Explore the vast range of titles available.

Social science research argues that differences in individuals’ literary and cultural tastes originate in social environments. Yet, it might be that these differences are partly associated with genetic differences between individuals. To address this possibility, we use nation-scale registry data on library borrowing among Danish twins ( N = 67,900) to assess the heritability of literary tastes. We measure literary tastes via borrowing of books of different genres (e.g., crime and biographical novels) and formats (physical, digital, and audio) and decompose the total variance in literary tastes into components attributable to shared genes (heritability), shared environments (social environment shared by siblings), and unique environments (social environments not shared by siblings). We find that genetic differences account for 45–70 percent of the total variance in literary tastes, shared environments account for almost none of the variance, and unique environments account for a moderate share. These results suggest that literary tastes are approximately as heritable as other human phenotypes (e.g., physical traits, cognition, and health). Moreover, heritability is higher for socioeconomically disadvantaged groups than for advantaged groups. Overall, our results suggest that research should consider the role of genetic differences in accounting for individual differences in literary and broader cultural tastes.

Theories in sociology argue that family background and individual experiences shape cultural tastes and participation. Yet, we do not know the relative importance of each explanation or the extent to which family background operates via shared genes or shared environments. In this article, we use new data on same-sex monozygotic and dizygotic twins from Denmark to estimate the total impact of family background (genetic and environmental) and individual experiences on highbrow and lowbrow tastes and participation and on omnivorousness in music and reading. We find that family background explains more than half of the total variance in cultural tastes and participation and in omnivorousness. Moreover, family background operates mainly via shared genes, with shared environments shaping cultural tastes to some extent, but not cultural participation. Our findings support theories claiming that family background is instrumental in shaping cultural tastes and participation but highlight the relevance of distinguishing genetic and environmental aspects of family background.

Theories in sociology argue that family background and individual experiences shape cultural tastes and participation. Yet, we do not know the relative importance of each explanation or the extent to which family background operates via shared genes or shared environments. In this article, we use new data on same-sex monozygotic and dizygotic twins from Denmark to estimate the total impact of family background (genetic and environmental) and individual experiences on highbrow and lowbrow tastes and participation and on omnivorousness in music and reading. We find that family background explains more than half of the total variance in cultural tastes and participation and in omnivorousness. Moreover, family background operates mainly via shared genes, with shared environments shaping cultural tastes to some extent, but not cultural participation. Our findings support theories claiming that family background is instrumental in shaping cultural tastes and participation but highlight the relevance of distinguishing genetic and environmental aspects of family background.

An article recently published in Sociological Science explored cultural tastes and practices in Denmark using a behavioral genetic lens. Using data from monozygotic and dizygotic twins, the authors concluded that shared familial environments mattered less than genetic factors, thus questioning the soundness of established sociological theorizations of cultural inequalities, including Bourdieu's Distinction. After a close methodological and conceptual examination, Lebaron et al's main conclusion is that social scientists should lend little credence to the claims put forth by Jæger and Møllegaard, which fall short of the methodological and conceptual standards habitually upheld in both sociology and behavior genetics. This is true whether they selectively focus on the cultural aspects while ignoring the genetic dimension, or vice versa. Logically, their combination only brings additional uncertainties.

Machine learning (ML) approaches show increasing promise in their ability to identify vocal markers of autism. Nonetheless, it is unclear to what extent such markers generalize to new speech samples collected e.g., using a different speech task or in a different language. In this paper, we systematically assess the generalizability of ML findings across a variety of contexts. We train promising published ML models of vocal markers of autism on novel cross-linguistic datasets following a rigorous pipeline to minimize overfitting, including cross-validated training and ensemble models. We test the generalizability of the models by testing them on i) different participants from the same study, performing the same task; ii) the same participants, performing a different (but similar) task; iii) a different study with participants speaking a different language, performing the same type of task. While model performance is similar to previously published findings when trained and tested on data from the same study (out-of-sample performance), there is considerable variance between studies. Crucially, the models do not generalize well to different, though similar, tasks and not at all to new languages. The ML pipeline is openly shared. Generalizability of ML models of vocal markers of autism is an issue. We outline three recommendations for strategies researchers could take to be more explicit about generalizability and improve it in future studies. Machine learning approaches promise to be able to identify autism from voice only. These models underestimate how diverse the contexts in which we speak are, how diverse the languages used are and how diverse autistic voices are. Machine learning approaches need to be more careful in defining their limits and generalizability.