The “Hook-Profet embryo-protection hypothesis” has formed the analytical scaffolding for a slew of scholarship. Biologists Paul Sherman and Samuel Flaxman of Cornell University have rounded up the evidence in several review articles. For example, questionnaire studies reveal that women who do experience nausea and vomiting report an onset around the fifth week; queasiness then peaks between the sixth and 12th weeks, and is rare after the 18th week. These patterns overlap with critical periods in which embryonic cells divide rapidly and differentiate.
When the authors asked which foods the women found off-putting, the most disgusting things were meat (that lair of microorganisms) followed by vegetables, coffee, tea (all of which are usually laced with secondary plant compounds) and alcohol (a teratogen).
The most important question for the theory, of those reviewed by Sherman and Flax, is the extent to which pregnancy sickness equates to better pregnancy outcomes. The results were mixed, and for that reason some scholars question the veracity of the Hook-Profet hypothesis. After performing an impressive meta-analysis using data from tens of thousands of pregnancies, the authors found that women who experienced nausea and vomiting were, just as Irving noticed back in 1940, significantly less likely to miscarry compared to women who had more comfortable pregnancies. On the other hand, there was no clear relationship between morning sickness and birth defects, which, at least on the surface, poses problems for the protection model. In fact, although the causality is unclear, women who were prescribed antihistamines (which happen to have an anti-nausea effect) for varying reasons during their first trimesters were statistically less likely to have children born with birth defects. (The use of thalidomide in the 1950s to suppress pregnancy sickness resulted in a notorious teratogenic catastrophe, however.) Given that these medicated women did not have the proposed protective benefits of morning sickness, and yet had healthy babies still—and perhaps even healthier babies than their nauseated cohorts—these findings fail to support the Hook-Profet model. In such cases, however, it’s often difficult to rule out other confounds that often go along with prescription drugs, such as maternal socioeconomic status and related access to prenatal health care, which may also underlie this difference in infant health.
There’s another question that cannot be answered comfortably by the protection model, and that is simply why, if it is indeed an evolutionary adaptation, does pregnancy sickness not occur in all (or at least, almost all) pregnant women? Women who neither gag nor barf in any substantive way during their first trimester may be in the minority, but they’re not a small minority. Profet and others didn’t ignore this lack of universality, and they postulated some factors that may account for individual differences, yet the absence of morning sickness in so many women (even within the same society and with very similar diets) is difficult to reconcile with the strong adaptationist protection model.
One scientist who has begun poking holes in the Hook-Profet hypothesis and piecing together an alternative theory is SUNY-Albany psychologist Gordon Gallup. (Full disclosure: Gallup and I are like-minded, and I’m often partial to his ideas. I even gave his work on human penis adaptations the titular spot in my new essay collection.) Gallup claims that even the best evidence of categorical food aversions in early pregnancy is not, in fact, as compelling as it’s usually made out to be.
First, pregnant women aren’t the only ones to say that animal products such as meat, fish, poultry, and eggs are the most disgusting foodstuffs—nonpregnant women answer surveys the same way. Also, the embryo-protection account doesn’t really distinguish between the relative dangers of the two main food categories—both plant toxins and meat-borne pathogens, after all, can disrupt organogenesis and make the immunosuppressed mother ill. But it’s also common for women to report meat cravings during this period, much more so than they do for vegetables. Furthermore, there is conflicting evidence regarding the universality of these aversion patterns; women in some cultures may actually be more likely to develop morning sickness to common starch-based carbohydrates that have a low potential for toxicity.
Another problem for the Hook-Profet model, Gallup argues, is the fact that the evidence for pregnancy nausea and vomiting in other species is extremely weak, at best. Not only would similar selection pressures have been operating on the evolution of other species’ pregnancies for exactly the same reasons, more importantly, it’s unlikely to have arisen for the first time in hominids without some phylogenetic precedent that can still be easily observed in modern nonhuman animals. Yet nausea and vomiting in pregnancy have only been reported—and anecdotally so—in two other species: domestic dogs and rhesus macaques.