific social support to psychopathology largely parallels the broader social support literature. Specifically, the frequency with which survivors report receiving negative social reactions is associated with worse psychopathology, the frequency with which survivors report receiving positive social reactions has nonsignificant associations with psychopathology, and more positive perceptions of social reactions are associated with lower psychopathology. Next, we review the literature on each of these types of support in relation to psychopathology.

Socioeconomic status

Although not considered in Belsky’s original model, SES and, to a lesser extent, minority status emerged across a variety of domains as important contextual factors influencing the magnitude of associations between predictors and parenting outcomes. Overall, low-SES and minority status (although likely confounded) appear to enhance relations between stressors and impaired parenting. Among mothers in a large meta-analysis with adequate financial resources, for example, the effect size of the negative association between depression and positive parenting was zero; among mothers who were financially disadvantaged, however, this effect size was moderate (Lovejoy et al., 2000). Similarly, a meta-analysis of paternal depression found that the association between depression and more negative parenting was stronger for non-Caucasian fathers. Even more striking, high SES was found to reverse the directionality of negative associations between child negative emotionality and parenting in multiple studies (Crockenberg, 1986; Paulussen-Hoogeboom et al., 2007, 2008). Specifically, child negative emotionality was associated with more sensitive parenting among high-SES samples, but with reduced sensitive parenting among less affluent samples. In addition to increasing the likelihood that stressors will be more strongly related to impaired parenting, low SES also appears to increase the salience of protective factors. Consistent with the buffering hypothesis, social support has been found to be more strongly associated with positive parenting in the context of stress, such as among low-income, homeless, and African American mothers (Lutz et al., 2012; Green et al., 2007; Woody & Woody, 2007).

SES is typically measured using a single factor, most often income or educational attainment (Hauser & Warren, 1997). However, when thinking about associations with parenting, SES, particularly at the lower spectrum, goes far beyond “amount of dollars” or “amount of education” that parents have attained. Low-SES is likely associated with a host of other factors that impact parenting, such as neighborhood quality, access to quality childcare, availability of developmentally-appropriate toys and learning materials in the home, stability of caregivers over time, household chaos, access to high quality and affordable food, and exposure to toxins (e.g., Duncan & Magnuson, 2012; Shaw & Shelleby, 2014). Additionally, low SES is likely related to higher levels of daily and chronic stress, undermining parents’ well-being (Baum, Garofalo, & Yali, 1999). Finally, it also is possible that there are certain traits that predispose individuals to moving into and remaining in low-income brackets and that could undermine parenting (e.g., personality, psychopathology). For example, lifetime history of low-SES status has been associated with high levels of neuroticism and low levels of conscientiousness (Jonassaint, Siegler, Barefoot, Edwards, & Williams, 2011). Thus, when considering direct associations between SES and parenting, or the moderating role of SES on associations between contextual predictors and parenting, it is important to keep in mind that SES itself it very unlikely to be a mechanism, and that it represents a broad and varied influence. Future research would benefit from further exploration of why and how SES influences parenting in early childhood.

It should be acknowledged that the pattern we observed regarding SES and parenting was based on a limited number of studies, and—as is a common problem in the field—there are likely studies reporting null effects of SES that have not been published in the literature or that we failed to find during our review of the literature. Still, we did not purposely search for studies related to SES, nor did we have any initial hypotheses about how SES would interact with context to predict parenting. Thus, we believed the pattern of results that emerged during the course of our review warranted comment. A more systematic review of the literature may be necessary to validate our initial hypothesis that SES increases the strength of associations between contextual factors and early childhood parenting.

4.3 Life-course variations and antenatal stress