Vertical cultural transfer effects — plausible but mostly not real

Since this topic comes up repeatedly and I have to find the same studies over and over, here’s a little review.

Vertical cultural transfer (various other names: vertical cultural inheritance, cultural transmission) is the main effect proposed by various social scientists for the similarity of parents and kids. It differs from the shared environment concept. Shared environment is really a combination of several environmental effects that all affect children growing up in the same family. These can be 1) vertical downstream (parents to kids), 2) vertical upstream (kids to parents), and 3) various horizontal (siblings to each other, or older sibs to younger sibs). So when you find a shared environment effect in some classical twin study, pause and think about whether it might more plausibly reflect (2) or (3) rather than necessarily (1).

The classical twin design — the one that compares MZs and DZs both reared together — cannot be used to study vertical transfer effects. However, various extended family designs can, most popularly the children of twins/extended twin design. Here’s 4 studies that have examined vertical transfer effects with these designs.

Martin et al 1986

  • Martin, N. G., Eaves, L. J., Heath, A. C., Jardine, R., Feingold, L. M., & Eysenck, H. J. (1986). Transmission of social attitudes. Proceedings of the National Academy of Sciences, 83(12), 4364-4368.

Data gathered in Australia and England on the social attitudes of spouses and twins are largely consistent with a genetic model for family resemblance in social attitudes. There is substantial assortative mating and little evidence of vertical cultural inheritance.

note: h = additive genetic, b = vertical cultural, u = assortative mating, apostrophe = female (when sexes are split)

The authors’ short discussion is worth quoting:

The problem with many “social” explanations of our data is that they do not lead to predictions about other kinds of relationship unless social interaction is based ultimately on genetic differences (e.g., see ref. 8). Our model can be used to predict the results of other studies. For example, we predict a zero correlation between foster parent and adult foster child for all our attitude scales. Our model (Table 3, model 3) predicts a parent-offspring correlation of ½2h2 (1 + A) = 0,52 for conservatism. We predict correlations of V4h2 (1 + h2 U)2 = 0.31 for the offspring of monozygotic twins and h = 0.62 for separated monozygotic twins. If our model of mate selection is right, we predict that the spouses of siblings should show a correlation of ½2h2,u2 (1 + h2A) = 0.20. The correlation between the spouse of one monozygotic twin and the co-twin is expected not to differ significantly from h2A = 0.42. All these data are obtainable and can yield further tests of our model.

If our model withstands the further tests we propose, it will have radical implications for our understanding of cultural inheritance in humans and undermine the naive assumption that the resemblance of family members can be interpreted in purely social terms. We began with psychometric instruments, which might have been expected to maximize our chances of detecting nongenetic transmission. We fitted a model that specifies both social and genetic components of vertical transmission and we have obtained estimates of the cultural parameter that do not differ significantly from zero in many cases. One interpretation of our finding is that our method and model are fundamentally wrong. If this is the case, then our predictions will be falsified by the data still to be gathered. The alternative possibility is that geneticists and social scientists have misconceived the role of cultural inheritance and that individuals acquire little from their social environment that is incompatible with their genotype. In no way does our model minimize the role of learning and social interaction in behavioral development. Rather, it sees humans as exploring organisms whose innate abilities and predispositions help them select what is relevant and adaptive from the range of opportunities and stimuli presented by the environment. The effects of mobility and learning, there- fore, augment rather than eradicate the effects of the genotype on behavior.

Eaves et al 1999

Measures of four dimensions of personality (Psychoticism, Extraversion, Neuroticism, and Lie scores) and six aspects of social attitudes (to sex, taxation, militarism, politics, religion and a general conservatism scale) were obtained by mailed questionnaire from 29 691 US subjects including adult twins (n = 14 761) their parents (n = 2360), their spouses (n = 4391), siblings (n = 3184) and adult children (n = 4800). After correction for the average effects of age, sex and source of sample, familial correlations were computed for 80 distinct biological and social relationships. The data allow for the estimation of the additive and non-additive effects of genes, assortative mating, vertical cultural inheritance and other non-parental effects of the shared environment on differences in personality and social attitudes. The interaction of genetic and environmental effects with sex may also be analyzed. Model-fitting analyses show that personality and social attitude measures differ markedly in major features of family resemblance. Additive and dominant genetic effects contribute to differences in both personality and attitudes, but the effects of the family environment, including vertical cultural transmission from parent to child, are much more marked for social attitudes than for personality. There is substantial assortative mating for social attitudes and almost none for personality. The causes of family resemblance depend significantly on sex for almost every variable studied. These findings clarify and extend the more tentative findings derived from previous twin, family and adoption studies.

The authors don’t emphasize the striking finding, but compare the values for total parental to total genetic and you see that total parental generally near zero while genetic is not. The only thing that’s somewhat notably shared environmental is political, which just means party voting. Since people vote somewhat randomly, and singular votes don’t really matter much, this finding is not that interesting. The conservatism scale, formed from the other others, shows no noteworthy parental effects as usual.

Wadsworth et al 2002

Background: The study aimed to conduct the first analysis of CAP parent–offspring resemblance for reading performance in children aged 7, 12 and 16 years, and to assess the etiology of individual differences in reading performance of children at 16 years of age. Method: The Reading Recognition subtest of the Peabody Individual Achievement Test was administered to children in the Colorado Adoption Project (CAP) at 7, 12 and 16 years of age, and to their adoptive and nonadoptive parents when the children were 7 years of age. Results: Resulting parent–offspring correlations in adoptive families were not significant at any age, but correlations between scores of nonadoptive control parents and their offspring were significant at all three ages. Conclusions: Results obtained from behavioral genetic model‐fitting analyses of data from parents and their children tested at age 16 are consistent with results of studies of twins and siblings indicating that individual differences in reading performance are due substantially to genetic influences. In contrast, environmental transmission from parents to offspring was negligible, suggesting that environmental influences on individual differences in the reading performance of children are largely independent of parental reading performance.

Note: extended family design based on adoptives.

And from conclusion:

In contrast, cultural transmission based on parental reading performance (m and f) appears to have little effect on individual differences in the children’s reading performance. This may seem surprising given the emphasis generally placed on reading to children. However, this does not imply that reading to one’s children does not influence the development of reading ability, or that parents have no effect on their children’s reading performance. Rather, its imply means that individual differences among parents’ reading performance, or factors correlated with reading performance of the parents, do not directly influence individual differences in reading performance among their children; this does not preclude the possibility of transmissible effects other than parental reading performance or its correlates.

Leeuwen et al 2008

In this paper we assess the presence of assortative mating, gene–environment interaction and the heritability of intelligence in childhood using a twin family design with twins, their siblings and parents from 112 families. We evaluate two competing hypotheses about the cause of assortative mating in intelligence: social homogamy and phenotypic assortment, and their implications for the heritability estimate of intelligence. The Raven Progressive Matrices test was used to assess general intelligence (IQ) and a persons IQ was estimated using a Rasch model. There was a substantial correlation between spouses for IQ (r = .33) and resemblance in identical twins was higher than in first-degree relatives (parents and offspring, fraternal twins and siblings). A model assuming phenotypic assortment fitted the data better than a model assuming social homogamy. The main influence on IQ variation was genetic. Controlled for scale unreliability, additive genetic effects accounted for 67% of the population variance. There was no evidence for cultural transmission between generations. The results suggested that an additional 9% of observed IQ test variation was due to gene–environment interaction, with environment being more important in children with a genetic predisposition for low intelligence.

Authors are explicit:

The present study design is not suited to uncover GE correlations other than one resulting from simultaneous genetic and cultural transmission. But what we can conclude is that if there is GE correlation, the role of parents seems limited to responding to the needs and interests as indicated by the child. We found no indication that intelligent parents provide their offspring with intelligence promoting circumstances. More likely, children with a genetic predisposition for either a low or a high IQ ask for a specific type of stimulation. In other words, an evocative gene – environment correlation (where individuals are reacted to on the basis of their genetically influenced phenotype) or an active GE correlation (where individuals seek or create environments correlated with their genetic inclinations) seems a more probable mechanism than a passive GE correlation ( Scarr & McCartney, 1983 ). Only the last type of correlation could in principle have been detected by our extended twin family design.

I also note that they found a GxE in their data (p val not that convincing), and cite Turkheimer (2003). Regarding that, see this post.

Hatemi et al 2010

Variance components estimates of political and social attitudes suggest a substantial level of genetic influence, but the results have been challenged because they rely on data from twins only. In this analysis, we include responses from parents and nontwin full siblings of twins, account for measurement error by using a panel design, and estimate genetic and environmental variance by maximum‐likelihood structural equation modeling. By doing so, we address the central concerns of critics, including that the twin‐only design offers no verification of either the equal environments or random mating assumptions. Moving beyond the twin‐only design leads to the conclusion that for most political and social attitudes, genetic influences account for an even greater proportion of individual differences than reported by studies using more limited data and more elementary estimation techniques. These findings make it increasingly difficult to deny that—however indirectly—genetics plays a role in the formation of political and social attitudes.

Again, vertical transfer was mostly indistinguishable from 0, except for voting behavior.

Boosma et al 2010


In contrast to the large number of studies in children, there is little information on the contribution of genetic factors to Attention Deficit Hyperactivity Disorder (ADHD) in adults.

To estimate the heritability of ADHD in adults as assessed by the ADHD index scored from the CAARS (Conners’ Adult ADHD Rating Scales).

Phenotype data from over 12,000 adults (twins, siblings and parents) registered with the Netherlands Twin Register were analyzed using genetic structural equation modeling.
Main outcome measures

Heritability estimates for ADHD from the twin-family study.

Heritability of ADHD in adults is estimated around 30% in men and women. There is some evidence for assortative mating. All familial transmission is explained by genetic inheritance, there is no support for the hypothesis that cultural transmission from parents to offspring is important.

Heritability for ADHD features in adults is present, but is substantially lower than it is in children.

Large study, but used crappy self-report measure, so low reliability and biased measure gives low heritability. This was shown in a recent study, Faraone and Larsson 2019. Aside from that, little evidence for vertical transfer as usual.

Vinkhuyzen et al 2012

Heritability estimates of general intelligence in adulthood generally range from 75 to 85%, with all heritability due to additive genetic influences, while genetic dominance and shared environmental factors are absent, or too small to be detected. These estimates are derived from studies based on the classical twin design and are based on the assumption of random mating. Yet, considerable positive assortative mating has been reported for general intelligence. Unmodeled assortative mating may lead to biased estimates of the relative magnitude of genetic and environmental factors. To investigate the effects of assortative mating on the estimates of the variance components of intelligence, we employed an extended twin-family design. Psychometric IQ data were available for adult monozygotic and dizygotic twins, their siblings, the partners of the twins and siblings, and either the parents or the adult offspring of the twins and siblings (N = 1314). Two underlying processes of assortment were considered: phenotypic assortment and social homogamy. The phenotypic assortment model was slightly preferred over the social homogamy model, suggesting that assortment for intelligence is mostly due to a selection of mates on similarity in intelligence. Under the preferred phenotypic assortment model, the variance of intelligence in adulthood was not only due to non-shared environmental (18%) and additive genetic factors (44%) but also to non-additive genetic factors (27%) and phenotypic assortment (11%).This non-additive nature of genetic influences on intelligence needs to be accommodated in future GWAS studies for intelligence.

We look for models PA-3 and PA-7 because:

Summarizing the results for both reduced models: model PA-3 included additive genetic factors (44%), genetic dominance (27%), phenotypic assortment (11%) and non-shared environmental factors (18%). Model PA-7 included additive genetic factors (58%), phenotypic assortment (23%), negative CT (8%), and non-shared environmental factors (11%); correlation between A and CT was −0.36. Comparing both reduced PA models (model PA-3 and model PA-7) showed no significant difference in likelihood, but showed lower AIC, BIC, and DIC for model PA-3. Based on these criteria, model PA-3 was to be preferred.

Thus, vertical transfer (CT) was either modeled as negative or zero, with the preferred model being zero.

Kandler et al 2012

In this study, we used an extended twin family design to investigate the influences of genetic and cultural transmission as well as different sources of nonrandom mating on 2 core aspects of political orientation: acceptance of inequality and rejecting system change. In addition, we studied the sources of phenotypic links between Big Five personality traits and political beliefs using self- and other reports. Data of 1,992 individuals (224 monozygotic and 166 dizygotic twin pairs, 92 unmatched twins, 530 spouses of twins, 268 fathers, and 322 mothers) were analyzed. Genetically informative analyses showed that political attitudes are genetically but not environmentally transmitted from parents to offspring and that a substantial proportion of this genetic variance can be accounted for by genetic variance in personality traits. Beyond genetic effects and genotypic assortative mating, generation-specific environmental sources act to increase twins’ and spouses’ resemblance in political beliefs. The results suggest multiple sources of political orientations in a modern democracy.

Kandler et al 2016

Despite cross-cultural universality of core human values, individuals differ substantially in value priorities, whereas family members show similar priorities to some degree. The latter has often been attributed to intrafamilial socialization. The analysis of self-ratings on eight core values from 399 twin pairs (ages 7-11) and their biological parents (388 mothers, 249 fathers; ages 26-65) allowed the disentanglement of environmental from genetic transmission accounting for family resemblance in value orientations. Results indicated that parent-child similarity is primarily due to shared genetic makeup. The primary source of variance in value priorities represented environmental influences that are not shared by family members. These findings do not provide evidence for parental influences beyond genetic influences contributing to intrafamilial similarity in value priorities.

So, medium-sized dataset. The main results table:

The m² + f² + 2mfµ column is the environmental impact of parents on kids, h² is genetics, c² is sibling-sibling effects (not the same meaning of c² as in classical twin design!), and e² is the residual, which can be taken as some mix of measurement error, biological randomness, and individualized environmental causes. The values in parentheses are adjusted for measurement error. So the story is basically that similarity is mostly due to genetics, and for specific values sometimes siblings copy each other. Variation at large though is dominated by random causes or individualized environmental causes.

Lyngstad et al 2017

Research on the intergenerational transmission of educational attainment, and transmission of social positions more generally, have long attempted to separate the relative roles of ascription and achievement. In these efforts, the bulk of research has ignored genetic inheritance. We use structural equations models and data on 4590 twin pairs and their parents to distinguish the roles of genetic and environmental influences on educational attainment in Norway, a country with high affordability and easy access to education at all levels. Our quantitative genetic models confirm the status quo; not of sociology, but of behavior genetics. Heritable factors play an important role in the transmission process, and the postulated direct effects of parents own educational attainments are negligible. The family environment does matter, but only those features that are shared between the twins themselves and not those that involve their parents. These results represent a challenge to conventional sociological theory on intergenerational transmission processes and the role of education in social stratification.

In table 4 , we show the complete results for the AFSE model including the variance component estimates for each component. The AFSE model allow for both direct cultural transmission as well as influences from the sibling environment, in addition to genetic and unique environmental influences. The most influential variance component is the additive genetic influences (A), which is estimated to contribute more than half of the variation in educational attainment. The second most influential factor is the unique environment. The sibling environment contributes about 12% of the total variation. The family environment does not contribute much variation at all to the phenotype.

Evaluation of hypotheses Our third hypothesis, which concerned the role of direct cultural transmission from parents to children, is not supported in our preferred model, based on the NTFD design. We are able to estimate m , the pathway for direct cultural transmission from parents to children in the AFSE model. This path represents the influence of the parent-driven home environment. Based on the extant literature, we expected this path coefficient to be positive and measurably different from zero. The path coefficient for m is estimated at -0.0733. This indicates that any direct cultural transmission from parents to children seems to be of miniscule magnitude and in fact negative. Its confidence interval ranges from -0.1313 to -0.0129, suggesting a small, negative effect of having highly educated parents on one’s own education.

Swagerman et al 2017

Reading is the processing of written language. Family resemblance for reading (dis)ability might be due to transmission of a genetic liability or due to family environment, including cultural transmission from parents to offspring. Familial-risk studies exploring neurobehavioral precursors for dyslexia and twin studies can only speak to some of these issues, but a combined twin-family study can resolve the nature of the transmitted risk. Word-reading fluency scores of 1100 participants from 431 families (with twins, siblings and their parents) were analyzed to estimate genetic and environmental sources of variance, and to test the presence of assortative mating and cultural transmission. Results show that variation in reading ability is mainly caused by additive and non-additive genetic factors (64%). The substantial assortative mating (r father–mother = 0.38) has scientific and clinical implications. We conclude that parents and offspring tend to resemble each other for genetic reasons, and not due to cultural transmission.

And conclusion:

In this study we aimed to test if the family resemblance which has been reported for reading ability and disability is caused by genetic or cultural transmission. To our knowledge, we were the first to explore this using a sample including twins, their parents and siblings. Secondly, we aimed to test if assortative mating is present. We found that individual differences in reading ability were mainly caused by genetic factors, both additive and non-additive. Environmental factors that are shared between parents and children did not contribute to familial resemblance and no evidence was found for cultural transmission from parents to their offspring. In the remainder we will start with limitations, followed by discussion of modeling findings and their scientific and clinical implications.

Kornadt et al 2018

Political participation (POP), social participation (SOP), and political interest (PI) are important indicators of social status and social inequality. Previous studies on related trait differences yielded genetic and environmental contributions. However, focusing on adult samples, classical twin designs, and convenience samples often restricts parameter estimation and generalizability, and limits the understanding of age differences. We investigated sources of variance in POP, SOP, and PI in late adolescence and early adulthood with an extended twin family design (ETFD). We analyzed data from over 2,000 representative German twin families. Individual environments not shared by family members reflected the major source of variance for all variables, but genetic influences were also pronounced. Genetic effects were mostly higher for young adults, whereas effects of twins’ shared environment were significant in adolescence. Our study deepens the understanding of the interplay between genetic and environmental factors in shaping differences in young persons’ integration in society.

Main table:

This is from their best fitting model. The second complex formula gives the vertical downstream transfer, which ranges from 0 to 8%, in line with the findings from the other studies.

Bell et al 2018

A new paradigm has emerged in which both genetic and environmental factors are cited as possible influences on sociopolitical attitudes. Despite the increasing acceptance of this paradigm, several aspects of the approach remain underdeveloped. Specifically, limitations arise from a reliance on a twins-only design, and all previous studies have used self-reports only. There are also questions about the extent to which existing findings generalize cross-culturally. To address those issues, this study examined individual differences in liberalism/conservatism in a German sample that included twins, their parents, and their spouses and incorporated both self- and peer reports. The self-report findings from this extended twin family design were largely consistent with previous research that used that rater perspective, but they provided higher estimates of heritability, shared parental environmental influences, assortative mating, and genotype-environment correlation than the results from peer reports. The implications of these findings for the measurement and understanding of sociopolitical attitudes are explored.

This is the same sample as above (total n = 2000), which the authors apparently decided to split into multiple publications on nearly identical topics. Main table:

As the authors note:

The estimate of the environmental influence provided solely by mothers (m) was small and nonsignificant, as was that provided by fathers (f). The variance explained by mothers’ and fathers’ joint environmental effects (m2+f2+2m fμ+2s2(m+f)) was very small: 0.9% (1.2% corrected). This suggests that parental influences on individual differences in liberal-ism/conservatism were primarily genetically mediated rather than purely socially transmitted, a finding consistent with previous work on Americans.

The strangest thing about this study was the very low heritability results for the peer report values, despite what others have found. Unfortunately, they did not combine self and peer report data for this study.

Hufer et al 2019

Political orientation is often assumed to be shaped by socialization processes; however, previous studies have shown substantial genetic variance components in party affiliation, political attitudes and behaviors, or closely related personality traits. The majority of these studies have relied on the Classical Twin Design, which comes with restrictive assumptions, some of which are easily violated. Moreover, most analyses lack a perspective of age-group differences. In this study, we investigated political orientation in adolescents (age: 16-18) and young adults (age: 21-25) in a cross-sectional Nuclear Twin Family Design. We used data of the German TwinLife project, including data from same-sex twins reared together, their biological parents, and nontwin full siblings. We found genetic variation in political orientation, which was significant in the older cohort, possibly indicating an increasing importance of active gene-environment correlation from adolescence to adulthood. Individual differences in political orientation because of passive gene-environment correlation and shared environmental effects were larger in the younger cohort, substantiating the same theoretical consideration and the importance of shared socialization contexts for adolescents’ political views. By running Nuclear Twin Family model analyses, and considering age-group differences, as well as the relationship of political orientation with the Big Five personality traits, our study extended previous work, and resulted in more robust and fine-grained estimates of genetic and environmental sources of variance in political orientation. Therefore, it contributed to a better understanding of the complex nature-nurture interplay that forms political orientation in emerging adulthood. (PsycINFO Database Record (c) 2019 APA, all rights reserved).

Final sample size was ~4200 persons. The main table of results is:

The component for parental transmission is x, which is 22% for age 17 and 0% for age 23. As authors name in text:

As expected, the additive genetic variance component (a2q) was larger in C23, and in contrast to C17, the estimate was significantly different from zero. The passive gene-environment correlation (w) and its contribution to the variance (2aw), as well as the sibling-specific shared environmental component (s2), tended to be higher in the younger cohort. However, CIs largely overlapped, and w was not statistically significant in either cohort, whereas s was significant in both cohorts. The variance component because of direct parental transmission (x) was, as expected, substantially larger in C17 (21.5 vs. 0%). A closer look at the specific parameters yielded that the latter was because of a substantial maternal effect in C17, whereas paternal transmission was not statistically significant in both cohorts. Finally, the twin-specific shared environmental component was not statistically significant in both cohorts, which was contrary to our expectations. The nonshared environmental component, including random error variance, also tended to be larger in the younger cohort. In C17, genetic effects accounted for 12.4% of the variance inpolitical orientation, whereas shared environmental components accounted for 38.0% and the passive gene-environment correlationfor 16.6%. In C23, 61.3% of the variance was accounted for by genetic effects, 17.8% by shared environmental effects, and 0% by the passive gene-environment correlation. Thus, our main hypotheses were supported in the sense that the contribution of the passive gene-environment correlation was present in C17, and that the genetic variance was substantially larger in C23. This pattern of results was consistent with the hypothesis of a shift from passive to active or reactive gene-environment correlation with advancing age (Scarr & McCartney, 1983).

Thus, the results are pretty much as usual. Political interests are sort of copied from parents when living at home, but falls off quickly when people get into adulthood.

There we have it, 14 studies spanning 33 years reach quite similar conclusions. Should be noted though that samples were (partially) shared between some studies but examined with different methods.

Leave a Reply