The Evolutionary Psychopathology of Human Pair-Bonding

Abstract

Evolutionary psychiatry views biological systems as evolved functional adaptations shaped by selection to promote survival and reproductive success (Darwinian fitness). However, even functional systems can malfunction under certain conditions resulting in pathological and psychopathological states. We argue that the human mating system based on pair-bonding and biparental investment is a fundamental human, species-typical characteristic that is supported by a bundle of evolved psychological adaptations. Here we describe a framework for understanding the biological roots of a subsection of recognised mental disorders as psychopathologies of the Human Mating System. We propose that this framework illuminates why such psychopathologies exist and why they present in the way they do. The framework also explains why such disorders are not found in species with alternative mating systems including those phylogenetically close to Homo sapiens. We argue that such functional, phylogenetic approaches can serve as a template for the understanding of range of other psychopathologies.

1. Introduction

Theoretical and empirical work within evolutionary psychology have led to significant advances in our understanding of the nature of human sexuality and mating strategies (e.g. Buss, 2008). It suggests that humans follow a dual strategy of short and long-term mating and that each of these strategies has its own distinct set of characteristics that can be gender specific (Buss, 2008). However, there has been less emphasis within this literature on identifying and specifying the human, species-typical mating system, exploring its phylogenetic roots and examining its full implications for shaping human psychology. As a result, an erroneous interpretation can arise that presumes that the short and long-term mating strategies are of equal significance for human reproduction and in shaping human mating psychology (Abed, 2025).

In all human societies studied by anthropologists, the primary human mating strategy involves some variant of long-term mating. The distinctive characteristics of the human mating system includes pair-bonding and paternal investment (Alger et al., 2020; Schacht and Kramer, 2019). We contend that humans demonstrate a species-typical mating/reproductive strategy the details of which can help make sense of patterns of human reproductive activity both cross-culturally and, over human evolutionary history, while acknowledging the extraordinary plasticity and diversity of the granular details of reproductive arrangements in response to individual ecological and cultural influences. This plasticity is further attested to (for instance) by variability in testosterone levels among men in relation to a range of factors such as mating effort and the dominant societal mating system (monogamous vs polygynous) (e.g. Alvergne et al, 2009).

We draw attention to the fact that the central element of the human mating system is the long-term breeding bonds (the pair-bond) between males and females which has a clear biological foundation in evolved physiological, neurobiological and psychological adaptations (Chapais, 2008). The biological foundation of the human pair-bonds is further attested to by the existence of a ‘transculturally invariant social core’ in all types of marital bonds and families (Chapais, 2008).

Furthermore, we argue that a novel examination of the biological foundations of the human pair-bond and its central importance in human mating and human evolution is through the identification of evidence of psychopathology that emerges from dysfunctional and/or disordered states of the evolved psychological adaptations that support pair-bonding. This serves the dual purpose of providing a better understanding of the evolutionary roots of a range of seemingly disparate psychopathological states as well as illuminating the adaptive function and significance of those systems in their non-pathological states.

This proposal must not be mistaken for a circular argument. The evidence for the human pair-bonded, long-term mating system with a significant role for fatherhood and paternal investment is evidenced independently through extensive primatological and anthropological studies and stands regardless of the existence of the disorders and dysfunctions described in this paper. However, in recognising these distinctive features of the human reproductive system, we pose the evolutionary question: what kind of breakdowns, dysfunctions and disorders can we expect in a mating system with the above characteristics?

We have arrived at the current framework through standard scientific methodology followed in evolutionary psychiatry that starts with abduction to the best possible explanation (hypothesis) then proceeds to prediction-making followed by the expectation of empirical testing (Griffin et al, 2025) although, as is the case with theoretical formulations, the third step (empirical testing) remains outstanding.

Such a Darwinian adaptationist perspective can organise a wide array of observations and specifically illuminate a subsection of mental disorders by providing an explanation as to why these disorders appear to be human-specific and have no clear analogues or exact homologies, in non-human great apes. This framework demonstrates the power of evolutionary theory as an epistemic tool in psychology and psychiatry. The approach enables a novel functional understanding of a range of psychological/psychopathological phenomena through identifying them as disorders of human mating psychology, helping to explain why these phenomena exist and why they present in the way they do.

In general terms, Darwinian fitness refers to the measure of an individual organism’s reproductive success or the ability of an organism to pass on their genes to the next generation within a given environment. It can be seen as having two major components, 1) survival and 2) the ability to successfully reproduce.¹ By inference any enduring rather than transient illness (especially during the reproductive years), will likely interfere with either or both of these facets of fitness. Hence, fitness can be affected by a range of physical and/or mental disorders. Whereas many enduring mental disorders may have secondary fitness costs and may indirectly affect the mating system, which requires many functional mental abilities, our focus is on mental conditions and disorders that arise within and primarily affect the human, species-typical mating system. In other words, we are examining disorders presenting to clinicians, from a different level of abstraction. Customarily disorders are grouped together in clinical symptom clusters for clinical purposes (DSM and ICD) and only some parts of the classification systems (e.g. organic, substance misuse), are based on aetiology or system dysregulation. This realigning of the relevant conditions with evolutionary, functional and phylogenetic roots has particular advantages in terms of future research and possible interventions.

2. The evolution of the pair-bonded humanmating system

The human mating system differs significantly from that of all other great apes and represents a radical departure from that of the chimpanzees and bonobos, our closest ape relatives. This observation indicates that the human mating system evolved within the hominin lineage after separation from the chimpanzee (and bonobo) line (Abed, 2024). Although human mating practices show significant variation across cultures and can include components of both short and long-term mating strategies (Schmitt, 2005), it is possible to identify common features that characterise a universal human, species-typical mating system. Importantly, the short and long-term mating strategies that are identified and extensively studied in the evolutionary psychological literature, are not of equal importance in understanding human reproduction and human mating psychology.

Anthropological studies have demonstrated that the distinctive characteristics of the human mating system cross-culturally, include pair-bonding and importantly, substantial paternal investment (Alger et al, 2020; Schacht and Kramer, 2019). However, whereas paternal investment varies significantly between human populations in both extent and significance (Hrdy, 2024), alloparenting (the care of offspring by individuals other than their biological mothers) is both a ubiquitous and distinctive/core feature of child care in human subsistence and traditional societies and is undertaken by a range of individuals including older siblings, grandmothers, other female relatives (and non-relatives) as well as fathers (Chaudhary et al, 2023).

Marriage, which is based on pair-bonding, is a unique human characteristic that is considered a universal, species-typical trait (Shenk, 2024). Also, while some polygyny occurs in the majority of societies studied by anthropologists, monogamy is the dominant marriage system in all societies at any given moment in time (Shenk, 2024). For example, studies of foragers show that on average, only 14% of married men are polygynous, and only 21% of married women have co-wives (Marlowe, 2005).

Studies in hunter-gatherers suggest that marriages, while not life-long, frequently endured over periods of years. For example, a study of the Hadza showed that 60% of marriages endured for 5 years, 40% for 10 years and 16% for 25 years (Blurton-Jones, 2016). Also, the average number of marriages in the Jo/’hoansi and the Hadza hunter-gatherers was two (Howell, 1979; Blurton-Jones, 2016).

Interestingly, only around 9% of all mammals are socially monogamous (Lukas and Clutton-Brock, 2013) and of these species, an even smaller fraction shows direct or indirect paternal care. However, paternal provisioning is common in human subsistence societies, while rare among primates and almost totally absent in the great apes (Alger et al, 2020). The prevalence of social monogamy among primates is significantly higher than the average for mammals at around 29% but, interestingly, there are no examples of this among the non-human great apes (Lukas and Clutton-Brock, 2013).

Thus, across all human societies the predominant mating system is monogamous, and humans are therefore unique in this regard among the great apes (Conroy-Beam et al, 2016). This has major psychological implications in humans.

While extra-pair paternity (which critically affects the reproductive significance of short-term mating strategies) is hotly debated (Scelza, 2024), current literature suggests that the levels in humans are particularly low at around 1-2% (Larmuseau et al, 2016). A more recent study using patrilineal genealogies from the Low Countries spanning a period of over 500 years and Y-chromosome genotyping of living descendants, revealed historical extra-pair paternity ranging from 0.4% to 5.9% (Larmuseau et al, 2019). In comparison, extra-pair paternity regularly exceeds 20% in monogamous birds (Schacht and Kramer, 2019).

The human reproductive system shows other species-typical peculiarities. These include infant weaning before nutritional independence and a particularly short inter-birth-interval (Kaplan et al, 2000). It has been estimated that it takes around 13 million calories to raise a single infant from birth to nutritional independence (Kaplan, 1994). Thus, unlike other great apes, it would have been impossible for ancestral human mothers to provide such a massive investment unaided and therefore provisioning needed to be supplemented by other adults within the group (Hrdy, 2009). The problem is compounded as dependent offspring in need of provisioning are ‘stacked up’ as a result of the shorter inter-birth-interval which multiplies the provisioning requirements (Hrdy, 2009). Two primary sources for this supplementary caloric investment stand out besides the mother, which are fathers and grandparents (Hooper et al 2015). As alluded to earlier, while men are high paternal investors compared to all other great apes and mammals generally, paternal care is however facultative rather than obligatory and the significance of the father’s presence to child survival varies across human societies, influenced by cultural and ecological factors (Marlowe, 2000; Shwalb et al., 2013). Nevertheless, unlike other great apes and most mammals, men generally produce a considerable surplus of calories (primarily through hunting) during their peak adult years and, a proportion of this surplus is potentially available for investment in mate and offspring (Kaplan et al, 2000).

It is important to underline that pair-bonding, monogamy and paternal care and investment are distinct phenomena and must not be conflated.

For example, it is notable that monogamy has evolved frequently in the absence of paternal investment (Muller and Pilbeam, 2017). Hence, the subset of species that exhibit both, paternal investment probably followed the adoption of monogamy rather than evolving with it (Chapais, 2013; Grueter et al, 2012). Also, some degree of paternal care has been reported in chimpanzees despite the lack of pair bonds and monogamy (Murray et al, 2016). Therefore, monogamy itself is not a good proxy for investment in parenting over mating (Del Giudice, 2025). In addition, pair-bonding can exist in monogamous and polygynous mating systems and both systems are compatible with paternal investment and provisioning or as Chapais (2008, p 177) put it ‘monogamy is maximally constrained polygyny’.

In summary, the importance of the long-term pair-bond in human evolutionary history cannot be overstated. The human pair-bond has been of pivotal importance in enabling paternity recognition, parental collaboration, the recognition of affines (in-laws) and uniquely, in the emergence of tribal level organisation in humans. Thus, the pair-bond has been instrumental in determining the deep and distinctive structure of human society (Chapais, 2008).

3. The biological foundations of the pair-bonded, human mating system

It is argued here that the human, species-typical pair-bonded mating strategy is supported by a bundle of evolved psychological and physiological-anatomical adaptations/traits. These were shaped by selection due to their contribution to fitness within the pair-bonded, long-term, paternal investing, human mating system. Examples of bodily adaptations that have arisen in the human lineage, include the distinctive, nubile female hour-glass shape and the permanently “swollen” female breasts (Jasienska et al, 2004; Lassek and Gaulin, 2019; Marlowe, 1998). These female physical traits appear to be adaptations that provide “honest” signals for reproductive potential where this is significant only in the context of a long-term mating system that involves significant male investment. In other great apes, reproductive potential seems to have little or no significance.

Interestingly, human males show a range of hormonal and brain changes when becoming fathers, not seen in those individuals who never become fathers (e.g. testosterone levels drop and the Anterior Cingulate Cortex becomes thicker) (Hrdy, 2024). These physiological changes are evidence that evolution has shaped men’s biology and psychology for parenting and provisioning of mate and offspring.

The focus of this article is on the particular psychological and behavioural adaptations related to long-term human mating and specifically on ways that some of these adaptations can create vulnerability to dysregulation and dysfunction under certain conditions. Given that the human mating strategy cluster is exclusive to our species and not shared by the other great apes, we argue that many of the manifestations of dysregulation of these adaptations are therefore also human-specific. These adaptations are human, species-specific and evolutionarily recent and novel. A non-exhaustive list of psychological adaptations includes:

• Romantic love (the psychological adaptation for pair-bonding) (e.g. Bode and Kushnik, 2021; Fletcher et al, 2015)

• Sexual jealousy (strategy for mate retention and paternity assurance) (e.g. Platek & Shackelford, 2006).

• Female intra-sexual competition for mates (evolved female strategy for competition for investing males, relevant to long-term mating strategy) (e.g. Abed, 2024; Fisher, 2017).

4. Mental disorders related to the pair-bondedhuman mating system²

The psychopathological conditions related to the adaptations for the human mating strategy may be viewed under two broad headings. These are:

A. Vulnerabilities arising from adaptations related to the human mating system:

• The Couvade syndrome (adaptive/maladaptive response to fatherhood)

• Depression secondary to breakup of romantic relationships (adaptive/maladaptive response to failure of pair-bonding)

• Postpartum depression in fathers (adaptive/ maladaptive response to fatherhood and to maternal postnatal state).

• Paternal depression after the death of an offspring (adaptive/maladaptive response to fatherhood).

B. Dysregulated adaptations related to the human mating system³:

• Morbid jealousy (dysregulation in mate retention/paternity assurance)

• Erotomania (De Clerambault’s syndrome) (dysregulation in mate attraction and romantic love)

• Eating disorders (dysregulation of female competition for mates)

• Borderline Personality Disorder (dysregulation of pair-bonding)

It must be noted that the framework utilised to identify and classify these disorders is based on ultimate rather than proximate causation (Tinbergen, 1963). Hence, we do not claim to identify the neurobiological systems or other mechanistic causes (brain circuits, neurotransmitter systems) for these disorders. We also recognise that the psychological adaptations related to the human pair-bonded mating system are complex and multi-faceted and therefore, this proposal should be viewed as a high-level classification based on ultimate function.

Nevertheless, we recognise that question-marks linger regarding what we mean by the term ‘disorder’ and whether evolutionary thinking can help clarify this frustratingly difficult-to-define concept (Nesse, 2001). Interestingly, both the DSM-5 and the RDoC refer to ‘dysfunctions’ as underlying disorders but neither system specifies what counts as a dysfunction (Del Giudice, 2018).

A promising and especially fruitful evolutionary concept is Wakefield’s ‘harmful dysfunction’ (HD) approach (e.g. Wakefield, 1992, 2015). In the HD approach, to fulfil the dysfunction criterion, a condition must be caused by a failure of a biological mechanism to perform its evolved function and additionally this must inflict harm or damage on the affected person as judged by socio-cultural norms. Although, the subject of considerable criticism, the HD framework provides a useful heuristic for thinking about disorders (Del Giudice, 2018). While strict adherence to Wakefield’s HD system would lead to the exclusion of a whole range of distressing conditions from classification systems, a more pragmatic approach would be to make a distinction between narrow-sense disorders and broad-sense disorders (i.e. undesirable conditions that are currently diagnosed as disorders but may or may not represent genuine dysfunctions) (Del Giudice, 2018).

Taking this view, the list of disorders considered in this paper includes a number of narrow-sense disorders that likely involve true HD’s (morbid jealousy, erotomania, moderate/severe anorexia and bulimia nervosa and severe borderline personality disorder). The 3 variants of depression listed are likely to represent broad-sense disorders which should be the object of interest by the caring professions but may or may not represent genuine dysfunctions. The Couvade syndrome, however, stands out as having an ambiguous status being characterised as benign and self-limiting (e.g. Brennan et al, 2007).

4.1 The couvade syndrome:

This is also known as sympathetic pregnancy, a psychological, psychosomatic or behavioural condition whereby expectant fathers experience physical and psychological symptoms similar to those of their pregnant partner. While not officially recognized as a psychiatric disorder, it is often described as a psychosomatic phenomenon. The Couvade syndrome can be considered a form of extreme male empathy for his pregnant partner. Its presentation is highly dependent on cultural context. However, the symptoms are usually benign, self-limiting, transient and resolve after childbirth. The term can also refer to certain rituals, varying in different cultures, that fathers adopt during their partner’s pregnancy. Couvade is first recorded in Ancient Egypt (Zogbe, 2007).

International studies reveal contradictory findings in the incidence and type of symptoms. This may reflect methodological problems in the syndrome’s definition or criteria and type of measurement across studies. The syndrome’s incidence in the U.K. is largely unknown, but the range of early estimates varies between 11% and 50% (Trethowan and Conlon,1965). In Sweden, Bogren (1984) found an incidence of 20%. However, higher incidence rates (up to 97%) were also reported in the USA by Clinton (1987) while Khanobdee et al. (1993) estimated the incidence in Thai men at 61% and Marilov (1997) reported an incidence of 35% in Russian men. A more recent literature review has similarly found wide variations in incidence cross-culturally (Couto et al, 2022). The extreme variability in incidence across studies is likely the result of the absence of the Couvade syndrome from all official classifications of diseases and hence differing criteria are used in different studies.

Couvade’s cultural-ritual origin can be examined in many traditional societies, where it involves fathers mimicking birth, such as fasting, going to bed, or even acting out labour. The proposed social benefits of such rituals include providing affirmation of paternity, fostering the belief that the rituals deflect spiritual harm from the mother onto the father (seen in groups like the Garifuna and Papuan societies), and, helping reinforce family and social bonds (Brennan et al., 2007).

There are some instructive cross-species parallels. For example, Wynne-Edward and Reburn (2000) reported that the males of monogamous marmosets and tamarins show ~10% weight gain when their mate is pregnant, due to hormonal changes. Also, it is of interest that in monogamous species like prairie voles and California mice, male partners exhibit hormonal shifts (prolactin, oestradiol, cortisol, testosterone) when the female is pregnant. These hormonal changes are tied to paternal behaviour and emotional bonding, sometimes leading to shared symptoms which sometimes may be viewed as pathological, like appetite changes, fatigue, or emotional support behaviours (Bales et al, 2007; Insel and Young, 2001).

Hence, the available literature suggests that the Couvade syndrome exists cross-culturally and over wide swathes of historical times although the exact presentation is greatly influenced by cultural context. The core features of the condition (symptoms and behaviours relevant to pregnancy of partner) are what interests us in the present context rather than the question of which features are adaptive or dysfunctional. The existence of the Couvade syndrome raises the question: Why should such a puzzling cluster of symptoms and behaviours arise repeatedly in a proportion of men in diverse social and historical settings? We suggest that the only plausible answer is that they are direct manifestations of the human pair-bonded mating system that involves male commitment and male investment in mate and offspring. This pair-bonding system has shaped various aspects of human psychology and biology (e.g. Storey et al, 2000) which can manifest themselves in both functional and dysfunctional states. It is likely that the Couvade syndrome includes both. It should be clear from the above that the Couvade cluster can make no sense and would not even be feasible in a non-pair-bonding species e.g. the promiscuous system of the chimpanzees.

Prediction 1: we predict that any analogue of the Couvade in non-human primates will occur exclusively in species that have a pair-bonded mating system. This may include a subset of the 29% of socially monogamous primates and will not occur in species with other mating systems.

4.2 Depression secondary to romantic rejection

Romantic rejection is a recognised risk for depression (Kelly et al, 2016; Rantala et al, 2017). However, as the focus of this paper is on psychopathology relevant to pair-bonding it is important to first understand the context of the romantic rejection. Mate rejection in the sense of denial of mating access can occur within a range of mating systems including those that do not involve any form of pair-bonding. We are particularly interested in depression secondary to relationship breakup, which is a specific subtype of romantic rejection that occurs after a bond has been established i.e. romantic relationship breakups that occur where bonds are formed then dissolved for whatever reason. We suggest that for the dissolution of romantic bonds to have significant psychopathological sequelae such as depression indicate that these bonds were psychologically significant to the subject and hence, point to the likelihood that such bonds have an evolved function and biological foundation. We argue that, as the pair-bond system is the linchpin of human reproduction, any serious threat to or dissolution of this bond represented a significant threat to fitness for ancestral humans, especially if this occurred within the context of infidelity or mate poaching. It is not surprising, therefore, that early stages of romantic love are associated with euphoria (signalling increased fitness prospects) whereas romantic rejection can lead to a profound sense of loss, negative affect and clinical depression (Fisher et al., 2010).

A number of studies suggest that romantic rejection/ romantic relationship breakups are associated with increased risk of distress and psychopathology. For example, a study examining the lifetime prevalence of anxious misery, fear, and externalizing disorders among mothers in the US National Comorbidity Survey (N=1,534) found that separated/divorced mothers compared to married mothers had increased odds of having any anxious-misery disorder, depression, dysthymia, generalized anxiety disorder and posttraumatic stress disorder (Afifi et al., 2006). Also, a large Canadian population health survey found that the high prevalence of major depression in separated or divorced individuals was due both to an increased risk of marital disruption in those with major depression, and also to the higher risk of this disorder in those with divorced or separated marital status (Bulloch et al., 2009).

While the psychiatric morbidity associated with divorce, relationship break-ups and romantic rejection are widely recognised by mental health professionals, the evolutionary roots of this vulnerability are likely to be underappreciated. It is an open question as to whether patients suffering from this kind of depression will find it therapeutically beneficial to reflect on this evolutionary inheritance.

Prediction 2: We predict that analogues for this type of depression in non-human primates will only be detected in species that have a pair-bonded mating system. This may include a subset of the 29% of primate species that are socially monogamous as well as pair-bonded polygynous species and will not occur in species with other types of mating systems.

4.3 Paternal post-natal depression

Paternal post-natal depression (PPND) is less well studied than maternal post-natal depression (PND). Nevertheless, there is growing recognition that it is a significant mental health issue. For example, Kim & Swain (2007) estimated the prevalence of PPND at between 4-25% with high co-morbidity with maternal PND, while a large meta-analysis estimated the prevalence of PPND to be around 10.4% within the 3-6 months following childbirth with a moderate correlation with maternal PND (Paulson & Bazemore, 2010). A more recent meta-analysis found an overall prevalence of PPND to be around 8% internationally with higher rates in the US of 12.8-14.1% (Cameron et al., 2016).

The significant comorbidity with maternal PND is of interest in the present context as it shows obvious similarity to Couvade (see above).

In addition, fatherhood has been found to be consistently associated with a range of hormonal changes. These most prominently include a reduction in testosterone levels (Gettler et al., 2011). Also, interestingly, there is evidence of parental cortisol synchrony during the prenatal period, with significant testosterone, prolactin, and progesterone covariations in the time leading up to childbirth. During the postpartum period, parental synchrony was reported for oxytocin, testosterone, and cortisol levels. The implications of these covariations were found to translate into adaptive parenting behaviours and the facilitation of the romantic bond (Daneshnia et al., 2024).

Questions regarding the possible adaptiveness or maladaptiveness of depression in new fathers do not directly concern us here. Our purpose is to establish whether the existence of PPND is evidence of an evolved adaptation for fatherhood via the pair-bonding system. We suggest that the available data amply supports the contention that PPND does indeed provide such evidence. The hormonal synchrony in mothers and fathers and drop in testosterone levels in fathers associated with partner’s childbirth are tangible biological changes and so is the significant correlation between maternal PND and PPND.

It is clear that PPND can only exist in a pair-bonded mating system where paternity and fatherhood are fairly identifiable. Additionally, in humans, fatherhood frequently comes with a set of additional provisioning and other parental responsibilities. We hypothesise that these roles and responsibilities are supported by a range of parenting psychological adaptations that served to increase the survivability of offspring in ancestral settings and that these evolved adaptations give rise to the vulnerability to PPND. Thus, we suggest that this formulation puts the phenomenon of PPND in its phylogenetic context and explains its evolutionary roots.

How the risk of PPND may differ in polygynous as opposed to monogamous marriages is currently unknown.

We appreciate that some evolutionary models of maternal post-natal depression propose that maternal depression functions to induce the pair-bonded partner to increase his support/investment (e.g. Hagen, 1999). However, there are equally credible, extant evolutionary formulations of maternal PND that do not require pair-bonding e.g. mismatch models (Hahn-Holbrook & Haselton, 2014). Hence, we have not included maternal PND in the list of disorders of the pair-bonded human mating system.

Prediction 3: We predict that any analogues to PPND that may be found in non-human primates will have both pair-bonds and paternal investment. This may include a subset of the 29% socially monogamous primates and will not occur in species with other kinds of mating systems.

4.4 Paternal depression after death of anoffspring

Another depression-related phenomenon, somewhat similar to PPND, is paternal bereavement after the death of a child. We should emphasise that the grief reaction is not a human-specific phenomenon and neither is it specifically related to any particular mating system. However, for paternal depression following the death of a child to occur certain conditions must apply. These include some degree of paternity identification and emotional bonding between father and child. These conditions do not and cannot exist in the promiscuous system of the chimpanzees and bonobos where paternity remains unknown and very little meaningful father-offspring bonding can or does take place. Maternal grief following death of an infant is not uncommon in a range of mammalian species. Interestingly, chimpanzee and other primate mothers have been observed to carry their dead infants for days or even weeks in a state analogous to human grief (Soldati et al, 2022).

However, in humans, fathers can experience very prolonged grief, depression, guilt, and existential crisis after infant loss (Aho et al, 2006; Vance et al., 1995). We therefore argue that this grief reaction and the associated risk of depression arises as a result of the pair-bonded human mating system where fatherhood is “known” and father-offspring emotional bonds are common.

Prediction 4: We predict that analogues to human paternal grief and depression will exclusively be seen in species that show significant paternal care and investment and will not occur in species with other types of mating systems.

4.5 Morbid jealousy

While many emotions can have multiple and complex functions that defy simple classification, sexual jealousy is an exception to this rule. Sexual jealousy is an emotional adaptation with clear design features aimed specifically at preventing infidelity in a romantic partner. It is activated by cues of probable infidelity of a romantic partner and specifically motivates a suite of behaviours aimed at minimising this risk (e.g. Platek & Shackelford, 2006). The evolutionary psychology dictum that ‘to understand the key you need to understand the lock’ (Sell & Sznycer, 2024) lends itself particularly well to the understanding of sexual jealousy, both in its functional and dysfunctional forms. However, unlike current evolutionary psychology analyses that begin with the adaptive problem of infidelity we take a phylogenetic perspective that views the human mating system in comparison with other great apes, primates and mammals generally. This reveals that the human pair-bonded reproductive system with its biparental investment is both unknown in other great apes and being a recently evolved system within the human lineage. Hence, we propose that both the adaptive problem (infidelity), and its solution(s) (e.g. mate guarding, vigilance, anger etc.) that correspond to the lock and its key are also recently evolved adaptations and human species-typical traits.

There have been suggestions that other monogamous primates (e.g. titi monkeys) (Maninger et al., 2017) show possible features of jealousy. However, given the phylogenetic distance from humans it is likely this has arisen through convergent evolution rather than homology where sexual jealousy (or its non-human equivalent) tends to evolve whenever monogamy, pair-bonding and paternal investment exists⁴.

It is important to note that ancestral males and females faced distinct recurrent challenges. Men risked cuckoldry (and investing in a rival’s progeny) whereas women risked being abandoned with one or more dependent offspring and left to provide for them unaided. Evolutionary psychology research has demonstrated that the design set up of sexual jealousy reflects these ancestral challenges with men being more sensitive to their partner’s sexual infidelity whereas women to their partner’s emotional infidelity (Buss et al., 1992; Buss et al., 1999). A similar pattern was identified in a clinical population diagnosed with morbid jealousy (Easton et al., 2007).

In its functional, non-pathological state, sexual jealousy can be successful as a solution to the perennial problem of infidelity, but it comes at a cost which includes the risk of intimate partner violence. Tragically, Buss & Abrams (2016) report that 13% of all homicides are spousal murders and jealousy is a leading factor in the majority of these cases, while Dell (1984) concluded that 17% of all homicides in the UK were related to jealousy.

Morbid jealousy, also known as pathological jealousy or the Othello Syndrome, refers to a state involving irrational thoughts and emotions, together with associated unacceptable or extreme behaviour, in which the dominant theme is a preoccupation with a partner’s sexual unfaithfulness based on unfounded evidence (Cobb, 1979). This has been equated with a delusional disorder where the key psychopathology is the delusion of infidelity and can arise in association with a wide range of mental disorders (Kingham & Gordon, 2004). In clinical practice, morbid jealousy is more common in men and is considered to be a condition with a particularly high risk of violence (Kingham & Gordon, 2004).

Although, morbid jealousy is not a unitary state, it can be viewed as an aberrant and/or dysfunctional state of an adaptation aimed at reducing the risk of infidelity of a romantic/pair-bonded partner/spouse. Both the functional and dysfunctional states of sexual jealousy can, therefore, only be properly understood within the context of the distinctive human reproductive system. Without pair-bonding and male investment there can be little meaning for infidelity, cuckoldry or abandonment and thus the evolutionary pressures that give rise to sexual jealousy would be weak or absent. Hence, we would not expect to find functional or morbid jealousy in species with reproductive systems not based on pair-bonds.

Prediction 5: While the delusional aspect of this disorder is uniquely human and will not be detected in non-human animals (as is the case for all delusional disorders due to dependence on language) analogues for sexual jealousy (including the morbid variety) will be detected exclusively in species with pair-bonded mating systems that show consistently high levels of paternal investment and will not be detected in species with other types of mating systems.

4.6 Erotomania (De Clerambault’s Syndrome)

Erotomania may be seen as a pathological distortion of an evolved human mating strategy. It is an uncommon psychiatric condition where an individual (usually female) harbours the delusional belief that someone, usually a higher social status male, is in love with them. This occurs despite the lack of empirical or corroborated evidence of any interest by the target or even contact with them. Furthermore, the patient often provides elaborate delusional explanations to justify the target’s lack of response or interpret their rejecting behaviours as affirmation of love. Commonly, the onset of erotomania is sudden, and the course is chronic. Erotomania may present as a primary mental disorder, or as a symptom of another psychiatric illness. With secondary erotomania, the delusions appear to be caused by other mental disorders such as bipolar I disorder or schizophrenia. Symptoms may also be precipitated by alcoholism, substance abuse (including cannabis use) and other drug use (e.g. Kelly, 2005).

From an evolutionary perspective the phylogenetic roots of such traits or adaptations are of primary interest and although the precise precipitant or proximate cause of pathology is important clinically, the nature of the psychological adaptations is our main focus. It is proposed that erotomania arises as a dysfunctional by-product (or maladaptive variant) of adaptive cognitive and emotional systems primarily in females that are related to finding and attracting a high-status long-term mate. These systems become disordered and dysfunctional when they override reality and wishful thinking becomes the truth.

The evolutionary roots of erotomania likely lie in the following adaptations related to the human pair-bonding system:

1. Seeking high-status mates,

2. Detecting romantic interest,

3. Forming attachments,

4. Inferring others' thoughts and feelings.

In erotomania, these systems become overactive, mis-calibrated, and/or entirely disconnected from reality, leading to what appears clinically as persistent delusional beliefs. From an evolutionary standpoint, it is a maladaptive extreme of traits that usually support reproductive success and pair bonding.

A further evolutionary perspective is viewing erotomania through the lens of signal detection theory (Nesse, 2019). This is a general algorithm that biases decisions towards increasing the possibility of false positives and minimising false negatives. Hence, we suggest that in erotomania the ‘romantic interest detection system’ becomes mis-calibrated to ensure that romantic interest from a high-status male is never missed which then runs the risk of identifying romantic interest when none exists. However, why this results in delusional beliefs is currently poorly understood.

In a review of nearly 250 cases, Brüne (2001) concluded that erotomania aligns with an extreme expression of the evolved female long-term mating strategy, pursuing a high-status partner, to the point of delusion. This helps explain why erotomania is more prevalent in women, and why the “love object” or target is often an extremely high-status male (e.g. pop star, movie star, royalty, billionaires etc.). The sex ratio in erotomania consistently shows that women are affected about three times more often than men. In Brune’s sample 69% were female. Whether there are sex differences in the details of the psychopathology that reflect men’s and women’s evolved mating strategies remains an open question. It is possible that men with erotomania are underrepresented in clinical referrals due to added features such as stalking or violence that lead to criminal rather than psychiatric intervention.

Although high status males are attractive to females in species with non-pair-bonding reproductive systems, there are no analogues in those species of ‘romantic love’ (by the target individual towards the subject) which forms a core feature of the (delusional) belief system of patients with erotomania. Hence, viewing erotomania as a disorder of the human mating system helps identify the dysfunctional psychological adaptation thus offering a deeper understanding of this psychopathological phenomenon.

Prediction 6: Erotomania is a disordered and misplaced state of romantic love which is related to pair-bonding. However, given that erotomania is based primarily on a complex delusional system that is uniquely human (due to dependence on language), we predict that no analogues for this disorder will be found in any non-human primate species regardless of their mating system.

4.7 Eating disorders (anorexia and bulimianervosa; henceforth AN & BN)

The focus of this section is limited to AN and BN. These two disorders share the core features of: morbid fear of fatness, distorted body image, and a pattern of behavior aimed at weight reduction, including purging, restriction of food intake, or excessive exercise (American Psychiatric Association, 2013). AN is characterized by low body weight, whereas BN is associated with binge eating and normal body weight.

Previously, female preponderance of EDs had been reported to be around 10:1 (Gordon, 1990). However, more recent lifetime prevalence estimates have suggested a female preponderance of around 3:1 for both AN and BN. BN is more common than AN in both females and males, with lifetime prevalence rates for AN of 0.9% in women and 0.3% in men, whereas BN has a lifetime prevalence of 1.5% for women and 0.5% for men (Hudson et al., 2007). Nevertheless, those presenting for treatment are overwhelmingly female, with males accounting for only 1%–5% of patients with AN and 5%–10% of patients with BN (National Centre for Eating Disorders, 2020).

In addition to the marked female preponderance, AN & BN show several notable epidemiological features. These include: the claim that they have emerged in recent times (Rantala et al., 2019), the claim that they are vastly more prevalent in developed, Western countries (Katzman et al., 2004), and the evidence that there has been a significant increase in all EDs in recent decades (Liu et al, 2025; Russell, 2000). Furthermore, there are claims that this increase is spreading to developing countries in association with industrialization, urbanization, and Westernization (e.g. Sugermeyer, 2020). Hence, a major paradox of EDs is that their incidence and prevalence are much higher in sociocultural environments where high-energy nutrition abounds (Nesse, 2017).

Although there is a multiplicity of theoretical models (both evolutionary and non-evolutionary) purporting to explain these puzzling disorders, we will limit the discussion in this section to the Sexual Competition Hypothesis (SCH) as this has face validity, significant explanatory power and it highlights the connection between these disorders and human mating psychology (Abed, 1998; Abed, 2024).

The SCH states that the modern western and westernized environments present a range of conditions that increase female competition to levels unknown in the ancestral environment (or other traditional environments). In western environments, females compete intensely through the display of signs of youth and physical attractiveness and as a result, a minority of females succumb to dysfunctional/disordered states that are identified as eating disorders (AN & BN). As the primary object of female competition is mate attraction and/or retention, this can only be understood in the context of a pair-bonded mating system. Female chimpanzees and bonobos who have a promiscuous mating system have no need to compete for access to males as they invest next to nothing in offspring, and males willing to copulate are not in short supply. Thus, it is not possible to understand the nature of female competition in humans without understanding the dynamics of pair-bonding, paternal investment and human long-term mating strategies. The SCH states that the evolutionary roots of vulnerability to AN & BN lie in the human mating system where mismatch plays an important role within the modern western environment.

An interesting extension to the SCH is the proposal that AN and BN reflect disorders at different points on the life-history strategy spectrum where AN represents competition for long-term mates and a slow life history strategy and BN competition for short-term mates and a fast life history strategy (Del Giudice, 2018).

Hence, according to this formulation, AN & BN represent a dysregulation/disorder of female competition related to the distinctive human mating system that is mismatched to the modern western environment (Abed et al, 2012). This uniquely explains cardinal features of eating disorders which are: Why the female preponderance? Why have they emerged/increased in recent times? Why the particular geographical distribution? (Abed, 2024).

Prediction 7: Given that female competition in eating disorders relies intensively on the mismatch created by modern cultural norms and beliefs⁵ around attractiveness and body shape that are heavily language-based (as well as other factors contributing to mismatch such as the consumption of novel, ultra-processed foods) all of which are uniquely human traits, we do not expect to encounter non-human analogues of AN or BN whether in monogamous primates or any other species. In other words, we predict that there are no non-human analogues for AN and BN.

4.8 Borderline personality disorder (BPD)

We appreciate that the inclusion of BPD on this list of disorders may seem contentious and novel. Nevertheless, given BPD’s particular negative effect on romantic relationships (and thus on the stability of pair-bonds) which can be considered a central feature of the disorder, the hypothesis that BPD is primarily a disorder of the human mating system can be justified. The thesis that we present here is that the human, species-typical mating system based on pair-bonding and biparental investment is supported by a cluster of adaptations (traits, systems) that enable mate selection, attraction and retention. Any of these processes can fail and in BPD they can frequently break down in multiple ways.

DSM-5 specifies the criteria for diagnosing BPD as follows (American Psychiatric Association, 2013):

A pervasive pattern of instability of interpersonal relationships, self-image and affects, and marked impulsivity beginning by early adulthood and present in a variety of contexts, as indicated by five (or more) core symptoms:

1. Frantic efforts to avoid real or imagined abandonment

2. A pattern of unstable and intense interpersonal relationships characterised by alternating between extremes of idealisation and devaluation

3. Identity disturbance: markedly and persistently unstable self-image or sense of self

4. Impulsivity in at least two areas that are potentially self-damaging (e.g. spending, sex, substance abuse, reckless driving, binge eating)

5. Recurrent suicidal behaviour, gestures, or threats, or self-mutilating behaviour

6. Affective instability due to a marked reactivity of mood (e.g. intense episodic dysphoria, irritability or anxiety usually lasting a few hours and only rarely more than a few days)

7. Chronic feelings of emptiness

8. Inappropriate, intense anger or difficulty controlling anger (e.g. frequent displays of temper, constant anger, recurrent physical fights)

9. Transient, stress-related paranoid ideation or severe dissociative symptoms

It is notable that several of the criteria above can have a direct negative impact on the quality and sustainability of long-term romantic relationships or pair-bonds, notably criteria: 1, 2, 4, 5, 6, 8 & 9.

Studies examining the relationship between BPD and romantic relationship quality suggest it is associated with decreased relationship quality compared to other personality disorders (Bouchard & Sabourin, 2009). Also, maladaptive interpersonal functioning is most severe in close relationships (Hepp et al., 2018).

Interestingly, while BPD has a prevalence rate of around 2.7% in the general population, up to 50% of distressed couples seeking treatment have at least one member with BPD traits or the full syndrome (Bouchard & Sabourin, 2009). In addition, the NIMH-funded McLean study of adult development found that the proportion of BPD patients who were married or cohabiting increased when they were in remission from 15.4% to 38% whereas in non-remitted BPD patients the proportion of co-habiting or married remained unchanged at 15% (Zanarini et al., 2005).

Moreover, patients with BPD exhibit higher sexual impulsivity and infidelity than non-BPD individuals. This manifests in increased rates of sexually transmitted diseases, unwanted pregnancies and commercial sexual relationships (Frias et al., 2016).

Given this evidence, we propose that there are sound reasons for considering BPD a disorder of mating strategy where the human species-typical pair-bonded and biparental system frequently becomes unstable, dysfunctional and prone to failure. We accept BPD is heterogeneous in its proximate causes. However, its phenotypic characteristics and effects in the realm of pair-bonded relationships is our primary focus within this framework. Also, it is worth noting that BPD’s heterogeneity at the genetic, neurobiological and developmental levels is not unique as the same applies to a whole range of mental disorders which are defined by their phenotypic characteristics rather than by their proximate aetiology. Our argument is that the characteristics of BPD that make it particularly pathogenic in the human case is primarily because it tends to disrupt the species-typical mating system based on long-term pair bonds.

A stark illustration of this would be to imagine BPD occurring in chimpanzees or bonobos with their promiscuous mating system. In such a scenario, with an absence of pair-bonds, we suggest that a considerable proportion of the burden of the BPD psychopathology could not occur.

Prediction 8: Analogues of BPD will only be detected in socially monogamous primates e.g. gibbons especially those with a traumatic early life.

Prediction 9: Borderline traits in non-pair-bonded species such as chimpanzees will have no effect on mating success but can affect the capacity for mothering.

5. Limitations

Science walks forward on two feet, namely theory and experiment...

Sometimes it is one foot that is put forward first,

sometimes the other, but continuous progress is only made by the use of both…

Robert Andrews Millikan, Nobel Lecture, May 23, 1924

In proposing the current framework, we have followed the hypothetico-deductive method starting with looking at the available data on human mating and surveying the various psychopathological states. This was followed by abduction to the best available inference that assembles the observable data into the most parsimonious conclusion (see Griffin et al, 2025). In addition, we have proposed a number of predictions that are testable and falsifiable.

However, we recognise that the primary limitation of this framework is that it is based on evolutionary or ultimate causation and does not specify the relevant proximate mechanisms (neurobiological and other direct causes) of these states. And given that interventions in medicine (including psychiatry) are based on manipulating proximate mechanisms, this limits the direct clinical applications of the current framework. Evolutionary theories and models of disease and disorder are answers to the ‘why’ questions rather than the ‘how’ and it is the ‘how’ that can more easily and readily be translated into clinical practice. However, this limitation is not unique to this particular framework but shared with all evolutionary models of disease across medicine to varying degrees. It may be argued that evolutionary models based on mismatch can have the most potential for direct clinical application but even then ‘mismatch reduction’ (e.g. through alterations to environmental factors) may be neither easy nor practical.

In this paper, we propose a novel way of grouping a set of apparently disparate psychopathological states together based on a high-level characteristic at the human species level but it should be recognised that at the lower, mechanistic (proximate) level there may be little in common between these states. However, this should not detract from the benefits accrued from understanding the ultimate function of a disordered system.

We would argue that a deeper understanding of the common evolutionary roots of these psychopathological states, many of which are uniquely human, provides invaluable information to both clinicians and researchers. We suggest that the framework helps make sense of why humans are vulnerable to these disorders in ways that would be inconceivable when these disorders are considered separately or when grouped together based on surface similarity (see section 6).

6. What are the implications and benefits ofthis evolutionary reformulation?

While the approach of current mainstream psychiatry is symptomatically descriptive, the evolutionary approach aims to understand mental disorders within the context of evolved human psychological and neurobiological systems. This has implications for therapy where insight into root causes that go beyond the immediate precipitants and effects of illness can be of benefit to patients. Evolution’s epistemic value is that it helps answer the question: What is the functional (non-pathological) system involved in a given psychopathology? We suggest that this question yields a non-trivial answer which currently remains largely unexplored within mainstream psychiatry.

Additionally, evolutionary thinking introduces the phylogenetic dimension, placing human traits in a comparative context both with closely related as well as with more distant species. This yields insights that are otherwise unattainable. For example, the reason why human females compete intensely for mates while chimpanzee (and bonobo) females do not has important implications both for normative human psychological functioning as well as psychopathological states.

Hence, we suggest that one practical benefit from this proposed framework is the potential for the incorporation of a phylogenetic perspective in the psychological therapies and social interventions, as well as the appropriateness of drug therapies, for all the disorders listed in this article. Patient knowledge and understanding of the characteristics of the human mating system that distinguishes it from that of other species can help make sense of a range of normative as well as troubling experiences.

In addition, taking a comparative phylogenetic approach helps clarify any future thinking regarding animal models for any of the above disorders. As a starting point, species that do not form pair-bonds can be summarily ruled out. This is an important insight with practical implications and the lessons from this can be generalised more widely to other aspects of mental health. Thus, the reframing of the above disorders based on an evolutionary analysis can prepare the ground for future work that draws out possible implications for the prevention and treatment of these mental disorders and can form a template for a better understanding of a range of other disorders.