A Gene for Harsh Parenting?
Do genetics and economics tell us why some people abuse their kids?
Posted Aug 07, 2013
Why do so many researchers keep trying to make human psychology simple and gene-based? A recent headline proclaims: “Researchers say a specific gene might make parents more likely to abuse their children during tough times.” The argument is that during bad economic times, or more specifically the perceived threat of bad times, mothers with a certain copy of the dopamine receptor gene DRD2 Taq1A are more likely to be harsh parents (spanking, slapping, shouting and threatening their children). These ideas come from a recent study entitled “The Great Recession, genetic sensitivity, and maternal harsh parenting” about to be published in the Proceedings of the National Academy of Sciences of the United States of America (PNAS). Understandably, this study and specualtion about its implications is burning up the Internet and radiowaves.
Can a bad dopamine gene be a core cause for Moms being harsher parents? Not likely. There is no way the world, the human body, and parenting behavior is that simple.
The basic data analyzed in the article comes from the very large Families and Child Wellbeing study (based in Princeton and nearly 15 years of data on ~5000 kids born between 1998-2000). The range of information in the larger study gives some really important insights into patterns and complexities in the structure and function of families termed “fragile” (at greater risk of breaking up and living in poverty). But if you take the data and add in some overly simplistic assertions about society, behavior, and biology you get the same old garbage: my genes made me do it.
The core premise in the article is that we can actually understand complex behavior via a gene by environment model (GxE) wherein looking at variations in the versions of a gene and correlating it with some sort of outcome via variation in environmental factors can tell us about causal relationships between the gene and the outcomes. While this works (more or less) in a the few cases of genetic sequences tied to certain outcomes like leaf shape or some disease function…it is not effective in complex behaviors and complex genetic systems (like humans and dopamine pathways). Basically GxE approaches for behavior give us overly simplistic, and horribly incomplete, answers.
The study in question compares variation in “maternal harsh parenting,” “city-level unemployment rates,” and the “national consumer sentiment index” and a few forms of the DRD2 gene. The authors state that “All models controlled for age, race/ethnicity, immigration status, educational attainment, poverty status, family structure, child sex, and child age (in months) at the time of interview.” Really? All that is controllable, and comparable, in one fell swoop? The actual graph they use has “harsh parenting” as one axis and “macroeconomic conditions” on the other.
To be fair, one version of the DRD2D gene (the Taq1A version) showed some significant patterns relative to the correlational analyses and other versions did not…and the authors do offer a few cautions in their discussion. But they have no physiological or neurological data, and only very generalized economic and behavioral data and still assert that “that mothers’ responses to adverse economic conditions are moderated by the DRD2 Taq1A genotype.”
Correlation, even a significant one, does not imply causation or any particular directional physiological connection. Can you really talk about the underlying processes of harsh parenting (or any parenting) without in-depth psychological and anthropological understandings of the mothers and their social contexts? Not to mention the problems with simplistic measures of economic conditions.
Do we want to sit back and accept, yet again, another simple explanation for complex human behavior? NO, we do not. Genes matter, but so do many, many, many other variables. Research that artificially tidies up human reality benefits no one. Being human is complex and messy and anyone who tells you otherwise is lying.