Addressing genetic essentialism: Sharpening context in behavior genetics

Abstract

Evidence of a causal role for genes in human behavior underpins genetic essentialism, the scientifically flawed and socially hazardous idea that heritable characteristics are immutable. Behavior geneticists can challenge this idea by designing research that brings the contextual dependence of heritability estimates into sharper focus, and by incorporating a relevant statement into research reports and public outreach.

In their well-argued paper, Madole & Harden take aim at concerns of some scientists that genome-wide association study (GWAS) research will reinforce the idea that inequalities in society are fixed and proper.

They declare that “such a picture…is unwarranted” on the grounds that “genetic causes for human behavioral traits are non-uniform, non-unitary and non-explanatory” (target article, sect. 3.3, para. 4). We agree that the concerns have no scientific bases, but consider that the concerns, and more generally worries about genetic essentialism (Dar-Nimrod & Heine, 2011), are grounded in reality. In this commentary, we explore that reality and suggest ways in which behavioral genetics might address its influence on public discourse and policy.

Genetic essentialism is a family of beliefs centered on the ideas that if genes are known to play a role in shaping human differences then those differences are, to a considerable extent, natural and entrenched. The differences cover not only society-wide structures that license privilege, so-called social Darwinism, but ones expressed in individuals, often enough focused on less favorable outcomes in health and behavior, which are also seen as hard to alter. Admittedly, this pessimism can go hand-in-hand with a sense of relief in that a genetically influenced difficulty can be seen as no-one's “fault.” This is the mixed-blessing model developed by Haslam and Kvaale (2015). But genetic essentialism can, and does, undermine the acceptance by individuals of efficacious interventions and hamper their implementation by professionals (Haslam & Kvaale, 2015; Lebowitz, Ahn, & Nolen-Hoeksema, 2013).

We propose two ways that behavior-genetic research community could contribute to combatting genetic essentialism, whose adverse effects are now well documented (Dar-Nimrod & Heine, 2011). One is in the design of the research agenda, the other in public communication of research results. Regarding research: The aphorism that “genes are not destiny” can be, and if possible should be, brought into clear focus by demonstrating its truth within the study of single phenotype. There exist useful demonstrations of different patterns of heritability across groups defined by location (e.g., lower heritability of early reading in Scandinavia compared to the United States and Australia, Samuelsson et al., 2008) and across groups defined by time (e.g., an increase in heritability of educational attainment in Norway for males born after 1940, Heath et al., 1985). In addition, demonstrations of gene–environment interplay, where it exists (see Grasby, Coventry, Byrne, & Olson, 2019), underline the dependence of some heritability estimates on environmental factors such as socioeconomic level. And in an additional and informative level of complexity, illustrations of different cross-national patterns of gene–environment interaction can deepen insights into environmental influences on phenotypes (Grasby et al., 2019; Tucker-Drob & Bates, 2016).

Researchers can, and should, also maximize opportunities afforded by analytic maneuvers to expose contextual influences on heritability estimates. For example, in a study using the classical twin design when samples are mixed in terms of location, time, or participant type, the choice to standardize scores within groups can generate higher estimates of genetic influence than the choice of not to standardize. In our cross-national study of early reading (Byrne, Olson, & Samuelsson, 2013), Scandinavian children in grade 1 read at a lower average level than those in the United States and Australia, because of the later start of reading instruction in Sweden and Norway. Employing within-country standardization, the genetic influence on reading comprehension was estimated at 0.62; in the absence of within-country standardization, the estimate dropped to 0.38 with a corresponding rise in the shared environment effect. As a thought experiment, imagine a whole-world twin study without country standardization; doubtless the biggest influence on variation in reading ability at any particular age will not be genes but curriculum policies and the resources available for education, showing up in analyses as shared environment. Such a picture would reinforce the value of projects like that of Lyytinen and colleagues (Ojanen et al., 2015), aimed at bringing the benefits of basic reading research to educational practice in less developed nations. Unthinking acceptance of the conclusion, derived from twin studies conducted largely in the United States, Europe, and Australia, that half or more of the variance in reading skill is heritable would tend to undermine efforts like Lyytinen's. Thus, where it is appropriate to privilege a humans in general perspective over humans within particular contexts, analytic choices can be made to do exactly that.

Thus, in designing behavior-genetic research, features that enhance the visibility of context on heritability estimates should be front and center. International cooperation would be one way to promote this goal, in cases at least where there might be reasons to believe that country differences could influence levels and expression of a phenotype. Geneticists working on devising polygenic risk scores for human conditions, especially diseases, may be ahead of behavior geneticists in that there are calls to increase the ancestral diversity of participant groups (Surigo, Williams, & Tishkoff, 2019) in response to (a) the predominance of populations with European ancestry in studies to date, and (b) accumulating evidence that scores derived from European samples do not always hold up in predicting disease status in African, Asian, Latino, and other ancestry groups (Belsky et al., 2013; Grinde et al., 2019). Behavior genetics would probably replace ancestral groups with sociocultural ones on the grounds that it is those groups that afford the most likely source of contextual influence on patterns of heritability, not ancestry-driven differences in linkage disequilibrium, and in the other sources of variability in polygenic risk score analyses (Surigo et al., 2019).

Armed with a substantial corpus of data on contextual dependence of heritability estimates, behavior genetics could develop a core statement to allay the concerns of “end-users” such as educators, clinical psychologists, criminologists, and lawyers that genes are indeed destiny. A group such as the Behavior Genetics Association could be tasked with crafting such a statement, which could then be incorporated, suitably modified for particular circumstances, into research publications and, more generally, into public outreach campaigns.