Thursday, July 7, 2022

Big meta-analysis: Religious people self-report higher well-being, specially in more religious societies

A many-analysts approach to the relation between religiosity and well-being. Suzanne Hoogeveen et al. Religion, Brain & Behavior, Jul 6 2022.

Abstract: The relation between religiosity and well-being is one of the most researched topics in the psychology of religion, yet the directionality and robustness of the effect remains debated. Here, we adopted a many-analysts approach to assess the robustness of this relation based on a new cross-cultural dataset (N=10,535 participants from 24 countries). We recruited 120 analysis teams to investigate (1) whether religious people self-report higher well-being, and (2) whether the relation between religiosity and self-reported well-being depends on perceived cultural norms of religion (i.e., whether it is considered normal and desirable to be religious in a given country). In a two-stage procedure, the teams first created an analysis plan and then executed their planned analysis on the data. For the first research question, all but 3 teams reported positive effect sizes with credible/confidence intervals excluding zero (median reported β=0.120). For the second research question, this was the case for 65% of the teams (median reported β=0.039). While most teams applied (multilevel) linear regression models, there was considerable variability in the choice of items used to construct the independent variables, the dependent variable, and the included covariates.

Keywords: Healthmany analystsopen sciencereligion

5. Summary

In the current project, 120 analysis teams were given a large cross-cultural dataset (, 24 countries) in order to investigate two research questions: (1) “Do religious people self-report higher well-being?” and (2) “Does the relation between religiosity and self-reported well-being depend on perceived cultural norms of religion?.” In a two-stage procedure, the teams first proposed an analysis and then executed their planned analysis on the data.

Perhaps surprisingly in light of previous many-analysts projects, results were fairly consistent across teams. For research question 1 on the relation between religiosity and self-reported well-being, all but three teams reported a positive effect size and confidence/credible intervals that exclude zero. For research question 2, the results were somewhat more variable: 95% of the teams reported a positive effect size for the moderating influence of cultural norms of religion on the association between religiosity and self-reported well-being, with 65% of the confidence/credible intervals excluding zero. While most teams used (multilevel) linear regression, there was considerable variability in the choice of items used to construct the independent variable, the dependent variable, and the included covariates.

A further discussion of these results including limitations and broader implications, as well as a reflection on the many-analysts approach is covered in the closing article (Hoogeveen et al.2022). There, we also address the commentaries written by some of the analysis teams.

Many-Analysts Religion Project: Reflection and conclusion

Many-analysts religion project: reflection and conclusion. Suzanne Hoogeveen et al. Religion, Brain & Behavior, Jul 5 2022.

In the main article on the Many-Analysts Religion Project (MARP) the results of the 120 analysis teams were summarized by taking each team's reported effect size and subjective assessment of the relation between religiosity and well-being, and the moderating role of cultural norms on this relation (Hoogeveen et al., 2022). The many-analysts approach allowed us to appraise the uncertainty of the outcomes, which has been identified as one of the pillars of good statistical practice (Wagenmakers et al., 2021). A downside of this approach, however, is that a fine-grained consideration of the details and nuances of the results becomes difficult. Summaries of the individual approaches are documented in the teams' OSF project folders, but time and space did not permit the inclusion of details on each of the individual analysis pipelines in the main article.

However, we believe the scope of the project and the effort of the analysis teams justify highlighting some more in-depth observations. Here, we aim to address these supplementary findings, taking the points raised in the 17 commentaries written by various participating analysts as a guideline. We identified three overarching themes in the commentaries and our own experiences. First, there was a need for more focus on theoretical depth and specificity. We refer to this aspect as “zooming in.” Second, multiple commentaries reflected on the broader implications of our results, elaborating on robustness and (the limits of) generalizability. We refer to this aspect as “zooming out.” Third, several commentaries addressed the appropriateness of the analysts' chosen statistical models given the MARP data.

In the following sections, we will first zoom in and address the issue of theoretical specificity. We will then zoom out and discuss to what extent the MARP results are robust and can be generalized. Subsequently, we discuss some methodological concerns, mostly related to the structure of the data. Finally, we will reflect on our experience of organizing a many-analysts project and highlight some lessons learned.

5. Concluding remarks

The main finding of the MARP is that religiosity and well-being are positively associated. This relation was established in a strictly confirmatory manner and seems robust against a plethora of different analytic decisions and strategies. In addition, the positive relation between individual religiosity and well-being appears stronger when religion is perceived to be normative in a particular country than when it is perceived as less normative. This moderating effect of cultural norms of religion was found consistently in the same direction but appears less robust than the main association between religiosity and well-being.

Many-analysts approaches are relatively new to the social sciences and we hope that they will become more widely adopted in the coming years. We believe the two main merits of a many-analysts approach are that it provides (1) an indication of the robustness of the effect on interest, and (2) a concrete demonstration of the variety of theoretical angles and statistical strategies that may be added to researchers' toolboxes. We would recommend the many-analysts approach especially for much-debated research questions that are tested using a fairly straightforward design (e.g., simple associations or effects from an existing theory instead of complex cognitive models for a new hypothesis).

We consider the MARP a positive example of team science and would like to thank the analysis teams for their efforts. In fact, we are intrigued by the creative contributions of the teams exploring different aspects of religiosity and well-being beyond our imposed research questions. We hope the MARP can serve as an inspiration for future many-analysts projects.

Gossiping was more about celebrities and ingroup members (over strangers); more about negative events overall, and yet for ingroup members, more positive gossiping; for content, more about moral topics

How ‘who someone is’ and ‘what they did’ influences gossiping about them. Jeungmin Lee, Jerald D. Kralik, Jaehyung Kwon, Jaeseung Jeong. PLoS, July 6, 2022.

Abstract: To understand, predict, and help correct each other’s actions we need to maintain accurate, up-to-date knowledge of people, and communication is a critical means by which we gather and disseminate this information. Yet the conditions under which we communication social information remain unclear. Testing hypotheses generated from our theoretical framework, we examined when and why social information is disseminated about an absent third party: i.e., gossiped. Gossip scenarios presented to participants (e.g., “Person-X cheated on their exam”) were based on three key factors: (1) target (ingroup, outgroup, or celebrity), (2) valence (positive or negative), and (3) content. We then asked them (a) whether they would spread the information, and (b) to rate it according to subjective valence, ordinariness, interest level, and emotion. For ratings, the scenarios participants chose to gossip were considered to have higher valence (whether positive or negative), to be rarer, more interesting, and more emotionally evocative; thus showing that the paradigm was meaningful to subjects. Indeed, for target, valence, and content, a repeated-measures ANOVA found significant effects for each factor independently, as well as their interactions. The results supported our hypotheses: e.g., for target, more gossiping about celebrities and ingroup members (over strangers); for valence, more about negative events overall, and yet for ingroup members, more positive gossiping; for content, more about moral topics, with yet all domains of social content communicated depending on the situation—context matters, influencing needs. The findings suggest that social knowledge sharing (i.e., gossip) involves sophisticated calculations that require our highest sociocognitive abilities, and provide specific hypotheses for future examination of neural mechanisms.


In this study we sought to test the extent to which gossiping behavior could be understood using a model of social information communication [7273]. This model is not unlike others, investigating the fundamental motives behind gossip behaviors [e.g., 20, 24, 26, 28]; however, our model extends the existing ones, providing a more comprehensive sociocognitive-neuroeconomic account of social information communication that more closely models the human mind/brain, allowing us to understand and explain the findings under one consistent and detailed framework. To test the six model predictions, we first asked participants to read various gossip scenarios (e.g., Person-X cheated on the final exam), and then asked whether they would gossip this information to others. The scenarios varied the target person that the gossip is about (ingroup, celebrities, or outgroup), valence (whether positive or negative), and content (eight different domains).

Main findings

The target main effect (celebrities > ingroup >> outgroup) provides evidence for interest more generally [491419], and relationship intimacy [223344953] and social influence [2425273548] more specifically as important factors driving people to gossip. Since intimacy or closeness implies having more meaning and influence in an individual’s day-to-day life, and influence or status implies social attention, power, and influence, both factors support the notion that functional significance strongly drives gossiping.

For valence, we found negative events to be spread more than positive ones. Indeed, the functional role of negative gossiping–e.g., to punish the target, protect receivers, and potentially promote oneself [2425273548839096]–appeared prominent and clear, especially given that we did not find evidence for negative gossiping being based simply on its intrigue or entertainment value (with ratings for ‘rarity’ comparable to positive scenarios and lower than positive scenarios for ‘interest level’).

Although, overall, negative valence promoted gossiping more than positively valenced scenarios, a persistent finding across the study nevertheless was that positively valenced scenarios perhaps rather surprisingly promoted gossip to a large degree: overall (i.e., not greatly lower than negative: 48.41% spreading for positive vs. 50.76% for negative), being higher than negative spreading for ingroup (target x valence), and being comparable to or even higher than negative spreading for specific content dimensions, even for celebrities and outgroup [61]. For example, the positive gossiping rate was especially high for carefairness, and altruism for all target groups, indicating that these positive acts resonate perhaps universally: such kindness and regard for others permeates the social network. Our results (including the ratings scores for subjective valence, interest and emotion) thus join the others that have found positive events to also strongly drive gossip [832575961]. Moreover, given that this occurred with all content domains (i.e., all showed significant positive gossiping), the results suggest that along with correction and ‘punishment’, positive feedback and information sharing comparably influence social behavior–not only for motives such as social bonding, but also for social control via strengthening and promoting it [26]. This is especially suggested given the high rates of positive gossiping across the moral domains. Indeed, rather surprisingly, we even found prosociality to produce more positive (care) than negative (harm) gossiping.

The results for content overall supported our hypothesis that, in general, moral dimensions would be spread most, and in particular, that prosociality (care/harm) and fairness (fairness/cheating) would be most spread. These align with previous findings that highlight the functional significance of gossip as it relates to cooperation, competition, and other moral dimensions, and we extend the findings to further dimensions of morality [253542646668]. Thus, almost diametrically opposed to the presumed trivial nature of gossip, we found the most impactful moral dimensions to be most spread, attesting to the importance of gossip on the regulation of societal members and the influence on future social interactions.

At the same time, scenarios representing all of the moral domains (and all content domains for that matter) were significantly spread at various rates, indicating the need to study these various domains and their differences more closely, with dominant paradigms that have focused on cooperation and competition, for example, useful in their own right, but not well representing other important social dimensions. For instance, the results for purity were particularly interesting, revealing that, on the one hand, the dimension appears to be universally meaningful (here: across all target groups), and yet, on the other hand, treated exceptionally, at least for perhaps particular targets (here: ingroup) and cultures (here: Korean). Indeed, gossip spreading about targets’ norm violation has been shown to depend on cultural context [6369]. Future research is therefore needed to clarify how and why individuals and societies respond the way they do to cases involving purity (e.g., gossiping vs. other means of communication, or perhaps even suppression)–and, in fact, for all individual dimensions of social interaction (i.e., moral and others). To be sure, our findings also point again to the significance of the relationships and interdependencies among the gossip parties themselves (i.e., gossiper-receiver, gossiper-target, and receiver-target) [20]. Moreover, we believe these questions are particularly ripe for additional modeling [97100] and neural imaging studies [101103] to help clarify the factors, their relationships, and the underlying mechanisms that drive the sociocognitive-neuroeconomic decisions involving social interaction and communication.

In addition to moral dimensions, we nonetheless also found that other types of knowledge may be highly valued under various circumstances; in other words, context is critical: for example, with more personal and day-to-day sociality (i.e., general social affairs and social-oriented) being more important with ingroup targets. Because this more seemingly mundane social information was differentially important for ingroup members, it suggests that its significance derives from the desire to be updated about the basic activities of those close to you [28]. From a sociocognitive perspective, it implies the need to maintain accurate knowledge of them: i.e., an accurate model of their minds, including their current knowledge, interests, intentions, activities, etc. [572739293]. Moreover, in doing this, a sense of solidarity and feelings of community among ingroup members also develops [2334464953].

A dominant finding in the two-way (target x valence) and three-way (content x valence x target) interactions was that spreading information about ingroup targets skewed positive, suggesting either that positive events were more meaningful and thus being spread more for ingroup targets [8390], or spreading negative information about them might be costlier [20]–and we found evidence for both–especially the latter. That is, for spreading negative information about ingroup members, the overall results suggest a general negativity-avoidance effect for ingroup targets. This result is in line with previous findings that people perceive a gossiper who shares favorable and therefore positive information about others more positively, which should be especially important among ingroup members [65104107].

Multiple factors likely dampen negative gossiping of ingroup members, including potential repercussions, other means to communicate to the target, and perhaps more leeway with and empathy toward equal and equivalently lower status individuals [202149108109]. Further empirical and computational research can help to delineate the potential influence of such factors on negative gossiping about ingroup members (including both the ‘ceiling’ and ‘floor’ components). In any case, this avoidance of negativity with ingroup members provides evidence for our predictions (Hypothesis 4b), and more specifically, that the net effect of the overall expected outcome influences the gossiper’s decision to gossip [2057].

Another dominant finding was that spreading information about celebrities generally skewed negative. This overall effect is likely at least partially due to the lowered risk of repercussions to the gossiper; but the evidence further suggests that the social status of celebrities is a major driving factor underlying the heightened negative gossiping, with the intention of lowering that of the celebrity (and thereby raising the gossiper’s own status, and perhaps the receivers’ as well), at least among those within the gossiper’s purview. This finding thus supports others showing that in an environment where vertical hierarchy exists, people with low status tend to gossip about high-status individuals with more power [2849]. Gossiping negatively about people with power (especially about moral contents) has been called a weapons-of-the-weak mechanism [110] or subordinate strategy [see 111] whereby low-status, relatively powerless individuals use gossip as a weapon to pressure more privileged individuals [26]. An examination of the results for the specific content domains further showed that for higher-status individuals (celebrities), loyalty and humility warrants action (i.e., gossiping), as do cases of trying to beat others in underhanded ways or otherwise cheating the system [83].

In contrast, the greater interest in altruism (over selfishness) and fairness (over cheating) for both ingroup and outgroup targets suggests that such selfless and fair acts are especially impressive when conducted by those of lesser status and means. Moreover, authority and positive competition with ingroup targets appear to generate action (receiving higher gossiping rates), suggesting that these topics are more relevant among relatively lower (compared to celebrities) and more equal status ingroup members, with positive competition suggesting scenarios of ambition or achievement resonate more. At the same time, selfish behavior of ingroup and outgroup people are relatively less spread, with such acts appearing to be more tolerated in those with less means and status. In sum, our results support our hypotheses and others findings that status is a major factor determining the extent of and types of gossip [2849]. Further empirical and computational work can extend our findings by delineating exactly how status interacts with other factors to promote social interactions, such as with social information spreading [e.g., 112]. In any case, the evidence for status considerations again attests to the importance of functional value driving gossiping behavior.

For outgroup targets, we generally found a relative lack of interest, with many results significantly weaker compared to ingroup and celebrities, and thus supporting our hypotheses (especially Hypothesis 1). Even this result may be a bit surprising if one generally construes the “outgroup” as outsiders, and thus potential threats, enemies, etc. However, comparable to the findings of others, we found that the usual use of the “outgroup” concept requires a more nuanced appreciation [113]. Those considered as viable threats likely evoke sufficient interest that warrants action, rather than generally being ignored. Yet it is also unfortunately probable that a relative lack of empathy can be seen with strangers in general, making it more difficult to care sufficiently in their affairs [114115].

At the same time, however, even for outgroup members (i.e., complete strangers), some content domains were generally important in our study, including a heightened rate of information spreading for both care and harm (i.e., the prosociality domain), fair and cheating (i.e., the fairness domain), and altruism and selfishness (i.e., the social-oriented domain). This heightened spreading for all three target groups indicates a strong interest in prosocialityfairness, and social-orientation that is worthy of disseminating to others regardless of the actor involved–i.e., prominent universals for all members of society. These results are in line with others that show that stories of strangers can also elicit interest and therefore produce gossip if the events can offer useful life lessons and strategies [428].

Study limitations

There are some study limitations that should be considered. First, we did not test our hypotheses in a more natural context where gossip triads are interacting freely in spontaneous situations. The clear advantage of field studies using various methods such as eavesdropping [68116], daily diary surveys [117], and experience sampling methods [59] is that a potentially rich set of observational data that reflects real-life gossiping behavior can be collected [see 5996]. And there are indeed cases where behavior observed in the laboratory may not appear outside it [118119], requiring all laboratory studies to consider the ecological validity. Here, we took several steps to minimize the gap between the natural and laboratory settings. First, we note that much of the information people learn about others (including ingroup members) these days comes from texts read on electronic devices (phones, computers). Second, prior to the experiment we asked participants to submit the names of their closest friends (i.e., ingroup) so that the ingroup scenarios would feel like actually receiving information about them. We also chose well-known Korean celebrities, and used foreign names as strangers for the outgroup. Third, we also gave participants thorough instruction to assume that every piece of information provided during the task is real; and we received several comments during the post-experiment verbal interview directly stating that the scenarios felt realistic, even being “shocked” by some of the extreme cases (e.g., with harm or degradation contents). Fourth, and importantly, our experiment was also designed to address ecological validity directly by asking participants to rate each scenario according to subjective valence, ordinariness, interest level, and emotion (after they chose whether to gossip or not). The scenarios participants chose to gossip were considered to have higher valence (whether positive or negative), to be rarer, more interesting, and more emotionally evocative—thus showing that they were meaningful to the participants. Fifth, and finally, the fact that many of our results match those of other studies helps to support the validity of our experimental paradigm.

Indeed, experimental paradigms such as ours help complement others by enabling clearer and more precise testing of relevant factors. Nonetheless, it is clear that future research is needed to further validate our findings, not only with methodologies described above, but others as well, such as curating and examining natural interactions on social media (such as Facebook or Twitter).

A second limitation is that the behavioral-based study is nevertheless limited in the precision it can achieve, not enabling tests of more detailed processes in our theoretical framework. Future research can hopefully utilize our framework and test paradigm to examine, for example, the neural processes underlying the gossip decision (such as the benefits vs. costs computations in the brain that drive gossiping). A few studies have investigated neural activity during gossip [101103], but many unknown factors remain. Third, it is clear that many more factors remain to be examined, such as how the choice of receiver influences the gossip decision, as the possible gossiper may keep quiet [20] or selectively expose the target information [49120] depending on the receiver’s identity. Other factors include more specific detail about the target and event content, especially to better characterize the more nuanced context effects.

A sigh of relief actually confers relaxation

The Psychophysiology of the Sigh: II: The Sigh From the Psychological Perspective. Elke V lemincx, Liza Severs, Jan-Marino Ramirez. Biological Psychology, July 6 2022, 108386.


• Sighs have essential regulatory functions.

• Sighs may function as psychophysiological resetters.

• Sighs may contribute to psychophysiological flexibility.


A sigh is a distinct respiratory behavior with specific psychophysiological roles. In two accompanying reviews we will discuss the physiological and psychological functions of the sigh. The present review will focus on the psychological functions of the sigh. We discuss the regulatory effects of a sigh, and argue how these effects may become maladaptive when sighs occur excessively. The adaptive role of a sigh is discussed in the context of regulation of psychophysiological states. We propose that sighs facilitate transitions from one psychophysiological state to the next, and this way contribute to psychophysiological flexibility, via a hypothesized resetting mechanism. We discuss how a sigh resets respiration, by controlling mechanical and metabolic properties of respiration associated with respiratory symptoms. Next, we elaborate on a sigh resetting emotional states by facilitating emotional transitions.

We attempt to explain the adaptive and maladaptive functions of a sigh in the framework of stochastic resonance, in which we propose occasional, spontaneous sighs to be noise contributing to psychophysiological regulation, while excessive sighs result in psychophysiological dysregulation. In this context, we discuss how sighs can contribute to therapeutic interventions, either by increasing sighs to improve regulation in case of a lack of sighing, or by decreasing sighs to restore regulation in case of excessive sighing. Finally, a research agenda on the psychology of sighs is presented.

Keywords: Sighsemotionsregulationflexibility

2. Sighing as a maladaptive behavior

Physiologically, complex breathing patterns serve to maintain blood gas levels consistent with metabolic need (Ben‐Tal & Tawhai, 2013). Breathing patterns resulting in deviations from normoxia and normocapnia, may cause respiratory symptoms. However, respiratory symptoms caused by dysfunctional breathing patterns can also occur in the presence of efficient gas exchange (Courtney and Cohen, 2006Hornsveld et al., 1996Hornsveld and Garssen, 1997Vlemincx et al., 2012). While the term ‘dysfunctional breathing’ has been long used for a variety of maladaptive breathing patterns, recently, a classification system has been proposed to better understand and define dysfunctional breathing patterns (Boulding, Stacey, Niven, & Fowler, 2016). One type of dysfunctional breathing is periodic sighing (Boulding et al., 2016), also known as sigh syndrome, sighing dyspnea or sighing breathing (Aljadeff et al., 1993Hurvitz and Weinberger, 2021Sody et al., 2008Wong et al., 2007Wong et al., 2009), which consists of frequent sighing leading to hyperventilation-induced hypocapnia and associated respiratory symptoms.

Sighing more frequently than normal has been associated with various disease states. Sigh frequency in chronic low back pain patients is higher than in depressed and healthy controls, and correlates with pain ratings across periods of sitting, standing, reclining and walking (Keefe & Hill, 1985). In patients with traumatic brain injury, sighing is a frequent pain behavior, observed more during nociceptive exposure (compression of the nail bed), than during baseline, recovery and non-nociceptive exposure (non-invasive blood pressure measurements) (Nazari et al. 2018). In an ambulatory study in rheumatoid arthritis patients, sighing was associated with depression, yet not experienced pain (Robbins, Mehl, Holleran, & Kasle, 2011). Furthermore, while not a disease state, sighing is considered a symptom of motion sickness related to nausea (Leung & Hon, 2019).

In addition, sighing is associated with respiratory disease. Frequent sighing was present in a case of difficult-to-treat asthma (Prys-Picard, Kellett, & Niven, 2006). In this patient, sighs increased during exercise, and decreased after breathing retraining. In patients with respiratory disease, sighing co-occurred with emotional states (Stevenson & Ripley, 1952). In patients with so-called ‘hyperventilation syndrome’, sigh rate was higher during quiet sitting compared to healthy controls (Han et al., 1997), and while listening to soft music compared to healthy controls and asthmatics (Hormbrey, Jacobi, Patil, & Saunders, 1988).

Furthermore, the association between anxiety disorders and frequent sighing has been well established in laboratory studies; persons with panic disorder, post-traumatic stress disorder and chronic anxiety, sigh more frequently than (healthy) controls during quiet sitting and resting periods (Abelson et al., 2001Abelson et al., 2008Abelson et al., 2010Blechert et al., 2007Han et al., 1997Schwartz et al., 1996Tobin et al., 1983Wilhelm et al., 2001aWilhelm et al., 2001b). Furthermore, the frequent sighing in panic disorder patients has been associated with chronic hypocapnia (Wilhelm et al., 2001a) and respiratory dysregulation indicated by high respiratory irregularity (Abelson et al., 2001Martinez et al., 2001Yeragani et al., 2002), suggesting that panic disorder patients sigh ‘excessively’ (i.e. sigh in excess of metabolic need). Importantly, however, it should be noted that these findings in anxiety disorders were observed in experimental laboratory studies, and have not been replicated in ambulatory studies. In ambulatory studies, there is no evidence that excessive sighing occurs in panic disorder (Pfaltz et al., 2009Pfaltz et al., 2010). In addition, an ambulatory study did not find evidence for a relationship between sighing and dispositional negative emotionality (Danvers et al., 2021).

These findings could potentially, in part, be explained by the fact that ‘baselines’ or quiet sitting, resting laboratory conditions do not necessarily represent daily life (Wilhelm & Grossman, 2010). While research labs are commonly unfamiliar to participants (e.g. unfamiliar rooms, equipment, procedures, experimenters), daily life is the opposite. Previous studies have shown that panic disorder patients show increased neurobiological stress reactivity in response to contextual stressors, among which novelty (Abelson, Khan, Liberzon, & Young, 2007). Accordingly, quiet sitting and resting baseline measurements in the laboratory may constitute contextual stress or anxiety in persons with panic disorder. This reasoning would support a relationship between excessive sighing and state anxiety (which we will discuss in more detail below), rather than dispositional anxiety.

This rationale may help to understand why frequent sighing occurs in the disease states described above. We propose that the sigh is not only an important regulator of respiration (e.g. blood gases and lung compliance), but also an often overlooked contributor to the homeostatic regulation of psychological states, including stress, emotions (e.g. anxiety), and perceived symptoms and sensations (e.g. pain, dyspnea, nausea) (Vlemincx et al., 2010Vlemincx et al., 2013). This would implicate that sighs occur frequently during these psychological states, and therefore, may become a byproduct of these psychological states associated with disease states. While sighs may be adaptive regulation mechanisms as long as their frequency matches metabolic demand, sighs may become maladaptive when occurring excessively. Excessive sighing may result in chronic hypocapnia (Wilhelm et al., 2001a), which may lead to widespread bodily symptoms (including autonomic, respiratory, cardiac, motor and emotional symptoms), overlapping with symptoms in a broad range of disease states, such as chronic pain, respiratory and cardiovascular disease, anxiety and neurological disorders. In turn, these bodily symptoms may require further regulation, and therefore further increase sighing, initiating a vicious cycle of emotions, symptoms and sighs. In other words, while sighs consistent with metabolic need may be adaptive to regulate psychological states, excessive sighs may exacerbate symptoms in various disease states. Below, we will detail the adaptive role of sighs.

Consistently over time, polygenic scores that predict higher earnings, education and health also predict lower fertility

Human Capital Mediates Natural Selection in Contemporary Humans. David Hugh-Jones & Abdel Abdellaoui. Behavior Genetics, Jul 6 2022.

Abstract: Natural selection has been documented in contemporary humans, but little is known about the mechanisms behind it. We test for natural selection through the association between 33 polygenic scores and fertility, across two generations, using data from UK Biobank (N = 409,629 British subjects with European ancestry). Consistently over time, polygenic scores that predict higher earnings, education and health also predict lower fertility. Selection effects are concentrated among lower SES groups, younger parents, people with more lifetime sexual partners, and people not living with a partner. The direction of natural selection is reversed among older parents, or after controlling for age at first live birth. These patterns are in line with the economic theory of fertility, in which earnings-increasing human capital may either increase or decrease fertility via income and substitution effects in the labour market. Studying natural selection can help us understand the genetic architecture of health outcomes: we find evidence in modern day Great Britain for multiple natural selection pressures that vary between subgroups in the direction and strength of their effects, that are strongly related to the socio-economic system, and that may contribute to health inequalities across income groups.


Previous work has documented natural selection in modern populations on variants underlying polygenic traits (Beauchamp 2016; Kong et al. 2017; Sanjak et al. 2018). We show that correlations between polygenic scores and fertility are highly concentrated among specific subgroups of the population, including people with lower income, lower education, younger first parenthood, and more lifetime sexual partners. Among mothers aged 22+, selection effects are reversed. Furthermore, the size of selection effects on a polygenic score correlates with that score’s association with labour market earnings. Strikingly, some of these results were predicted by Fisher (1930), pp. 253-254. The economic theory of fertility gives a parsimonious explanation for these findings. Because of the substitution effect of earnings on fertility, scores are selected for when they correlate with low human capital, and this effect is stronger at lower levels of income and education.

Polygenic scores which correlate with lower earnings and less education are being selected for. In addition, many of the phenotypes under positive selection are linked to disease risk. Many people would probably prefer to have high educational attainment, a low risk of ADHD and major depressive disorder, and a low risk of coronary artery disease, but natural selection is pushing against genes associated with these traits. Potentially, this could increase the health burden on modern populations, but that depends on effect sizes. Our results show that naïve estimates can be affected by sample ascertainment bias. There may be remaining sources of ascertainment bias after our weighting; if so, we expect that, like the sources of ascertainment we have controlled for, they probably bias our results towards zero. Researchers should be aware of the risks of ascertainment when studying modern natural selection.

We also do not know how estimated effect sizes of natural selection will change as more accurate polygenic scores are produced, or whether genetic variants underlying other phenotypes will show a similar pattern to those studied here. Also, effects of polygenic scores may be inflated in population-based samples, because of indirect genetic effects, gene-environment correlations, and/or assortative mating (Lee et al. 2018; Selzam et al. 2019; Kong et al. 2018; Howe et al. 2021), although we do not expect that this should change their association with number of offspring, or the resulting changes in allele frequencies. Although effects on our measured polygenic scores are small even after weighting, individually small disadvantages can cumulate to create larger effects. Lastly, note that our data comes from people born before 1970. Recent evidence suggests that fertility patterns may be changing (Doepke et al. 2022). Overall, it is probably too early to tell whether modern natural selection has a substantively important effect on population averages of phenotypes under selection.

Because selection effects are concentrated in lower-income groups, they may also increase inequality with respect to polygenic scores. For example, Figure 8 plots mean polygenic scores for educational attainment (EA3) among children from households of different income groups. The blue bars show the actual means, i.e. parents’ mean polygenic score weighted by number of children. The grey bars show the hypothetical means if all households had equal numbers of children. Natural selection against genes associated with educational attainment is stronger at the bottom of the income distribution, and this increases the differences between groups. Overall, natural selection increases the correlation of polygenic scores with income for 28 out of 33 polygenic scores, with a median percentage increase of 16.43% in the respondents’ generation (Appendix Table 5). If inequalities in polygenic scores are important for understanding social structure and mobility (Belsky et al. 2018; Rimfeld et al. 2018; Harden 2021), then these increases are substantive. Similarly, since many polygenic scores are predictive of disease risk, they could potentially increase health inequalities. In general, the evolutionary history of anatomically modern humans is related to disease risk (Benton et al. 2021); understanding the role of contemporary natural selection may help researchers to map the genetic architecture of health disparities.

Fig. 8

Mean polygenic score for educational attainment (EA3) of children by household income group. Blue is actual. Grey is hypothetical in the absence of selection effects (Color figure online)

Existing evidence on human natural selection has led some to “biocosmic pessimism” (Sarraf and Feltham 2019). Others are more sanguine, and argue that natural selection’s effects are outweighed by environmental improvements, like those underlying the Flynn effect (Flynn 1987). The evidence here may add some nuance to this debate. Patterns of natural selection have been relatively consistent across the past two generations, but they are not the outcome of a single, society-wide phenomenon. Instead they result from opposing forces, operating in different parts of society and pulling in different directions.

Any model of fertility is implicitly a model of natural selection, but so far, the economic and human genetics literatures have developed in parallel. Integrating the two could deepen our understanding of natural selection in modern societies. Economics possesses a range of theoretical models on the effects of skills, education and income (see Hotz et al. 1997; Lundberg and Pollak 2007). One perennial problem is how to test these theories in a world where education, labour and marriage markets all interact. Genetic data, such as polygenic scores, could help to pin down the direction of causality, for example via Mendelian randomization (Smith and Shah 2003). Conversely, economic theories and empirical results can shine a light on the mechanisms behind natural selection, and thereby on the nature of individual differences in complex traits and disease risk.