The Steroid/Peptide Theory of Social Bonds: Integrating testosterone and peptide responses for classifying social behavioral contexts

TL;DR: The Steroid/Peptide Theory of Social Bonds (S/P Theory), which integrates T and peptides to provide a model, set of predictions, and classification system for social behavioral contexts related to social bonds, is presented.

About: This article is published in Psychoneuroendocrinology.The article was published on 2011-10-01 and is currently open access. It has received 251 citations till now. The article focuses on the topics: Aggression & Poison control.

Summary

1. Introduction

• The close ties between hormones and pair bonds have been studied in a variety of species, with most research focusing on the peptides oxytocin (OT) and arginine vasopressin (AVP).
• The authors goal is to bring T and peptides together to better understand their joint evolutionary significance in pair bonds and other social bonds via the Steroid/Peptide Theory of Social Bonds (S/P Theory).
• Using partner cuddling as a test case, the authors next detail the utility of the S/P.
• Theory for classifying “tricky” behavioral contexts whose “kind” is not immediately apparent or can be misleading.
• A major goal of this paper is to discuss how neuroendocrine responses provide a proximate means for addressing evolutionary questions about pair bonds and other social bonds.

2.1. Evolutionary significance of pair bonds

• Pair bonds may be evolutionarily adaptive, enhancing biparental care and parent—offspring bonds when they promote parent or offspring fitness in some way (Snowdon, 2001; Fernandez-Duque et al., 2009).
• Pair bonds are typically understood to exist between one female and one male.
• The two-point assumption is undertheorized, however, as the presence of a pair bond between two individuals does not necessarily preclude additional pair bonds (see Fig. 1).
• There are a host of species in which multiple females appear to bond with one male.

2.2. Proximate mechanisms for pair bonds

• Hormones often are the proximate mechanisms by which social traits critical to life-history trade-offs are expressed and can be helpful in gaining insights into ultimate function.
• Moreover, researchers have shown that hormone responsivity, specifically, is notably important in studying the comparative evolution of mating systems (McGlothlin et al., 2010).
• Pair and parent-offspring bonds may overlap evolutionarily in aspects of function and mechanism, but also clearly diverge as pair bonds additionally facilitate sexual contact and reproductive potential: then the existence of two systems underlying/contributing to pair bonds is supported.
• As far as the authors know, only humans engage in sexual activity with the intent to reproduce, but most human sexuality is not oriented around reproduction and instead around sexual pleasure, power, relationship maintenance, intimacy, displays of gender, or resource acquisition.

3. The Steroid/Peptide Theory of Social Bonds

• The authors put forth an overarching conceptual umbrella they call the Steroid/Peptide Theory of Social Bonds (S/P Theory).
• Theory has two major aspects and one minor feature, which the authors introduce briefly here but are best examined in the forthcoming sections when noted.
• The first major aspect of the S/P Theory is the S/P Model (Fig. 3), which details how a specifically defined set of behavioral contexts affect T, OT, and AVP, and how these in turn facilitate or inhibit social bonds.
• In contrast, the S/P Model is useful for delineating how two sets of behavioral contexts can affect T and peptides, to either facilitate or inhibit social bonds.
• The Challenge Hypothesis posits a trade-off between low T and parenting (including pair bonds) versus high T and challenges (including mating/sexuality).

3.1.1. The Offspring Defense Paradox

• The Challenge Hypothesis positions all parental behaviors in a low T category (Wingfield et al., 1990), but offspring defense has been linked to high T (Teichroeb and Sicotte, 2008), leading to what the authors call the “Offspring Defense Paradox” (i.e., that parental behaviors are linked to low T, but infant defense increases T).
• Theory, the authors hypothesize that low T is linked only to parental contexts that are perceived/ experienced as nurturant, and high T is implicated in those parental contexts that invoke the need for defensive or protective responses.
• In the S/P Framework (Fig. 4), the authors classify behaviors on the basis of their apparent or evolutionary motivation, and any behavioral context linked to competition, including defending resources (and offspring certainly represent resources), is linked to high T.

3.1.2. The Aggression Paradox

• As the Offspring Defense Paradox (i.e., that all parental behaviors are linked to low T, but offspring defense increases T) exposes a conceptual gap in the social role for T, aggression exposes a similar conceptual gap in the social role of peptides.
• Peptides are most generally tied to intimacy within social bonds (Carter, 1998), yet they also contribute to certain types of aggression.
• In other words, some types of aggression can increase peptides and thus facilitate social bonds, and other types of aggression can increase T and inhibit social bonds.
• The authors speculate that peptides are more salient than T for social bonds, and that this could be reflected via a number of mechanisms.
• Permissive effects of T on AVP (Carter, 2007) also might be at play such that stimuli that increase both T and AVP lead to an androgenic upregulation of AVP (and thus higher AVP).

3.1.3. The Intimacy Paradox

• The Challenge Hypothesis holds that sexual activity should be tied to high T and pair bonds to low T (Wingfield et al., 1990), and does not address links between sexuality and pair bonds directly.
• Sexual activity actually facilitates pair bonds and very often occurs within a pair bond context (Carter, 1998; Snowdon et al., 2006; van Anders et al., 2007a).
• The authors speculate that sexual intimacy is a more salient stimulus for T than nurturant intimacy.
• In other words, there have been stronger selective pressures for a T response to sexual intimacy because of its importance in reproduction.

3.2. The Steroid/Peptide Framework of Social Bonds

• The S/P Framework lays out a more general conceptual hypothesis for the role that T and OT play in social bonds (Fig. 4).
• Though the authors focus on OT, they expect that other peptides and estradiol will fulfill similar roles (Lynch and Wilczynski, 2006; Heinrichs et al., 2009).
• The S/P Matrix (Table 1) makes a set of predictions specifically about the interactions between T, OT, and AVP for behavioral contexts related both to social bonds and their obverse (e.g., missing or dysfunctional bonds).
• There are some “tricky” behaviors that are either difficult to categorize or easy to miscategorize based on cultural (mis)-conceptions, and the S/P Framework can be used to make hypotheses about the evolutionary role of these behaviors based on the hormonal response they elicit.
• Below (in Section 3.3), the authors discuss cuddling as a test case for this use of the S/P Framework.

3.2.1. Testosterone and social goals

• High T is related to competition, that is, social behavioral contexts that involve acquiring or keeping resources.
• Resources are broadly defined and include traditionally conceptualized resources like territory, status, etc., but also sexual/mating opportunities and offspring.
• Actual examples of competitive behavioral contexts could include sexual activity, offspring defense, mate/partner guarding including jealousy, acts designed to acquire status or territory (e.g., fights and status-oriented maneuvering), etc.
• This in many ways overlaps with Wingfield et al.’s (1990) concept of challenge, and recent reviews provide support for links between these behavioral contexts and high T (Hirschenhauser and Oliveira, 2006; van Anders and Watson, 2006b).
• As such, low T is linked to nurturant behavioral contexts—those that involve loving warm contact with others.

3.2.2. Oxytocin and bonding goals

• High OT is tied to social behavioral contexts that involve social bonds, and this can relate to the anticipation of these contexts, the need/desire for these contexts, or their existence (e.g., initiation, development, maintenance).
• Perpetrating rape or offenses that hurt others without the concomitant experience of empathy for victims might also lead to or reflect dysregulation (that itself could sometimes have resulted from early abuse/neglect: Pedersen, 2004).
• Low OT is instead linked to dysregulated social bonds, which differs conceptually from “poor” social bonds.

3.3. Cuddling: a test case of the S/P framework for “tricky” behavioral contexts

• As the authors note above and in Fig. 3, sexual and nurturant intimacy are linked to high OT.
• Theory, the authors were able to develop a hypothesis that cuddling involves sexual intimacy, leading to a better classification of partner cuddling, and supporting the S/P Theory.

4. Gender/sex and the S/P Theory of Social Bonds

• Though there are increasing exceptions, research on nonhuman species with peptides and T has largely been gendered, with OT measured in females, and AVP in males; in contrast, peptide administration studies in humans are more often conducted with men.
• There is evidence for overlapping function (Carter, 2007), as OT and AVP facilitate protective aggression (Bosch, 2011), and both peptides seem to function in males and females (Gouin et al., 2010; Snowdon et al., 2010).
• The authors find that sexual stimuli are more likely to increase T in women than men (van Anders et al., 2007a; Goldey and van Anders, 2011, submitted for publication).
• Thus, there may be gender/sex sensitivities in the specificity or type of hormonal response.
• As an example, certain activities may be high T competitive for men but not women because men have been taught to value those activities as competitive, while women have not (e.g., handshake grip strength); or women have not been encouraged or allowed to engage in those activities, while men have been.

5. Conclusion

• The authors have put forth the Steroid/Peptide Theory of Social Bonds (S/P Theory), highlighting its relevance for understanding evolutionary pathways to pair bonds and other social bonds, as well as its importance in classifying and conceptualizing behavioral contexts.
• Theory provides a way to conduct theoretically rooted experimental work with humans (or other species) that does not rely on more invasive approaches, including administration, that is, by testing hormonal responses to social behavioral contexts.
• In addition, the S/P Theory reflects gaps in the literature in that research has largely focused on OT rather than AVP or prolactin (PRL), and T rather than other steroids.
• By partitioning intimacy and aggression, the S/P Theory also resolves several paradoxes apparent in the literature on social bonds and hormones: the Offspring Defense Paradox, Aggression Paradox, and Intimacy Paradox.
• Role of the funding source – K.L.G. was supported by a Graduate Research Fellowship from the National Science Foundation (grant no. DGE 0718128).

Figures

Figure 3. The Steroid/Peptide Model of Social Bonds. This model demonstrates pathways from aggression and intimacy to T, AVP, and OT and how these hormones in turn affect social bonding. We partition aggression into antagonistic and protective aggression; both types increase testosterone (T), but only protective aggression increases vasopressin (AVP), while antagonistic aggression decreases or has no effect on AVP. We partition intimacy into sexual and nurturant intimacy; both types increase oxytocin (OT), but only sexual intimacy increases T, while nurturant intimacy decreases T. OT and AVP promote, while T inhibits, social bonding. Aggression involves direct conflict that is instigated to be (1) antagonistic: obtaining new territory, status, mates, dominance, etc.; (2) protective: protecting or safeguarding social bonds, e.g. infant defense, partner/mate guarding or defense, etc. Only AVP can be used to distinguish aggression types, since both increase T. Intimacy involves direct close physical contact that is (1) sexual: sexual contact between individuals that may be oriented around pleasure, reproduction, power, etc.; (2) nurturant: warm loving contact between individuals that enhances social bonds. Only T can be used to distinguish intimacy types, since both increase OT. AVP and OT are represented separately for visual purposes, but likely overlap with each other to a large extent (also possibly with PRL); estradiol may also work similarly to OT, and progesterone to AVP. “+” indicates a positive effect; “–“ indicates a negative effect.

Figure 5. The percent change in testosterone (T) in response to exercise, intercourse, and cuddling, with standard error bars. The dashed line indicates 0% change, or no change; points above this line indicate an increase in T from baseline, while points below this line indicate decreases in T from baseline. “*” indicates a significant difference at p < .05; “<” indicates a trend for a statistical difference at p < .10. Adapted with permission from van Anders et al. (2007).

Figure 4. Steroid/Peptide Framework of Social Bonds. This framework lays out a conceptual map of behavioral contexts relevant to social bonds linked to high and low T and OT. The lighter boxes are associated with social bond facilitation/promotion; the darker boxes are associated with the inhibition/dysregulation of social bonds. The framework is divided into a top section for T (social goals) and a bottom section for OT (bonding goals), with the left column for high levels and the right column for low levels of each hormone. The boxes immediately below the high/low T/OT represent the “kind” of behavioral context, or role, for that level of that hormone. Below, at the lowest level for each hormone, are smaller boxes representing a nonexhaustive list of examples of each kind/role.

Figure 1. Examples of pair bonds in multiple configurations: (a) “two-point,” as is typically conceptualized, where each individual is in only one pair bond; (b) “embedded twopoint” for L and N, where L and N are in only one pair bond each, but their partner is not exclusively bonded to them; “multi-point” for M, where M has two pair bonds, and each of these pair bond partners is pair bonded only with M; and (c) “embedded two-point” for L and N; “embedded multi-point” for M and P, where M and P have two pair bonds each, which include a mixture of partners exclusively bonded with them and bonded with others. Letters represent separate individuals, and arrows represent the direction of the bond. An example of (b) is polygyny or polyandry, and of (c) is polyamory for individuals M and P (and potentially L and N, depending on definitions).

Table 1. The steroid/peptide matrix of social bonds. Up and down arrows are relative to population norms but could relate to individual norms as well. “T” indicates testosterone.

Figure 2. Ultimate functions and proximate mechanisms of pair and parent-offspring bonds. Predictions based on this model would be (1) if pair and parent-offspring bonds elicit similar hormonal responses, then a shared evolutionary history is supported and (2) if sexual and nurturant intimacy within pair bonds elicit divergent hormonal responses,

Frequently asked questions

Q1. What are examples of nurturant behavioral contexts?

Actual examples of nurturant behavioral contexts could include grooming, feeding, pair bond existence, huddling, and other close warm contact. 

Q2. What is the role of hormones in the evolution of pair bonds?

Given that parent-offspring bonds are likely to be evolutionarily older, pair bonds may be predicated upon a neuroendocrine system that evolved to support parent-offspring bonds but in general promotes nurturance (Fisher, 1992; Carter, 1998; Fernandez-Duque et al., 2009). 

Q3. What did the authors test in the follow-up study?

The authors used the S/P Theory to develop a hypothesis about the tricky behavioral context of cuddling, a hypothesis which the authors tested in a follow-up study. 

Q4. What is the definition of pair bonds?

Like others, the authors define pair bonds as long-lasting affiliations involving intimacy, sexualcontact, preferential proximity, and emotional attachment with relative exclusivity (Hawkes, 2004). 

Q5. What is the role of OT in human social bonds?

Actual examples of social bonding could include sexual intimacy, nurturant intimacy, loneliness (i.e., the need for social bonding), social conflict resolution (e.g., arguments with loved ones that arise out a of need to strengthen a bond), etc. Consistent with the above, OT facilitates social cognitions and empathy in humans (Bos et al., 2011). 

Q6. What are the characteristics of pair bonds?

They are generally defined by social and sexual “monogamy,” and though extra-pair sexual contacts occur, pair bonds still limit sexual access to others. 

Q7. What is the importance of incorporating T into research on intimacies?

Their example with cuddling highlights the importance of incorporating T into research on intimacies, even though intimacy is typically studied only in conjunction with peptides, and T is only studied in conjunction with competition. 

Q8. What is the reasoning for a low T nurturant behavior?

This reasoning should apply only to partner cuddling, and the authors predict that parent-child cuddling should decrease T as a low T nurturant behavior (unless it is experienced as infant defense, which should accordingly increase T; a testable viable alternative hypothesis). 

Q9. What is the role of OT in social bonding?

Though findings link high OT with pair bonds and partner closeness, other research demonstrates a complementary role for OT, that is, as tied to the need or desire for social bonds. 

Q10. What is the relationship between low T and nurturance?

Low T is related to nurturance, i.e. social behavioral contexts that involve loving warm contact with others (e.g. partners/mates, offspring, friends, etc.) (and this may possibly transpire via conversation of T to estradiol, and estrogenic facilitation of peptides). 

Q11. What is the role of T in the evolution of pair bonds?

T is a favorable candidate for testing these distinct evolutionary functions, since it is positively linked to sexuality but negatively linked to nurturance (van Anders and Watson, 2006b; Ziegler et al., 2009), and because it is so strongly implicated in tradeoffs relevant to pair bonding (Wingfield et al., 1990; Ketterson et al., 2005; Bales et al., 2006; van Anders and Watson, 2006b).