This study investigates the links between life history strategy (LHS), IQ, and individualism/collectivism (IND/COL) across Russian regions. It is the first to examine regional differences in LHS, illustrating how biological trade-offs are linked to cultural values within a country and separately considering the role of cognitive abilities in this relationship. We hypothesize that cultural differences between regions can be understood through human behavioral ecology, specifically the trade-offs associated with LHS. Data from 83 Russian regions were used to create indices for slow LHS (sLHS) and IND/COL. The sLHS index included indicators like teenage fertility rates, rates of third or higher births, average height, educational attainment, and interest in human sexual behavior (using Google Trends data). The IND/COL index was constructed from indicators like the proportion of multigenerational and single-person households, divorce rates, and search data indicating ingroup identity expression. Regional IQ scores were derived from a large-scale online test administered to over 230.000 individuals across these regions. Our findings reveal strong positive correlations between sLHS, IQ, and IND/COL across Russian regions. Mediation analysis suggests that IQ likely fully mediates the relationship between sLHS and IND/COL. Geographical analysis showed clear patterns of spatial clustering, with gradients linked to latitude and altitude. Additionally, five latent regional profiles emerged from the data, indicating distinct patterns among the regions. These results, while acknowledging certain limitations, underscore the importance of LHS in understanding regional cultural differences. They also point to the need for Russian social policies to adapt to the unique characteristics of each region.

individualism/collectivism, regional IQ, life history strategy, cultural values, phenotypic plasticity, human behavioral ecology, regional differences, Russia

Culture sets the fundamental principles for human interaction, which makes it quite natural to focus on the study of cultural characteristics when examining similarities and differences between societies. In the social sciences, the approach of cultural dimensions dominates in these contexts (Gallyamova and Grigoriev 2023). For a long time, the majority of studies focused on cultural differences between countries, with the primary focus being national culture. Indeed, national cultural differences provide important insights into the functioning of society, such as helping to answer why countries differ in institutional quality and levels of economic development.

It is important to note that a significant portion of the research focuses on the individualism–collectivism dimension (IND/COL), which is a reliable predictor of socio-economic development (Bakhtigaraeva et al. 2021). Generally, individualistic (IND) values contribute to economic growth and foster qualitative changes in social structures, promoting greater egalitarianism and economic efficiency. Overall, recent studies aimed at clarifying various cultural models show that this dimension is primary and encompasses the widest range of cultural differences (Gallyamova and Grigoriev 2023). The IND/COL dimension reflects whether the individual or the group is the central element of social interaction in society. This balance influences the formation of social institutions, which, in turn, either support social cohesion or promote individual autonomy regulated by universal norms (Minkov et al. 2023a). Notably, within some countries, there may also be significant regional differences in IND/COL that deviate from the national cultural average. This has recently led to increased interest in studying regional cultures within a single country (Ebert et al. 2024), as this approach enables researchers to identify cultural differences that may be overlooked when focusing solely on national cultures (Sokolov et al. 2025).

Current cross-cultural research often overlooks the complex role of the ecological context in shaping cultures, resulting in a limited understanding of the mechanisms and causes that determine behavior within cultural groups (Berry 2019). At the same time, culture is largely viewed as a subjective construct (Minkov and Kaasa 2021) rather than as a specific set of collective adaptations to the environment. To address this issue, we conceptualize culture within the framework of life history theory as the result of trade-offs in the allocation of limited resources (including energy and time), which are expressed in behavioral traits and strategies. This approach helps explain how individuals in groups collectively allocate resources throughout their lives, balancing competing biological needs such as survival, growth, reproduction, and mating (Figueredo et al. 2017), thereby generating a specific culture within a specific behavioral ecology (Figure 1).

Figure 1.

The human life history strategy framework. Source: Adapted from Komyaginskaya et al. (2024), who modified the original concept from Stott et al. (2024), applying the model to human subjects

To understand regional culture through the lens of human behavioral ecology, it is first necessary to consider the evolutionary basis of human behavior. If the primary goal of all biological species is gene transmission, then life history strategy (LHS) is the set of behavioral trade-offs required for successful gene transmission (Figueredo et al. 2005). LHS represent the behavioral adaptations of different species to environmental pressures that may vary between or within species. Thus, behavioral and cognitive differences between groups are shaped by various environmental factors, such as climate and pathogen prevalence, which exert selective pressure on individuals. Environmental harshness – in the form of threats, deprivations, and unpredictability – shapes a strategy for allocating limited resources (including energy and time) that is reflected in behavioral repertoires. Adaptive traits such as longevity, number of offspring, or birth intervals are selected for through multilevel selection and adapt via phenotypic plasticity (Figueredo et al. 2017). The LHS can be viewed as a continuum ranging from relatively fastest to slowest strategies and can be generalized to a fundamental trade-off involving the allocation of limited resources across two dimensions: (1) the choice between immediate versus delayed parenthood and (2) the choice between investing in increasing the number of offspring versus enhancing the upbringing of existing offspring (i.e., the trade-off between offspring quantity and offspring ‘quality’).

Under conditions of high external (uncontrolled) mortality, the fast life history strategy (fLHS) is more adaptive, as it focuses on maximizing the use of all available resources in the short term. With fLHS, resources, energy, and time are directed primarily toward reproduction rather than raising offspring, resulting in a larger number of children (Figueredo et al. 2005). In contrast, under conditions of low external mortality, it becomes feasible to invest more resources in development, which increases competitiveness and supports the formation of a slow life history strategy (sLHS). Moreover, under such conditions, it is more practical to use resources in the long term, investing more in the development of children than in reproduction and mating, thereby increasing the phenotypic quality of offspring (Figueredo et al. 2017). Although humans as a biological species generally gravitate toward the pole of sLHS by the continuum, closer examination reveals significant differences between populations within a single species.

Human populations around the world have evolved in diverse environmental and social contexts, making population-level differences in LHS virtually inevitable. These differences in LHS have become a fertile area of scientific research. Psychological studies of human LHS often focus on the influence of environmental factors on the formation of adaptive trade-offs, which manifest in personality traits and behavioral patterns. However, behavioral repertoires also shape social norms that govern the functioning of society. Accordingly, the integrated model of social biogeography (Figueredo et al. 2021) explicitly addresses differences in social structure, where environmental conditions influence behavioral traits associated with LHS. This model highlights how these traits lead to variations in social structures and their consequences, such as social equality, intragroup harmony, economic development, levels of cooperation and conflict, and the stability of social institutions.

According to the integrated model of social biogeography, when sLHS prevails, selection favors individuals with high generalized prosociality and the ability to form stable relationships characterized by a high level of cooperation. In such populations, societies with strong intragroup cooperation and economic specialization develop, leading to greater social equality and economic advancement. Conversely, when fLHS prevails, individuals who can endure constant competition are better adapted; in these contexts, cohesion and cooperation are limited to a narrow group, while individuals outside this group are perceived as threats. In such societies, high levels of intragroup and intergroup competition lead to challenges in economic development (e.g., increased corruption), and crime and violence are more prevalent.

Thus, at the national level, variations in the speed of an individual’s life history have been found to be associated with factors such as cognitive abilities, as assessed by IQ tests, and human capital (Boutwell et al. 2013; de Baca and Figueredo 2014; Figueredo et al. 2017). If we consider the Flynn effect – the improvement in IQ scores from generation to generation – through the lens of LHS theory, we can conclude that this effect results from a gradual slowdown in life history speed during the 20^th century, as environmental conditions significantly improved over this period. For instance, the 20^th century saw a general decline in fertility rates and a trend toward increasing care and investment in children (the reproduction/parenting trade-off). Consequently, with greater investment in offspring, subsequent generations have the opportunity to develop for longer periods, resulting in higher cognitive abilities within the range of phenotypic plasticity (Woodley of Menie et al. 2024).

It is important to understand that, although differences in cognitive abilities are largely due to biological predispositions (Figueredo et al. 2021), this cannot be the sole explanation in the case of humans. On the one hand, the prevalence of genes associated with higher cognitive abilities is necessary for elevated cognitive indicators at the national level. On the other hand, it is equally important that group motivation, rooted in cultural values, be directed toward creating a social environment that maximizes the intellectual potential of individuals. For example, research has found that IND is positively associated with national IQ (Zengrui et al. 2017). Thus, national differences in cognitive achievement may be partly attributable to cultural motivation (Minkov et al. 2016). At the same time, it is important to note a research gap in this area. If culture fosters an environment that enhances intellectual potential, then there are likely to be reciprocal relationships among LHS, IQ, and IND/COL. However, due to the lack of time-series data, this hypothesis cannot yet be directly tested. Therefore, this study proposes a simple directional relationship: from population response to environmental conditions (LHS), through the development of a specific cognitive phenotype (IQ), to the extended phenotype and cultural niche construction (IND/COL).

Although LHS and culture are interrelated phenomena, it is important to recognize that, unlike animals that rely solely on biological mechanisms to manage their resources, humans have evolved the ability to create cultural mechanisms and shape niches that support collective adaptations. These cultural adaptations include the formation of social structures that enable more efficient allocation of time and energy within a group, allowing for a level of social organization that far surpasses that of other species. In other words, human societies are unique in their ability to collectively regulate biological trade-offs through social norms. For example, in China, the biological trade-off between offspring quantity and quality was deliberately managed through government restrictions on family size, leading to a sharp decline in fertility rates (Gallyamova and Grigoryev 2025). Therefore, the connection between cultural orientations and regional demographic characteristics may be attributed to the influence of cultural values and attitudes on self-preservation, reproductive, and matrimonial behaviors, as well as on individuals’ migration intentions. In this way, cultural orientations can contribute to shaping the demographic landscape of a region.

Thus, it is not accurate to claim that culture is merely a derivative of LHS; rather, it comprises specific institutions developed through ultrasociality that enable individuals to cooperatively address biological challenges. These cultural institutions facilitate resource allocation within groups, supporting survival, reproduction, maintenance, and growth. We therefore propose viewing culture as a set of collective solutions to these LHS trade-offs. In other words, human societies, unlike other species, can regulate biological trade-offs collectively through social norms. Consequently, differences in IND/COL may largely reflect population-specific mechanisms for resource allocation.

In this study, we focus on regional differences within Russia, a unique context where regions with diverse ecological conditions and historical backgrounds coexist. Early studies on IND/COL in Russia indicate that most Russians today support IND norms, likely due to the strong European influence on Russian culture. Moreover, it is suggested that the level of IND in Russia may gradually increase over time (Latova 2016).

However, at the regional level, the degree of IND can vary significantly; for example, Stavropol exhibited IND levels similar to those in India, while Tyumen was more comparable to Canada and the Netherlands (Latov and Latova 2007). Furthermore, studies by Minkov et al. (2023a) and Sokolov et al. (2025) on regional differences in IND/COL in Russia have reinforced the concept of ‘regional culture’ as a valuable analytical framework. Specifically, these differences in IND/COL have been linked to variations in electoral behavior and political protest activity (Gallyamova and Miller 2024; Shcherbak 2023). Additionally, several studies have explored the historical continuity between the social structures of the Russian Empire and those of modern Russia. For instance, in regions where nuclear families predominated according to the 1897 census, residents were more likely to support democratic candidates in the elections to the Constituent Assembly in 1917 – favoring the Cadets – as well as in the presidential elections of 1996–2000 – supporting G. Yavlinsky (Kravtsova and Libman 2023).

Several studies have focused on regional differences in IND/COL within Russia. Data from representative sociological surveys conducted in 14 regions revealed that IND/COL exhibited the most significant regional variation among all of Hofstede’s cultural dimensions (Bakhtigaraeva et al. 2021; Bryzgalin and Nikishina 2020). Additionally, gross regional product and climatic demands were positively associated with levels of IND (Maklasova 2020). These findings were corroborated by a larger study, which found positive correlations between IND levels and latitude, gross regional product, the percentage of ethnic Russians in a region, and the level of innovation (Minkov et al. 2023b), as well as with water availability and average wages (Sokolov et al. 2025).

However, previous studies on this topic have several limitations. In some, the analysis is restricted to a small number of regions, and the IND/COL indices are derived from population survey data (Bakhtigaraeva et al. 2021; Minkov et al. 2023b). In others, certain indicators selected to assess IND/COL do not always capture the essence of this dimension or may be entirely misleading (e.g., population distribution by per capita income; crude rate of natural increase) (Maklasova 2020). A key strength of our study is that we are the first to apply the LHS trade-offs concept to explain regional cultural differences within a single country. We anticipate positive relationships between regional sLHS, IQ, and IND/COL measures, as well as a positive indirect effect of sLHS on IND/COL through IQ. Additionally, we aim to test whether spatial clustering is evident in Russia for these variables to form more reliable and detailed conclusions about the relationships between biological trade-offs and culture. Moderate spatial autocorrelation has previously been observed for variables such as crime rates, infant mortality, and urbanization, while weak spatial autocorrelation has been found for income levels, educational attainment, and migration in Russian regions (Grigoriev 2018).

The analysis included 83 regions of Russia. Using regional data, indices for LHS and IND/COL were constructed by calculating the sum of weighted indicators based on the results of principal component analysis.

The following data were used as indicators for the sLHS index, reflecting the main trade-offs of LHS – namely, reproduction (mating versus parenting, quantity versus quality), maintenance, and development (Figueredo et al. 2021):

1. Age-specific fertility rate for women aged 15–17 for the period 2012–2022.
2. Total fertility rate of third and subsequent births for the period 2018–2022.

Both sourced from the interdepartmental database of the Federal State Statistics Service (Rosstat) (https://rosstat.gov.ru/emiss).

1. Average height (in centimeters) based on the “Sample Survey of the Population’s Diet,” conducted by Rosstat in 2018 (https://gks.ru/free_doc/new_site/food18/index.html).
2. Education level indicator – averaged for the period 2021–2023 and calculated according to the methodology approved by Russian Government Resolution No. 542 of April 3, 2021.
3. The number of Google search queries on the topic “Human sexual behavior,” estimated on a scale from 0 to 100, where 100 corresponds to the highest relative popularity of the query. A higher score indicates a greater share of related queries within the region’s total search volume.

The sLHS index, where low values indicate a tendency toward fLHS and high values indicate a tendency toward sLHS (see Figure 2), accounted for over 55% of the variance across the included indicators, with a reliability coefficient (Cronbach’s α) of 0.81. The index showed a correlation of .34 with population density.

The following data were used as indicators for the IND/COL index, in accordance with the literature (see Gong et al. 2021; Vandello and Cohen 1999; Yamawaki 2012; see also Pelham et al. 2022):

1. The proportion of multi-generational households among all private households. This includes households with one or both parents of the spouses, children, other relatives, and/or unrelated persons.
2. The proportion of single-person households among all private households.

Both indicators were calculated based on data from the 2010 All-Russia Population Census (https://rosstat.gov.ru/free_doc/new_site/perepis2010/croc/results.html).

1. The crude divorce rate, averaged over 2006–2023, calculated as the ratio of the number of divorces to the average annual population of the region, according to Rosstat data (https://rosstat.gov.ru/emiss).

These data were adjusted for regional housing affordability using regression analysis, retaining only the residuals – that is, the portions of the data not explained by housing affordability.

Additionally, statistics on Google Trends queries were used to reflect the degree of ingroup identity expression:

1. Queries on the topic of “People”.
2. Queries on the topic of “We”.

The IND/COL index, where lower values indicate greater COL and higher values indicate greater IND (see Figure 3), accounted for over 61% of the variance in the selected indicators, with a reliability coefficient of 0.84. The correlation of our index with IND/COL indices from other studies was relatively high: 0.91 (Minkov et al. 2023b; Sokolov et al. 2025), and 0.62 (Bakhtigaraeva et al. 2021).

Finally, the used regional IQ estimates were gathered through a voluntary online military fitness test for contract service (Sugonyaev et al. 2018). This cognitive test, available on the Russian Ministry of Defense’s website from September 2012 to December 2017, consisted of 30 tasks assessing verbal, numerical, spatial, and perceptual abilities, with a 15-minute time limit. After excluding incomplete and repeated entries, the final sample comprised 238.619 men aged 18–40 years from all Russian regions. Regional average IQ scores were calculated and standardized for comparison (see Figure 4). In multiple reviews, the authors confirmed the validity of these estimates by analyzing psychometric characteristics, comparing results with data from testing in controlled conditions, and addressing potential threats to data reliability.

Figure 2.

Map of Russian regions by sLHS index (N = 83). Source: Created by the authors

Figure 3.

Map of Russian regions by IND/COL index (N = 83). Source: Created by the authors

Figure 4.

Map of Russian regions by average IQ score (N = 83). Source: Created by the authors

The rankings of Russian regions based on the focal variables are presented in Table 1.

There was a strong positive correlation (p < 0.001) among the variables: between IND/COL and sLHS (r = 0.71), IND/COL and IQ (r = 0.81), and sLHS and IQ (r = 0.85). Additionally, Moran’s I indices indicated moderate to strong spatial autocorrelation for each variable: IND/COL (I = 0.72), sLHS (I = 0.48), and IQ (I = 0.61).

The results of the mediation analysis are shown in Figure 5. Model diagnostics indicated that, despite significant autocorrelation in the variables, there was no significant autocorrelation in the residuals of the regression model according to the Durbin – Watson criterion (DW = 2.13, p = 0.600), nor was there evidence of multicollinearity (VIF < 3.64). The proportion of explained variance for IND/COL with sLHS as the sole predictor was 51%; with both predictors, accounting for both direct and indirect effects, it rose to 66%. The proportion of explained variance for IQ was 72%, with IQ showing full mediation in the relationship between sLHS and IND/COL, accounting for 87% of the effect.

Moreover, there was a positive correlation with latitude for IND/COL (r = 0.63, p < 0.001), sLHS (r = 0.34, p = 0.002), and IQ (r = 0.58, p < 0.001). Conversely, a negative correlation was observed with altitude for IND/COL (r = –0.61, p < 0.001), sLHS (r = –0.63, p < 0.001), and IQ (r = –0.72, p < 0.001), as well as with longitude for sLHS (r = –0.33, p = 0.002) and IQ (r = –0.22, p = 0.041), except for IND/COL (r = 0.12, p = 0.288).

We then used Geographically Weighted Regression (GWR), which analyzes local subsets of data around each point of interest, to examine the relationship between IND/COL, sLHS, and IQ across different locations. Unlike conventional regression, which provides a single global estimate of associations between variables, GWR generates local estimates that vary across locations. This approach allows us to identify spatial non-stationarity, meaning it reveals how relationships between variables change across geographic regions.

Using approximately 37–38 nearby points for each location provided the best model fit. The association between sLHS and IND/COL was consistently positive, ranging from 0.03 to 1.05, indicating a uniformly positive relationship, though with varying strength across locations. In contrast, the association with IQ was generally positive but ranged from –4.33 to 37.18, meaning that in some areas, higher IQ was strongly associated with higher IND, while in others, the effect was weaker or even slightly negative. Overall, the GWR model explained about 85% of the variance in IND/COL, indicating a strong but locally varying association. Similarly, for IQ, the association with sLHS was generally positive, ranging from –0.01 to 0.03, with variations in strength across locations. This model explained approximately 78% of the variance in IQ.

Finally, Spatial Lag Models (SAR) were calculated, and Moran’s I for the residuals revealed no significant spatial autocorrelation, indicating that the models fully account for spatial relationships. When sLHS and IND/COL were analyzed separately, both the direct effect of sLHS within a region (p < 0.001) and its indirect effect through neighboring regions (p = 0.006) were significant. However, when IQ was added to the model, both the direct (p = 0.116) and indirect (p = 0.986) effects of sLHS became non-significant. For IQ, only the direct effect on IND/COL was significant (p < 0.001), while the indirect effect through neighboring regions was not significant (p = 0.521). When examining the association of sLHS with IQ, both the direct and indirect effects of sLHS remained significant (p < 0.001).

In summary, our results showed that IND/COL, sLHS, and IQ were strongly positively related, with moderate (for sLHS) to strong (for IQ and IND/COL) regional clustering, as well as notable latitudinal and altitudinal gradients. Furthermore, the relationships among IND/COL, sLHS, and IQ varied across regions: while IND/COL and sLHS were always positively related, the strength of this relationship differed by location. The relationship between IND/COL and IQ was generally positive, but could be slightly negative in some areas. Effects of sLHS were observed both within regions and via neighboring regions; however, these effects disappeared when IQ was included in the analysis. Conversely, there was a strong relationship between sLHS and IQ both within regions and via neighboring regions. Overall, our findings support the interpretation of IQ as a mediator in the relationship between sLHS and IND/COL.

In addition to examining the relationships between variables, we conducted a latent profile analysis using maximum likelihood modeling to classify regions into distinct profiles. A variable class diagonal parameterization model (with variable variance and zero covariance) was fitted to the IND/COL, sLHS, and IQ data, revealing that the model with five profiles provided the best fit. The model’s entropy was 0.933 (maximum = 1), indicating high accuracy in profile differentiation; in other words, all regions were clearly classified into the appropriate profiles with minimal risk of overlap or misclassification. The results of the regional profile groupings are presented in Figure 6.

Figure 5.

Mediation regression model (N = 83). Note: * p < 0.001. Source: Author’s calculations

Figure 6.

Map of Russian regions’ profiles based on the constellation of IND/COL, sLHS and IQ (N = 83). Source: Created by the authors

This study represents a first step toward understanding the role of life history trade-offs in shaping culture and contributes to the analysis of regional differences in IND/COL, both globally and within Russia. Our results show that sLHS, IQ, and IND/COL are positively related, with a strong overall relationship that varies somewhat across regions. Five latent regional profiles with high classification accuracy were identified based on the variables examined. Additionally, geographical gradients were observed: the variables included in the analysis were positively associated with latitude and negatively associated with longitude (except for IND/COL) and altitude.

In studying between-group differences in LHS, countries are most often the primary unit of analysis, leaving regional differences within countries largely unexplored, despite clear foundations for such studies. First, regional differences in LHS are also observed within populations of other species (Ruttenberg et al. 2005). Second, as communities become more complex and integrated, they develop a greater diversity of niches (cultures) (Figueredo et al. 2013). These niches, subject to distinct environmental pressures, have varying capacities to respond to these pressures due to their unique characteristics. This diversity is likely to result in regional differences in group biological adaptations, which are subsequently reflected in cultural characteristics.

Our study contributes significantly to understanding how regional differences in biological trade-offs within a single country relate to cultural variation. In contrast to our approach – which used geographic indicators (latitude, altitude, longitude) exclusively – existing studies on this topic have typically employed the brumal factor, emphasizing the influence of low temperatures on LHS formation. For example, in European countries like Italy and Spain, the brumal factor was positively associated with sLHS, whereas in Mexico, the association was negative (de Baca and Figueredo 2014; Figueredo et al. 2017). In Russia, our findings revealed mixed results: sLHS was positively associated with latitude, consistent with findings at the national level (Figueredo et al. 2017), but negatively associated with altitude.

Previous studies have also shown that higher IQ was characteristic of countries with a sLHS (de Baca and Figueredo 2014; Figueredo et al. 2021). At the regional level, IND was positively associated with long-term orientation, which aligns with sLHS characteristics (Minkov and Kaasa 2021). Additionally, higher IQ was associated with a prevalence of IND (Zengrui et al. 2017). Interestingly, in our results, when IQ was accounted for, the effects of sLHS on IND/COL became non-significant. This phenomenon may be attributable to the notion that high IQ fosters the development of self-control, an important trait for implementing sLHS. Stable sLHS conditions promote high cognitive abilities, which, in turn, creates a foundation for high-quality human capital and societal modernization. Modernization leads to an increase in IND values, as people begin to prioritize personal freedom and move away from rigid group norms. In resource-limited societies, individuals tend to support their group as a means of securing access to resources through power or loyalty. However, under conditions of economic growth, group bias becomes less significant, as it becomes easier to consider the interests of all parties. This fosters a system where laws apply equally to all, reflecting the values of IND (Latova 2016).

On the other hand, the relationship between sLHS and IND aligns with the strategic differentiation-integration hypothesis (Figueredo et al. 2013), which suggests that the speed of LHS affects the degree of interrelationship between its components: with sLHS, traits are less closely interrelated than with fLHS. Under fLHS conditions, where survival depends on constant cooperation to counter external threats, integration increases, making COL norms more adaptive by supporting unity and synchronization of group goals. In contrast, in stable and predictable environments, sLHS fosters social differentiation and the formation of diverse socioecological niches – specialized subgroups with distinct functions. Here, IND norms play a crucial role by supporting autonomy and a diversity of interests, enabling each niche to fulfill its unique role effectively. In such communities, high cognitive abilities and specialized skills become particularly important. The cumulative effect of these factors enhances the need for IND values, as they facilitate the functioning of a society with diverse niches and skill sets, which, in turn, stimulates further development of cognitive abilities.

The latitudinal gradient of IQ and IND aligns with previous findings, which explain this relationship in two ways. First, colder climates pose greater survival challenges, requiring higher cognitive abilities to adapt successfully in such environments (Lynn 2006). Second, a cooler climate with regular precipitation provides greater existential security by reducing risks related to disease and water scarcity (Welzel et al. 2021). This security fosters the development of IND, as people in these conditions can afford to be more autonomous and less distrustful of strangers.

The negative correlation between altitude and IQ mirrors patterns observed at the national level. Interestingly, however, this relationship is not evident in the United States. This may be explained by the fact that the United States is a country of settlers, whose cultural norms and achievements were “imported” from various regions and do not necessarily reflect long-term adaptations to local conditions. This has allowed settlers to cope with environmental challenges without the need for significant transformation. It is important to note that theoretical explanations for these associations remain underdeveloped, as the classical explanation involving the negative impact of chronic hypoxia lacks convincing empirical support (León 2013).

We suggest that at higher altitudes, lower population IQ may be due to geographic and socio-economic factors: lowland regions with extensive river networks and proximity to the sea were more likely to foster urban development. These cities facilitated population concentration, the growth of industry, trade, and education, and attracted diverse populations, thereby stimulating knowledge exchange and innovation. The expansion of shipping, trade, and communication networks in these areas also promoted economic growth, urbanization, and the establishment of educational and scientific institutions, which in turn attracted new ideas, technologies, and cultural values, raising overall IQ levels. Conversely, remoteness from the sea and high altitude may limit access to global networks and resources, potentially reducing urbanization and educational opportunities. This aligns with prior studies that emphasize the impact of social and economic factors on the development of regions at varying altitudes (León and Avilés 2016). Additionally, mountainous terrain has been shown to support authoritarian rule, further affecting regional development patterns (Conway et al. 2017).

In addition, moderate (for sLHS) and strong (for IQ and IND/COL) regional clustering is observed in Russia. This suggests that these variables remain stable within certain spatial distances between regions. The sLHS is more influenced by environmental conditions, which in practice can vary widely from one region to another, while IQ and IND/COL measures are more susceptible to social influences and selective migration. Our results align with patterns found in previous studies (e.g., Grigoriev 2018; see also Hassall and Sherratt 2011). Neighboring regions, on the one hand, often share similar levels of socioeconomic development, while on the other, they experience intertwined historical and demographic processes, such as migration and the spread of cultural norms. Consequently, regions offering attractive conditions for individuals with high IQ and IND values tend to act as gravitational centers, gradually forming larger clusters with similar indicators. For instance, the degree of remoteness of districts of Moscow region from Moscow itself is predictive of a decrease in IQ, further supporting the assortative nature of regions based on intellectual potential (Sugonyaev et al. 2020).

Interestingly, in Russia, relatively low spatial autocorrelation was found for educational achievements based on Unified State Exam results across all specialties at universities within each region. This outcome largely reflects the level of educational infrastructure in each region rather than the intellectual potential of its residents (Grigoriev 2018). In the case of IND, it is essential to consider the role of social influence, as social norms can shape and encourage certain behaviors, leading individuals to internalize these norms and act accordingly (Rentfrow and Jokela 2016). Modern conditions enable people not only to adhere to historically established cultural traditions, but also to actively select or change their place of residence based on personal preferences and lifestyle. This dynamic fosters the formation and reinforcement of clusters with diverse cultural orientations, each adapted to specific environmental characteristics, such as infrastructure, population density, labor market conditions, and other factors (Gallyamova and Grigoryev 2022). Consequently, the clustering of Russian regions largely reflects the patterns of cultural norm dissemination and the assortative processes in cognitive abilities.

Despite the strengths of our study, it has several limitations. First, the cross-sectional design does not allow us to draw conclusions about causality or the direction of the relationships. Although our hypotheses suggest that sLHS contributes to cultural differences, we cannot definitively assess this with the current data. Our statistical model, which assumes full mediation, does not contradict this interpretation but does not strongly support it either. Additionally, we cannot evaluate reciprocal influences – namely, the possibility that IND or IQ may not only influence the slowing of sLHS but may themselves be subject to a reverse effect from sLHS, further reinforcing sLHS. It is possible that interactions among these variables align with the concept of multilevel selection, in which selection pressures generate both biological and social outcomes (de Baca and Figueredo 2014; Figueredo et al. 2017).

In our work, we rely on objective indicators to assess cultures, but it is important for future research to examine how LHS relates to subjective cultural values (Minkov et al. 2023b) as well as to other objective behavioral indicators (Minkov and Kaasa 2021). Additionally, we used only geographic indicators to explore their relationship to LHS. Future studies should expand the set of factors explaining regional differences, as LHS is shaped not only by physical ecology but also by social ecology. Social tensions, historical and political contexts, and periods of economic crisis all play significant roles. For instance, a factor predicting higher fLHS in Poland (Koljević 2024a) and the Czech Republic (Koljević 2024b) was forced displacement after World War II. Given that the development of Russian society in recent decades – and even centuries – has been marked by similar upheavals, it can be assumed that this factor will also hold significant predictive power.

This study is the first in Russia to examine regional differences in LHS. Additionally, we integrated biological explanations into our understanding of cultural formation, identifying the role of IQ as a mediator between biological trade-offs and the social mechanisms that regulate them. The study also makes a significant contribution to the analysis of regional differences in LHS, IQ, and IND/COL measures, as well as the clustering of regions by these measures. Importantly, one practical outcome of this research was the identification of five groups of regions exhibiting similar patterns in both biological and cultural characteristics. These insights may prove useful for incorporating regional features into the development and adjustment of Russian social policy.

The authors express their gratitude to Mikael, a subscriber of the YouTube channel “Zachem my takie,” for his assistance in explaining the relationship between regional IQ and altitude.

The publication was prepared within the framework of the Academic Fund Program at HSE University (grant № 24-00-013 “Adaptive Foundations of Culture: Toward Understanding Cultural Orientations through the Lens of Life History Trade-Offs”).

Albina Gallyamova – Junior Research Fellow at the Center for Sociocultural Research. HSE University, 20 Myasnitskaya St., Moscow, 101000, Russia. E-mail: aagallyamova@hse.ru

Elizaveta Komyaginskaya – Research Intern at the Center for Sociocultural Research. HSE University, 20 Myasnitskaya St., Moscow, 101000, Russia. E-mail: ekomyaginskaya@hse.ru (corresponding author)

Anna Vasyunina – Research Intern at the International Laboratory for Population and Health Research. HSE University, 20 Myasnitskaya St., Moscow, 101000, Russia. E-mail: avasyunina@hse.ru

Dmitry Grigoryev – Research Fellow at the Center for Sociocultural Research. HSE University, 20 Myasnitskaya St., Moscow, 101000, Russia. E-mail: dgrigoryev@hse.ru