Related to phylogeny, we summarize theories of romantic love’s evolutionary history & show that romantic love probably evolved in concert with pair-bonds in our recent ancestors

Proximate and Ultimate Perspectives on Romantic Love. Adam Bode and Geoff Kushnick. Front. Psychol., April 12 2021. https://doi.org/10.3389/fpsyg.2021.573123

Abstract: Romantic love is a phenomenon of immense interest to the general public as well as to scholars in several disciplines. It is known to be present in almost all human societies and has been studied from a number of perspectives. In this integrative review, we bring together what is known about romantic love using Tinbergen’s “four questions” framework originating from evolutionary biology. Under the first question, related to mechanisms, we show that it is caused by social, psychological mate choice, genetic, neural, and endocrine mechanisms. The mechanisms regulating psychopathology, cognitive biases, and animal models provide further insights into the mechanisms that regulate romantic love. Under the second question, related to development, we show that romantic love exists across the human lifespan in both sexes. We summarize what is known about its development and the internal and external factors that influence it. We consider cross-cultural perspectives and raise the issue of evolutionary mismatch. Under the third question, related to function, we discuss the fitness-relevant benefits and costs of romantic love with reference to mate choice, courtship, sex, and pair-bonding. We outline three possible selective pressures and contend that romantic love is a suite of adaptions and by-products. Under the fourth question, related to phylogeny, we summarize theories of romantic love’s evolutionary history and show that romantic love probably evolved in concert with pair-bonds in our recent ancestors. We describe the mammalian antecedents to romantic love and the contribution of genes and culture to the expression of modern romantic love. We advance four potential scenarios for the evolution of romantic love. We conclude by summarizing what Tinbergen’s four questions tell us, highlighting outstanding questions as avenues of potential future research, and suggesting a novel ethologically informed working definition to accommodate the multi-faceted understanding of romantic love advanced in this review.

Discussion

Romantic love is a complex and multifaceted aspect of human biology and psychology. Our approach in this review has been to highlight how Tinbergen’s (1963) “four questions” can help us to synthesize the important strands related to the mechanisms, development, fitness-relevant functions, and evolutionary history of this phenomenon. Here, we synthesize what this review has presented in each level of explanation and suggest what this indicates about other levels of explanation. We then highlight some gaps in our knowledge that could be filled with future research and present a new ethologically informed working definition of romantic love.

What Do Tinbergen’s Four Questions Tell Us?

One of the benefits of using Tinbergen’s four questions as a framework to describe a complex trait such as romantic love is its ability for one level of explanation to provide insights into the other level of explanation (see Tinbergen, 1963; Bateson and Laland, 2013; Zietsch et al., 2020). In particular, an understanding of the proximate causes of romantic love has provided insights into the functions and phylogeny of romantic love although an understanding of the ultimate level of explanation provides some insights into the mechanisms of romantic love.

Multiple mechanistic systems involved in romantic love suggests it may serve multiple functions and may be a suite of adaptations and by-products rather than a single adaptation. We found that romantic love is associated with activity in a number of neural systems: reward and motivation, emotions, sexual desire and arousal, and social cognition. It is also associated with activity in higher-order cortical brain areas that are involved in attention, memory, mental associations, and self-representation. We also found that romantic love is associated with a number of endocrine systems: sex hormones, serotonin, dopamine, oxytocin, cortisol, and nerve growth factor. This is consistent with our position that romantic love serves mate choice, courtship, sex, and pair-bonding functions. Reward and motivation system activity may be particularly involved in the mate choice function of romantic love. Cortisol may be particularly indicative of the courtship function of romantic love, which overlaps with pair-bonding. Neural areas associated with sexual desire and arousal and the activity of sex hormones may play a particular role in the sex function. Finally, reward and motivation regions of the brain (rich with oxytocin receptors) and activity of the oxytocin system may play a particular role in the pair-bonding function of romantic love. Our understanding of the biological mechanisms that cause romantic love supports our description of romantic love’s functions.

Mechanistic similarities between romantic love and mother-infant bonding suggest that romantic love may have evolved by co-opting mother-infant bonding mechanisms. This articulates one hypothesis about the evolutionary history of romantic love that complements the predominate theory of independent emotion systems (Fisher, 1998, 2000; Fisher et al., 2002). This is supported by the psychological similarities between romantic love and early parental love.

Evidence of substantial activity of oxytocin receptor rich brain regions and the oxytocin endocrine system in romantic love lends weight to the position that romantic love only evolved after the neural circuitry associated with mate choice, specifically, regions of the mesolimbic reward pathway and dopamine rich areas, became populated by oxytocin receptors specifically receptive to stimuli from mating partners. That played a role in the evolution of enduring social attraction and pair-bond formation (Numan and Young, 2016). This supports our claim that romantic love probably evolved in conjunction with pair-bonds in humans. As a result, we are bolstered when we contend that romantic love emerged relatively recently in the history of humans.

The duration of romantic love also raises questions about the functions of romantic love. It has been said that the psychological features of romantic love can last from 18 months to 3 years in reciprocated romantic love. However, in our evolutionary history, romantic love would have usually occurred in the context of pregnancy and child birth. Mother-infant bonding becomes active around the time of childbirth. We are not aware of any research that has investigated whether romantic love can occur at the same time as mother-infant bonding or whether it must subside for mother-infant bonding to become active. Answering this question would elucidate if the functions of romantic love extinguish once reproduction has been successful. The existence of long-term romantic love also raises questions about the functions of romantic love. It has been posited that long-term romantic love is “part of a broad mammalian strategy for reproduction and long-term attachment” (Acevedo et al., 2020, p. 1). This may indicate that long-term romantic love serves similar functions to romantic love that lasts a shorter period of time.

Just as the multiple biological mechanisms involved in romantic love suggests a variety of functions, the functions of romantic love specified in our review suggests specific biological mechanisms are involved. As outlined above, specific functions may be associated with specific mechanisms and this should be an area of targeted research.

The possibility of romantic love evolving by co-opting mother-infant bonding mechanisms raises a number of possibilities in relation to the proximate causes of romantic love. It suggests that social activity associated with mother–infant bonding (e.g., filling of needs, specific cues) may be particularly important precursors to, or features of, romantic love. It suggests that many of the genes and polymorphisms involved in causing romantic love may have been present in mammals since the emergence of mother–infant bonding, making comparative animal research using mammals relevant. It also suggests that further research into shared neural activity between romantic love and mother–infant bonding is warranted.

We contend that romantic love probably emerged in conjunction with pair-bonds in humans or human ancestors. As such, further information about the similarities and differences between romantic love (pair-bonding) and companionate love (established pair-bonds) is needed. In particular, information about any role of the mesolimbic pathway (see Loth and Donaldson, 2021) or regions associated with sexual desire in companionate love would help to shed light on the evolutionary history of pair-bonding and pair-bonds. Specifically, this could shed light on if, as has been suggested (see Walum and Young, 2018), romantic love and pair-bonds are inextricably linked.

Areas of Future Research

One issue with research into the mechanisms of romantic love is that it has, with some exceptions (e.g., Fisher et al., 2010), utilized samples of people experiencing romantic love who are in a relationship with their loved one. Romantic love serves a mate choice and courtship function, and as a result, a large proportion of people experiencing romantic love are not in a relationship with their loved one (e.g., Bringle et al., 2013). A small number of studies have directly investigated unrequited love (e.g., Tennov, 1979; Baumeister et al., 1993; Hill et al., 1997; Aron et al., 1998; Bringle et al., 2013), but none of these investigated the mechanisms that cause romantic love. Studying such people might identify the specific contributions of particular mechanisms to particular functions. For example, the mechanisms associated with the pair-bonding function of romantic love may not be active in individuals who are engaging in courtship and the mechanisms involved in courtship may not be present in lovers who are already in a relationship with their loved one. Research would benefit from considering the mechanisms that underlie related psychopathologies and it would be useful to understand the relationship between mate preferences and romantic love.

Molecular genetics research, such as that undertaken by Acevedo et al. (2020), could further identify contributions of genes in people experiencing romantic love. Resting state fMRI provide an opportunity to investigate networks characteristic of psychopathology related to romantic love. Research should investigate the automatic/internal emotional regulatory network and the volitional/external regulatory network associated with mania/hypomania in people experiencing romantic love. Further research is required into the endocrinology of romantic love. In particular, further research is needed into the role of opioids, corticotropin-releasing factor, glutamate, acetylcholine, and vasopressin in romantic love. Efforts should be made to combine psychological and mechanisms research. For example, differences in neural or endocrine activity may be present in people experiencing romantic love who display elevated symptoms of depression compared to those who display reduced symptoms. As a result, neuroimaging and endocrinological studies could categorize people experiencing romantic love according to their levels of depression or type of hypomanic symptoms.

Given the large number of fMRI studies, interpreting the neuroimaging literature can be overwhelming. It has been nearly 10 years since the last meta-analysis of fMRI studies including romantic love. It is time for another one that focuses solely on romantic love. There is also a pressing need to attempt to replicate and extend endocrine studies and to specifically investigate the oxytocin system in people experiencing romantic love using validated measures of romantic love. As with many areas of psychological research (Henrich et al., 2010), and specifically in areas related to mating psychology (Apicella et al., 2019; Scelza et al., 2020), there is a pressing need to ensure that samples used in research are not exclusively Western, educated, industrialized, rich, and democratic.

Limited ontogeny research has elucidated the mechanisms causing romantic love across the lifespan. The literature that has (e.g., Luoto, 2019a), has focused on mate choice, rather than romantic love, per se. We know nothing about the neurobiology or endocrinology of romantic love in children or about the endocrinology of long-term romantic love. It would be useful to investigate how the functions of romantic love differ according to age of individuals or the duration of romantic love. Internal and external factors influence romantic love, although there has been surprisingly little research into this topic. It would be prudent to continue to develop a more detailed understanding of the factors that lead to romantic love (e.g., Riela et al., 2010, 2017). It would be useful to better understand the relationship between attachment styles and romantic love. Research should investigate if romantic love can occur at the same time as mother-infant bonding, or if they are mutually exclusive states.

Research into the functions of romantic love is sparse. There is a need for clear, evidence-informed definitions and descriptions of each of the functions of romantic love. It is likely that different mechanisms moderate different functions, and research should attempt to determine the contribution of specific genetic, neural, and endocrine activity to each individual function (see Zietsch et al., 2020). The advent of contraception and the adoption of family planning strategies means romantic love now serves more of a sex function than a pregnancy function in some environments. This is particularly the case early in a relationship. Pregnancy may become a feature as a relationship progresses and the fitness consequences of romantic love need to be investigated. Romantic love’s role as a suite of adaptations and by-products should be investigated. There is theoretical support for the notion that romantic love serves a health-promoting function (e.g., Esch and Stefano, 2005); however, there is a limited number of studies demonstrating a health-promoting effect of romantic love.

The relative infancy of genetic research, the lack of a clear fossil record, and the small number of species with which comparative analysis can be undertaken, means novel and creative means of investigating the phylogeny of romantic love must be undertaken. There is a need to pin-point the phylogenetic emergence of romantic love and the factors that caused it. To do this, more research into the genetics of romantic love must be conducted, and this should consider the phylogeny of specific genes and polymorphisms (e.g., Acevedo et al., 2020; see also Walum and Young, 2018). Efforts to assess the contribution of sexual selection to the evolution of romantic love are warranted. Studies of newly discovered fossils can help to identify shifts in sexual dimorphism that are indicative of pair-bonds. Further observational and experimental research into romantic love in hunter-gatherer tribes could tell us more about how romantic love functioned in our evolutionary history. Comparative research still has much to contribute. Research should explore the possibility that initial changes to the ancestral mammalian physiology that led directly to human romantic love arose in response to selection on both mating and non-mating-related behavior, such as pro-sociality (e.g., Barron and Hare, 2020; Luoto, 2020) or unique aspects of our species’ parenting repertoire. It might be fruitful to further investigate the relationship between romantic love and life history theory (e.g., Olderbak and Figueredo, 2009; Marzec and Łukasik, 2017). Finally, efforts should be made to elaborate and test the theory that romantic love emerged by co-opting mother–infant bonding mechanisms.

A New Working Definition of Romantic Love

The introduction to this review provided four definitions or descriptions of romantic love. For decades, most definitions (Hendrick and Hendrick, 1986; Sternberg, 1986; Hatfield and Rapson, 1993) of romantic love have informed research into the cognitive, emotional, and behavioral characteristics of romantic love. The past two decades, however, have seen an increasing focus on the biology of romantic love. Only recently has an evolution-informed definition been proposed (Fletcher et al., 2015). That working definition, however, does not incorporate much of the research that provides insight into the proximate and ultimate causes of romantic love.

We believe that the analytical approach taken in this review has identified sufficient information to justify the development of a new ethologically informed working definition of romantic love. The purpose would be to create an inclusive definition that is useful for researchers in varied disciplines investigating romantic love’s psychological characteristics, genetics, neurobiology, endocrinology, development, fitness-relevant functions, and evolutionary history. It may also be of use to psychologists and psychiatrists attempting to understand the experience and etiology of romantic love in their practice. It should be sufficiently precise and descriptive to both guide and link research. We provide, here, a working definition of romantic love:

Romantic love is a motivational state typically associated with a desire for long-term mating with a particular individual. It occurs across the lifespan and is associated with distinctive cognitive, emotional, behavioral, social, genetic, neural, and endocrine activity in both sexes. Throughout much of the life course, it serves mate choice, courtship, sex, and pair-bonding functions. It is a suite of adaptations and by-products that arose sometime during the recent evolutionary history of humans.

We situate the study of romantic love within the context of existing human mating literature. Our definition recognizes that romantic love is experienced across the lifetime of an individual, that research has shed light on the social, psychological, genetic, neural, and endocrine characteristics associated with it, and that it occurs in both sexes. Our definition also recognizes that romantic love serves a variety of functions and that these functions may vary across the lifespan. It does not exclude long-term or unrequited romantic love from the definition. Health is not identified as a function of romantic love in our definition despite being considered in our review. If more evidence comes to light, this definition can be amended to incorporate health.

Our definition has similarities and differences with the definition proposed by Fletcher et al. (2015). This is appropriate given both are informed by evolutionary approaches which differ somewhat. We do not specifically define romantic love as being a commitment device or reference passion, intimacy, and caregiving. In our review, we recognize that romantic love is a commitment device and serves to display commitment and signal fidelity as part of its courtship function. We believe that reference to romantic love’s behavioral activity and courtship and pair-bonding functions sufficiently encapsulate this concept. Sternberg’s (1997) definition of romantic love and Fletcher et al.’s (2015) definition include references to passion and intimacy. Caregiving (e.g., provision of psychological and emotional resources, sharing resources), while associated with pair-bonding, is not sufficiently definitive of romantic love using Tinbergen’s four questions as a framework to include in our definition.

We do not reference the universality of romantic love. While some experts assert its universality (e.g., Fletcher et al., 2015; Buss, 2019), we believe that the finding of Jankowiak and Fischer (1992) leaves enough uncertainty for it to be prudent to omit this aspect from our definition. Their research has found no evidence of romantic love in fifteen cultures (see Jankowiak and Paladino, 2008, for update to the original investigation) although this is probably the result of lack of data rather than evidence to the contrary. Once this matter is settled, which could be achieved by further investigating those societies where no evidence of romantic love was found, the definition can be amended. Fletcher et al. (2015) state that romantic love is associated with pair-bonds. We do the same by stating that pair-bonding is one of the functions of romantic love.

We also do not make specific reference to romantic love suppressing the search for mates. We recognize this as a cost in our review, but do not believe that this is so definitive of romantic love to include in our definition. Rather, we believe that our reference to “behavioral” activity and the “mate choice” function of romantic love in our definition sufficiently accommodates this feature. Our definition provides more detail than that provided by Fletcher et al. (2015) by including elements derived from substantial research into the mechanisms, ontogeny, functions, and phylogeny of romantic love. Like the Fletcher et al. (2015) definition, our definition recognizes that romantic love has distinct psychological characteristics and that we know about some of the proximate mechanisms that regulate it. However, as explained above, we do not include reference to the health-promoting effects of romantic love.

As more information about romantic love is gathered, we anticipate the definition to develop. However, we believe that this definition is an improvement upon previous definitions and adequately captures what is currently known about romantic love’s proximate and ultimate causes. It would be useful for researchers investigating romantic love from myriad perspectives. This definition should be critiqued and improved, and we welcome any such efforts from researchers and theorists across the spectrum of academic disciplines.

Politically misaligned bureaucrats incur in greater cost overruns, are generally less motivated, & are perceived as less loyal to their missions

Ideology and Performance in Public Organizations. Jorg L. Spenkuch, Edoardo Teso & Guo Xu. NBER Working Paper 28673, April 2021. DOI 10.3386/w28673

Abstract: We combine personnel records of the United States federal bureaucracy from 1997-2019 with administrative voter registration data to study how ideological alignment between politicians and bureaucrats affects the personnel policies and performance of public organizations. We present four results. (i) Consistent with the use of the spoils system to align ideology at the highest levels of government, we document significant partisan cycles and substantial turnover among political appointees. (ii) By contrast, we find virtually no political cycles in the civil service. The lower levels of the federal government resemble a "Weberian" bureaucracy that appears to be largely protected from political interference. (iii) Democrats make up the plurality of civil servants. Overrepresentation of Democrats increases with seniority, with the difference in career progression being largely explained by positive selection on observables. (iv) Political misalignment carries a sizeable performance penalty. Exploiting presidential transitions as a source of "within-bureaucrat" variation in the political alignment of procurement officers over time, we find that contracts overseen by a misaligned officer exhibit cost overruns that are, on average, 8% higher than the mean overrun. We provide evidence that is consistent with a general "morale effect," whereby misaligned bureaucrats are less motivated.

---

Panel A of Figure 7 shows that the overrepresentation of Democrats increases as we move up the hierarchy. Among employees in grades 1-12 of the GS, we find about 50% of Democrats (30% of Republicans and 20% of independents), which rises to approximately 56% at the top of the GS (grades 13-15), and to 63% among career SES.34 Intriguingly, this finding appears to be driven in large part by selection on observables.  First, Democrats have, on average, higher levels of human capital than Republicans. In Table 4, we report estimates from regressing indicators for educational attainment on a bureaucrat's political affliation. In order to measure education at entry, we restrict the sample to the first quarter in which the employee is observed.35 According to our results, Democrats are 6.6 p.p. more likely than Republicans to hold a college degree (column 1), and 8.3 p.p. more likely to have some form of post-graduate education (column 4). We continue to observe differences in human capital after controlling for bureau (columns 2 and 5) and pay-level fixed effects (columns 3 and 6)|although the gap between Democrats and Republicans does narrow. The pattern of coefficients in Table 4, therefore, suggests that higher human capital allows Democrats to be hired in bureaus and occupations that require more advanced skills as well as at higher steps of the hierarchy (see also Appendix Table A3).  Moreover, the fact that there do remain residual differences after accounting for bureau and pay grade implies that, even within comparable jobs, Democrat civil servants tend to be positively selected.

In addition to being positively selected at the time of hire, Democrats are more likely to be promoted after they enter the bureaucracy. In Table 5, we present estimates of partisan differences in promotions from grades 13{15 of the GS to the career SES (columns 1{3), as well as promotions from grades 1{12 of the GS to grades 13-15 (columns 4{6).36 Given that promotions are rare events at the quarterly level, all estimates in Table 5 are multiplied by 1,000. The results show that Democrats are more likely than Republicans to be promoted to higher levels of the hierarchy (columns 1 and 4), with a sizeable share of the gap being attributable to differences in educational attainment and the bureaus in which they serve (columns 2{3 and 5{6).37 The second factor that helps to explain greater overrepresentation of Democrats at higher levels of the bureaucracy is their lower propensity to exit. To illustrate this, Panel A of Figure 8 plots survival curves by partisan affiliation. While about 5% of civil servants of either party exit after the first quarter, the share of those who remain within the federal government as time progresses is significantly higher for Democrats. In Panel B of Figure 8, we repeat the exercise in regression form, controlling for bureau  quarter-of-entry fixed effects. After 10 years, Democrats are about 4.5% more likely than Republicans and independents to be still employed in the civil service.

5 Conclusion

A central question in the governance of any organization is how to align the objectives of leaders with those of their subordinates. In this paper, we turn to the U.S. federal bureaucracy to study the role of mission alignment in organizations.

To this end, we combine administrative data on the near universe of federal government workers with data on all registered voters in the U.S. The resulting dataset allows us to shed some of the first light on the ideological leanings of a large number of individual civil servants, and thereby peek into the black box of \bureaucratic politics." We establish three stylized facts. First, politicians do use the limited power they have over personnel policies in order to achieve greater ideological alignment between themselves and the upper echelon of the bureaucracy. The political cycles in our data are consistent with the use of the spoils system to better align the highest layers of the bureaucracy with the goals of the president. Second, we find a remarkable degree of political insulation among career civil servants. In contrast to political appointees, we see virtually no political cycles in the civil service. Our findings, therefore, suggest that, at its lower levels, the federal government resembles a \Weberian" bureaucracy, which is largely protected from political interference. Third, Democrats make up the plurality of civil servants. In addition, we show that Democratic civil servants are especially overrepresented in higher layers of the bureaucracy. Any observed difference in career progression, however, is in large part due to selection on observables. Democratic-leaning bureaucrats have on average higher levels of educational attainment, and they are less likely to exit the civil service, which results in a greater accumulation of experience. Both of these two facts are consistent with the idea that Democrats have a higher proclivity for public service.

The existence of an impartial and politically insulated career civil service is often seen as the hallmark of good governance and a \Weberian state." While the insulation of the career civil service prevents political interference, civil servants may have their own preferences and ideological leanings, which can conflict with those of the president. As a consequence, to implement an administration's agenda, politicians and department heads often need to work with bureaucrats whose personal values are not aligned with the present mission of the organization. To shed light on the costs of such misalignment, we focus on a subset of civil servants who work across all departments of the government and for whom we can measure performance: procurement ofcers. Linking procurement contracts to the matched personnel and voter registration data allows us to study the mission-alignment of procurement ofcers across nearly all departments of the federal bureaucracy. Strikingly, we find that political misalignment increases cost overruns by 8%. We provide evidence that suggests that a general \morale effect" is an important mechanism behind this finding, whereby bureaucrats who are ideologically misaligned with the organizational mission have lower motivation. As political turnover leads to sizable mission-misalignment between politicians and civil servants, our findings provide direct evidence on the costs of political insulation of the bureaucracy, which should be traded off against the benefits of avoiding political interference. As more and more organizations embrace a mission-driven focus, our findings may have implications beyond the public sector.

From 2007... Among the married, a higher discrepancy between men's and women's number of previous intercourse partners was related to lower levels of love, satisfaction, and commitment in the relationship

From 2007... Matching in Sexual Experience for Married, Cohabitating, and Dating Couples. Luis T. Garcia & Charlotte Markey. The Journal of Sex Research, Volume 44, 2007 - Issue 3, Pages 250-255. Dec 5 2007. https://doi.org/10.1080/00224490701443817

Abstract: This study examined heterosexual romantic partners' number of intercourse partners prior to the initiation of their relationship to determine if a significant positive correlation (matching) occurred between partners, and if this matching was associated with their level of love and satisfaction with and commitment to the relationship. One hundred and six couples who were dating, cohabitating, or married participated in this study. Results indicated that, with the exception of cohabitating couples, romantic partners showed a significant level of matching in the prior number of intercourse partners. Further, among the married couples, a higher discrepancy between men's and women's number of previous intercourse partners was related to lower levels of love, satisfaction, and commitment in the relationship.

Dynamic network perspective represents major departure from localist models; instead of cognitive functions mapping to discrete neural regions/connects, mental operations are suggested to be supported by unique conjunctions of distributed brain regions

Neuroimaging evidence for a network sampling theory of individual differences in human intelligence test performance. Eyal Soreq, Ines R. Violante, Richard E. Daws & Adam Hampshire . Nature Communications volume 12, Article number: 2072. Apr 6 2021. https://www.nature.com/articles/s41467-021-22199-9

Abstract: Despite a century of research, it remains unclear whether human intelligence should be studied as one dominant, several major, or many distinct abilities, and how such abilities relate to the functional organisation of the brain. Here, we combine psychometric and machine learning methods to examine in a data-driven manner how factor structure and individual variability in cognitive-task performance relate to dynamic-network connectomics. We report that 12 sub-tasks from an established intelligence test can be accurately multi-way classified (74%, chance 8.3%) based on the network states that they evoke. The proximities of the tasks in behavioural-psychometric space correlate with the similarities of their network states. Furthermore, the network states were more accurately classified for higher relative to lower performing individuals. These results suggest that the human brain uses a high-dimensional network-sampling mechanism to flexibly code for diverse cognitive tasks. Population variability in intelligence test performance relates to the fidelity of expression of these task-optimised network states.

Introduction

The question of whether human intelligence is dominated by a single general ability, ‘g’1, or by a mixture of psychological processes2,3,4,5,6, has been the focus of debate for over a century. While performance across cognitive tests does tend to positively correlate, population-level studies of intelligence have clearly demonstrated that tasks which involve similar mental operations form distinct clusters within a positive correlation manifold. These task clusters exhibit distinct relationships with various sociodemographic factors that are not observable when using aggregate measures of intelligence, such as ‘g’2,7.

Recent advances in network science offer the potential to resolve these contrasting views. It has been proposed that transient coalitions of brain regions form to meet the computational needs of the current task8,9,10. These dynamic functional networks are thought to be heavily overlapping, such that any given brain region can express flexible relationships with many networks, depending on the cognitive context8,9,11,12,13. This dynamic network perspective represents a major departure from localist models of brain functional organisation. Instead of cognitive functions mapping to discrete neural regions or specific connections, mental operations are suggested to be supported by unique conjunctions of distributed brain regions, en masse. The set of possible conjunctions can be considered as the repertoire of dynamic network states and the expression of these states may differ across individuals and relate to cognitive performance.

This conceptual shift motivates us to propose a network sampling theory of intelligence, which is conceptually framed by Thomson’s classic sampling theory14. Thomson originally proposed that ‘every mental test randomly taps a number of ‘bonds’ from a shared pool of neural resources, and the correlation between any two tests is the direct function of the extent of overlap between the bonds, or processes, sampled by different tests’. Extending this hypothesis, network sampling theory views the set of connections in the brain that constitute a task-evoked dynamic network state to be equivalent to Thomson’s ‘bonds’; therefore, the set of available brain regions is equivalent to the ‘shared pool of neural resources’. The distinctive clusters within the positive manifold reflect the tendency of operationally similar tasks to rely on similar dynamic networks2,15,16,17. From this perspective, the general intelligence factor ‘g’ is proposed to be a composite measure of the brain’s capacity to switch away from the steady state, as measured in resting-state analyses, in order to adopt information processing configurations that are optimal for each specific task. When recast in this framework, classic models of unitary and multiple-factorial intelligence1,14 are reconciled as different levels of summary description of the same high-dimensional dynamic network mechanism. The notion of domain-general systems such as ‘task active’ or ‘multiple-demand’ cortex is also reconciled within this framework. Specifically, each brain region can be characterised by the diversity of network states they are active members of. Brain regions that classic mapping studies define as ‘domain-general’ place at one extreme of the membership continuum, whereas areas ascribed specific functions, e.g., sensory or motor, place at the other extreme. The aim of this study was to test key predictions of network sampling theory using 12 cognitive tasks and machine learning techniques applied to functional MRI (fMRI) and psychometric data. First, we test the hypothesis that cognitive tasks evoke distinct configurations of activity and connectivity in the brain. We predicted that these configurations would be sufficient to reliably classify individual tasks, and that this would be the case even when focusing on brain regions at the domain-general extreme of the network membership continuum. We then tested Thomson’s hypothesis that similarity between cognitive tasks maps to the ‘overlap’ of the neural resources being tapped. Subsequently, it was predicted that the ability to classify pairs of tasks would negatively correlate with their behavioural-psychometric similarity, with tasks that are less similar being classified more reliably. Next, we hypothesised that individual functional dynamic repertoires would positively correlate with task performance, with the top performers expressing task configurations that would be more reliably classified. We also tested the prediction that classification success rates should have a basis in a combination of the distinct visual (VS), motor and cognitive sub-processes of the tasks. Finally, we hypothesised that task performance would be associated with optimal perturbation of the network architecture from the steady state, and that certain features within the network would have more general and more prominent roles in intelligence test performance.

Discussion

The results presented here are highly compatible with a network science interpretation of Thomson’s sampling theory14. Indeed, as has been noted by others, the relationship between the classic notion of a flexible pool of bonds and the analysis of the brain’s dynamic networks as applied a century later is striking17. Thomson proposed that mental tests tap bonds from a shared pool of neural resources, which is confirmed by our observation that different cognitive tasks tend to recruit unique but heavily overlapping networks of brain regions. Furthermore, when testing Thomson’s proposal that the correlation between any two tasks is a function of the extent of overlap between their bonds, we confirmed that the similarities of tasks in multi-factor behavioural psychometric space correlated strongly with the similarities in the dynamic network states that they evoked. These findings corroborate the key tenets of network sampling theory, further predictions of which were tested utilising a combination of machine learning techniques applied to the fMRI and psychometric data.

From a network science perspective, our results showing that the tasks were 12-way classifiable with high accuracy based on their dynamic network states is highly relevant. Indeed, the 74% accuracy achieved by the CRTX stack model was surprising, given that chance was 8.3% and we used just 1 min, comprising 30 images, of task performance data per classified sample. Although activity and connectivity provided complementary information when combined in the stack models, classification accuracy was consistently higher for connectivity when the measures were analysed independently. These results strongly support the hypothesis that the human brain is able to support diverse cognitive tasks because it can rapidly reconfigure its connectivity state in a manner that is optimal for processing their unique computational demands8,9,12,17. A key finding was that the task-evoked dynamic network states were consistent across individuals; i.e., our trained 12-way classification models operated with high accuracy when applied in a robust CV pipeline to data from individuals to whom they were completely naive. This was with an out of the box classifier with no CV optimisation, which is important, because it means that the features that drove accurate classification must reflect on a fundamental level how networks in the human brain are prewired to flexibly support diverse tasks.

At a finer grain, these task-optimised network states are most accurately described as a perturbation away from the RSN architecture of the brain12,29. More specifically, it was not simply the case that the relative levels of activity or connectivity within each RSN change, i.e., reflecting different mixtures dependent on task demands; instead, the features that were most specific to a given task-evoked state were predominantly the inter-RSN connections. Put another way, task-evoked states are not a simple blending of RSNs, but a dissolution of the RSN structure. This extends the findings of another recent study, where we used a similar analysis pipeline to examine how different aspects of working memory affected brain activity and connectivity12. Mirroring the current findings, we found that behaviourally distinct aspects of working memory mapped to distinct but densely overlapping patterns of activity and connectivity within the brain. Taken together, these results do not accord well with the hypothesis that the human brain is organised into discrete static networks. Instead, it would appear that the dynamic network coding mechanism is very high-dimensional, relating to the greater number of possible combinations of nodes8,9. There are dependencies whereby some nodes operate together more often than others, but these canonical network states, which are consistently evident in data-driven analyses of the resting state brain, are statistical rather than absolute. Our more holistic interpretation of the relationship between network states and cognitive processes is further supported by the analysis of the classifiability of task clusters when grouped according to their behavioural dimensions. Specifically, when grouped by psychometric, motor or VS characteristics, the clusters were more classifiable than random task groupings in all cases. It was notable though that psychometric and motor characteristics provided a stronger basis for classification. This is interesting, because it pertains to how the most prominent factors of human intelligence differ operationally. For example, it accords well with process overlap theory17, which proposes that general intelligence relates most closely to processes that are common across many different cognitive tasks.

More generally, the fact that inter-individual differences in the classifiability of the tasks predicted variability in a general measure of behavioural task performance provides further evidence that cognitive faculties relate to the way in which the brain expresses these task-optimal network states. Previous research into the neural basis of human intelligence has typically emphasised the role of flexible FP brain regions2,30,31,32. In this context, our focused analysis of the INTR ROI set warrants further discussion. Brain regions within the INTR ROI set belong to the classical MD cortical volume, which has been closely associated with general intelligence. MD includes the FP brain regions that have the broadest involvement in cognitively demanding tasks19,20,30; this includes executive functions, which enable us to perform complex mental operations33,34 and that have been proposed to relate closely to the ‘g’ factor17. From a graph theoretic perspective, MD ROIs have been reported to have amongst the broadest membership of dynamic networks of any brain regions35 and it has been shown that inter-individual variability in the flexibility of MD nodes, as measured by the degree of their involvement in different functional networks, correlates positively with individuals’ abilities to perform specific tasks, e.g., motor skill learning36 and working memory37. Collectively, these findings highlight a strong relationship between the flexibility of nodes within MD cortex and cognitive ability.

Here, we reconfirmed that MD ROIs were amongst the most consistently active across the 12 tasks. However, we also demonstrated that these ROIs were highly heterogeneous with respect to their activation profiles across those tasks. Furthermore, in many cases they were significantly active for most but not all tasks. This variability in the activation profiles even amongst the most commonly recruited areas of the brain aligns with the idea that MD cortex flexibly codes for diverse tasks in a high-dimensional manner. More critically, the internal activity and connectivity of the INTR ROI set was not strongly predictive of behavioural task performance. Nor did it provide the most accurate basis for classification overall, or correspondence to psychometric structure. Extending to the MDDM set provides an improvement, but it was inclusion of the whole cortex ROI set that provided the best predictor of task and behavioural performance. Furthermore, connections between the core set of INTR regions and the rest of the brain featured prominently in all of the above cases. This finding accords with bonds theory, insofar as that theory pertains to the wide variety of bonds that contribute to diverse behavioural abilities. It also accords particularly well with the core tenet of network science that cognitive processes are emergent properties of interactions that occur across large-scale distributed networks in the brain10,12.

An intriguing aside pertains to the phenomena of ‘factor differentiation’. It was originally noted by Spearman38 that ‘g’ explains a greater proportion of variance individuals who perform lower on intelligence tests. This finding has been robustly replicated over the subsequent century5. Our results provide a simple explanation for factor differentiation. When individuals of higher intelligence perform different cognitive tasks, the dynamic network states that they evoke are more specific. Therefore, there is less overlap in the neural resources that they recruit to perform the tasks. Given the relationship observed here between network similarity and behavioural-psychometric distance, this would be expected to reduce bivariate correlations in task performances and produce a corresponding reduction in the proportion of variance explained by ‘g’.

The boosted ensemble of regression trees provided a simple way to extend the individual differences analysis in order to capture not just mixtures but also interactions between network features when predicting behavioural performance. We observed that increased connectivity between DA and VS systems strongly associated with better performance, whilst increased connectivity within the DMN combined with decreased connectivity between either DA to VS or DM to FP associated with lower performance. This accords well with previous studies that have shown that these networks update their connectivity patterns according to the task context35,39,40,41,42,43. However, it was particularly notable that inter-RSN connections again played the most prominent role insofar as they formed the roots of all of the trees, meaning they had the broadest relevance across individuals. This further accords with the view that task-evoked network states are best described as a perturbation from the RSN architecture12,29.

In summary, we validated multiple key predictions of network sampling theory. This theory can potentially explain key findings from behavioural psychometrics, experimental psychology and functional neuroimaging research within the same overarching network-neuroscience framework, and bridges the classic divide between unitary and multi-factorial models of intelligence. Given that our machine learning analysis pipeline aligns naturally with multivariate network coding whereas more commonly applied univariate methods do not, we believe that the analysis of multivariate network states as applied here has untapped potential in clinical research; e.g., providing functional markers for quantifying the impact of pathologies and interventions on the brain’s capacity to flexibly express task optimised network states11,29.

Is the Psychopathic Brain an Artifact of Coding Bias? A Systematic Review

Is the Psychopathic Brain an Artifact of Coding Bias? A Systematic Review. Jarkko Jalava et al. Front. Psychol., April 12 2021. https://doi.org/10.3389/fpsyg.2021.654336

Abstract: Questionable research practices are a well-recognized problem in psychology. Coding bias, or the tendency of review studies to disproportionately cite positive findings from original research, has received comparatively little attention. Coding bias is more likely to occur when original research, such as neuroimaging, includes large numbers of effects, and is most concerning in applied contexts. We evaluated coding bias in reviews of structural magnetic resonance imaging (sMRI) studies of PCL-R psychopathy. We used PRISMA guidelines to locate all relevant original sMRI studies and reviews. The proportion of null-findings cited in reviews was significantly lower than those reported in original research, indicating coding bias. Coding bias was not affected by publication date or review design. Reviews recommending forensic applications—such as treatment amenability or reduced criminal responsibility—were no more accurate than purely theoretical reviews. Coding bias may have contributed to a perception that structural brain abnormalities in psychopaths are more consistent than they actually are, and by extension that sMRI findings are suitable for forensic application. We discuss possible sources for the pervasive coding bias we observed, and we provide recommendations to counteract this bias in review studies. Until coding bias is addressed, we argue that this literature should not inform conclusions about psychopaths' neurobiology, especially in forensic contexts.

Discussion

Neurobiological reviews of PCL-R and PCL:SV psychopathy significantly under-report null-findings in sMRI research, indicating widespread coding bias. The majority (64.18%) of original sMRI findings were nulls, whereas nulls made up a small minority (8.99%) of effects in review literature. Reviewers, in other words, preferentially reported data supporting neurobiological models of psychopathy. We found no evidence that the reporting imbalance was due to factors other than bias: systematic, narrative, and targeted reviews all reported disproportionately few nulls (though meta-analyses reported too few effects to evaluate), the pattern was stable across time, and not driven by exploratory research or outliers. Notably, reviews calling for forensic application of the data, such as treatment, criminal responsibility, punishment, and crime prediction, were no more accurate than purely theoretical reviews. Applied reviews were, however, more likely than theoretical reviews to conclude that the data supported neurobiological bases of psychopathy. These findings are surprising, as applied reviews in other fields—such as those examining drug safety and efficacy—typically face the highest burden of proof and are thus most likely to emphasize limitations in the data [see e.g., Köhler et al. (2015)].

Our study is the first to systematically examine coding bias in cognitive neuroscience. Although our findings are limited to structural imaging in psychopathy, they suggest that coding bias should be considered alongside more widely recognized Questionable Research Practices (QRPs) such as p-hacking, reporting bias, publication bias, citation bias, and the file drawer problem. QRPs in original research filter out null-findings at early stages of the research and publication process, while coding and citation bias further distort the state of scientific knowledge by eliminating null findings from reviews. In addition to coding bias, we found evidence of reporting bias during our review of sMRI studies. Null-findings in the original literature were rarely reported in the study abstracts and were frequently not reported fully in results sections. Nulls often appeared only in data or supplemental tables, and in some cases they had to be inferred by examining ROIs mentioned in the introduction but not in the results section. This illustrates how QRPs are not mutually exclusive, and the presence of one QRP may also signal the presence of another [see e.g., Agnoli et al. (2017)].

The coding bias we observed may have a number of explanations. First, reviewers may have been subject to confirmation bias. Confirmation bias refers to the tendency to weigh evidence that confirms a belief more heavily than evidence that does not (Nickerson, 1998). Reviewers in our study may have assumed neurobiological abnormalities in psychopaths—perhaps from previous reviews—and looked more carefully for data to confirm that assumption. Confirmation bias has been cited as a possible explanation for under-reporting of null-findings in original research (Forstmeier et al., 2017). Our findings suggest that it may play a role in review literature, where null-findings would be especially difficult to square with theories presuming group differences [see e.g., Sterling et al. (1995) and Ferguson and Heene (2012)], and reporting bias would make it very hard to locate disconfirming (null) findings. Second, reviewers may have been following convention. The earliest review studies did not generally include null-findings, and later reviews may have interpreted this as a precedent to follow. Third, explicit and tacit publication preferences may increase coding bias. Research tracking original studies from grant proposal to publication show that most null-findings are not even written up for publication, and that journals—particularly top-tier journals—show a marked preference for strong positive findings (Franco et al., 2014; Ioannidis et al., 2014). Similarly, review authors may have declined to submit reviews with inconclusive findings. Given the extent of publication bias, it is also possible that journal editors may have been more likely to reject inconclusive reviews in favor of those summarizing consistent, positive findings.

Coding bias observed in our study has a number of potential effects. Aside from distorting the true state of knowledge about structural brain abnormalities in psychopaths, it may also have led at least some researchers and courts to believe that the abnormalities are consistent enough for forensic application. This may have encouraged practitioners to de-emphasize or overlook more reliable, behavioral indicators of criminal responsibility, future dangerousness and treatment amenability in favor of less reliable predictors, such as brain structure. Neuroprediction of crime has a number of empirical shortcomings, such as unknown measurement error and inadequate outcome variables (Poldrack et al., 2018). Using MRI data to predict crime can thus introduce substantial error into an already imperfect process (e.g., Douglas et al., 2017). Neurobiologically-informed assessments and treatments are even less likely to be effective if the population's neurobiology is fundamentally misunderstood. Given the extent of coding bias in the psychopathy literature, such interventions may in fact be harmful.

More broadly, coding bias may have contributed to reverse inference [see Scarpazza et al. (2018)] whereby reports of brain abnormalities are taken as proof that psychopathy is a legitimate diagnostic category [for an argument such as this, see e.g., Kiehl and Hoffman (2011)].5 Similarly, some researchers have suggested that psychopathy diagnoses could be enhanced by neuroimaging evidence (e.g., Hulbert and Adeli, 2015). Arguments of this sort can detract from problems in other aspects of the PCL-R, particularly in its psychometric properties. Recently, these critiques have intensified, with authors raising concerns about the reliability of the PCL-R, its utility in forensic contexts (DeMatteo et al., 2020), its factor structure, and its predictive validity (Boduszek and Debowska, 2016). Using neurobiology to validate psychopathy as a diagnostic category is doubly problematic: not only are presumed brain abnormalities in psychopathy broad and non-specific [for problems in reverse inference, see Poldrack (2011) and Scarpazza et al. (2018)], but as we have shown here, their consistency appears to be largely misunderstood as well.

In light of our findings, we recommend the following: First, published review literature on sMRI studies of PCL-R and PCL:SV psychopathy should be approached with caution, especially when the literature is used to influence forensic decisions. Second, we recommend that guidelines for conducting review literature be revised to include explicit guidance for avoiding coding bias. Although the problem of un- and under-reported null-findings is recognized [e.g., Pocock et al., 1987; Hutton and Williamson, 2000; guidelines for accurate reporting in review literature also exist; see Petticrew and Roberts (2008), American Psychological Association (2008), and Moher et al. (2015)], the role of coding bias, by and large, is not. Third, we recommend that review literature pay careful attention to the a priori likelihood of null-findings in their data. In our example, both the PCL-R (DeMatteo et al., 2020) and neuroimaging methods (Nugent et al., 2013) have relatively low reliability. The likelihood that sMRI research on psychopathy should yield more than 91% positive findings is therefore not realistic [for more extended discussions relating to fMRI, see Vul et al. (2009) and Vul and Pashler (2017)]. Fourth, we recommend that the production of new data should be complemented by closer examination of data already published. Among the 45 reviews we evaluated, we found a single study (Plodowski et al., 2009) that comprehensively reported all nulls in the original literature. Unfortunately, it was also among the least cited reviews, suggesting that accuracy and scientific impact do not necessarily go together. Finally, we recommend that reviewers pay close attention to potential biases—such as publication and reporting bias, p-hacking, and the file drawer problem—in the original literature, and take measures to compensate for them. Currently, it appears that reviews largely magnify them instead.

Limitations

Our study has a number of important limitations. First, in order to focus on forensically relevant studies, we limited our analysis to PCL-R and PCL:SV psychopathy. We also excluded studies that reported on PCL-R Factor scores only (e.g., Bertsch et al., 2013), that did not use case-control or correlational method (Sato et al., 2011; Kolla et al., 2014), and that included youth samples. It is possible that the excluded studies were reported more accurately in review literature than those we included. Second, we excluded original and review studies not published in English. This may have introduced a selection bias of our own, as it is possible that non-English publications use different standards of reporting and reviewing than those published in English. Third, our findings may have underestimated the extent of the bias. For example, one whole-brain analysis reviewed here (Contreras-Rodríguez et al., 2015) only reported positive findings, which means that the remaining brain regions were unreported nulls. Had these unreported null-findings been included in our analysis, the true percentage of nulls in the original studies would have been greater than 64.18%. Further, we did not account for possible publication bias. Since null-findings are presumed to be less likely than null-rejections to be published, the percentage of true nulls in the field is essentially unknown, though it may be significantly higher than we estimated (review literature examined here did not report any unpublished null-findings). Finally, we excluded fMRI and other imaging methods entirely. Future research could evaluate whether coding bias is present in reviews of this literature as well.

Those with low levels of conscientiousness, life satisfaction, & self-esteem, as well as high levels of neuroticism, used more drugs on average; in contrast, found little evidence for personality change following substance use

How Does Substance Use Affect Personality Development? Disentangling Between- and Within-Person Effects. Lara Kroencke et al. Social Psychological and Personality Science, July 7, 2020. https://doi.org/10.1177/1948550620921702

Abstract: Little is known about the effects of substance use on changes in broad personality traits. This 10-year longitudinal study sought to fill this void using a large, representative sample of the Dutch population (N = 10,872), which provided annual assessments of drug use (tobacco, alcohol, sedatives, soft drugs, ecstasy, hallucinogens, and hard drugs), Big Five personality traits, life satisfaction, and self-esteem. Using multilevel models, we examined the longitudinal associations between drug use and personality both between and within persons. Results indicated that individuals with low levels of conscientiousness, life satisfaction, and self-esteem, as well as high levels of neuroticism, used more drugs on average (between-person effects). In contrast, we found little evidence for personality change following substance use (within-person effects). We discuss these findings in the context of previous empirical and theoretical work and highlight opportunities for future research.

Keywords: substance use, drug use, personality development, life satisfaction, self-esteem

Discussion

This research examined the 10-year longitudinal associations between broad personality traits, life satisfaction, and self-esteem and use of different legal and illegal substances in a representative sample of the Dutch population. The purpose was to disentangle stable between-person effects from within-person associations to advance our understanding of the sources that may drive personality change. In what follows, we discuss our findings with respect to the previous literature and highlight their implications.

Consistent with our preregistration and past research, we found evidence for moderate between-person associations between drug use and personality traits. Specifically, individuals who were high in neuroticism and low in conscientiousness were more likely to consume drugs. These findings were mirrored by associations with life satisfaction and self-esteem (participants lower in life satisfaction and self-esteem were more likely to report substance use). As expected from our power analysis, even small to moderate effects (B > .30) were typically significant, except for infrequently consumed substances.

The fact that conscientiousness was related to use of nearly all substances is consistent with its association with a wide range of health behaviors (Bogg & Roberts, 2004). The relationships between substance use and neuroticism may indicate attempts of self-medication among neurotic individuals in an effort to decrease negative affective states (e.g., Khantzian, 1997). Interestingly, these between-person effects were more pronounced for less frequently consumed substances.

Regarding personality change, our study is among the first to fully disentangle between- from within-person effects and hence represents a more conservative test for the hypotheses at hand. Contrary to previous studies, we found few within-person effects of drug use on subsequent personality change. Even when significant, these effects were considerably smaller than the between-person effects, and none of the effects were predicted based on the existing literature. The within-person effects for the more malleable variables life satisfaction and self-esteem were also small and rarely significant, highlighting the robustness of the results. Below, we will discuss several possible reasons for the lack of predicted within-person effects.

First, our study was limited by selective attrition and somewhat lower power for rarely consumed drugs. Importantly, our power for relatively frequently consumed drugs was adequate even for small within-person effects. As such, the null findings for those effects are unlikely to represent Type II errors.

Second, we investigated whether a drug was consumed during the last month, but we did not measure substance use over longer periods of time, neither did our measures account for intensity and context of usage. We tried to control for these limitations (e.g., by investigating the effects of repeated use), but future studies should replicate our results using alternative measures of drug use.

Third, the intervals between personality and drug use assessments were relatively long, preventing us from examining transitory effects (less than 200 days). Our analyses were also restricted by the limited number of assessments per person. Future studies should include more measurement points and examine both the bivariate trajectories of substance use and personality and the effects of certain substance use life events (e.g., first onset of use) on personality trajectories.

Our findings have important theoretical implications. First, drug use has been proposed as a candidate mechanism for changes in personality that may be mediated via biological pathways (Costa et al., 2019) as well as behavioral or social mechanisms. Although theoretically plausible, we found little evidence for such effects. Second, we observed large variability in the associations between substance use and personality (i.e., random effects), indicating that, despite the lack of strong main effects, there are significant individual differences in within-person associations between substance use and personality. In other words, substance use might have negative effects for some people but no effects or even positive effects for others. Future studies should examine which moderator variables explain these different trajectories.

To our knowledge, this is the first large-scale study examining the impact of a wide range of drugs on the Big Five personality traits, life satisfaction, and self-esteem. We analyzed data from more than 10,000 individuals that were collected over a period of more than 10 years with an average of three assessments for each participant, using highly reliable personality measures. In addition, we used statistical models that effectively distinguished between- and within-person effects. Overall, our study provides strong evidence for between-person relationships between substance use and personality differences but little evidence for within-person changes in personality following substance use.