The Parasite Stress Model of the Trivers-Willard Hypothesis
Parental investment is debatably one of the most important topics in evolutionary psychology. Although not present in all species, it is essential for the survival of human offspring. Without parental care and protection throughout early childhood, it would be almost impossible for any individual to live long enough to reproduce. It is not surprising, therefore, that much research has investigated this phenomenon.
One of the major pillars of parental investment theory is the Trivers-Willard hypothesis. It revolutionized the research in this field and provided a framework which allowed for greater organization. Much research has investigated this hypothesis ever since it was first proposed. This paper begins with a brief explanation of the major claims of the hypothesis.
The Trivers-Willard Hypothesis
In 1973, Robert Trivers and Dan Willard published their seminal work on varying degrees of parental investment based on the sex of the offspring. They proposed what would later become known as the Trivers-Willard Hypothesis. The concept is simple. When parents invest equally in both male and female offspring, natural selection tends to result in an equilibrium state where there is a one-to-one ratio of male-to-female offspring (Trivers, 1972). However, as this is usually not the case, Trivers and Willard (1973) formulated a rule that predicted bias in parental investment towards offspring of a specific sex. This rule was based on three assumptions. The first assumption is that the condition of the offspring when they have reached an age where they no longer require parental investment from the mother can be accurately predicted by the mother’s condition while she was providing parental care. Secondly, offspring that display more gains at the end of the term of parental investment tend to retain those gains as they continue into adulthood. Thirdly, due to the fact that males are subject to higher variability in reproductive success than females, male offspring benefit from parental investment more than female offspring as the overall condition tends towards optimality (Trivers & Willard, 1973). In layman’s terms, since male offspring can either father no children or leave behind a substantial progeny, males constitute an evolutionary gamble that parents should only invest in when conditions are good. On the other hand, female offspring are less of a risk as they rarely fail to produce at least one child of their own.
From the time the Trivers-Willard Hypothesis was first proposed, a substantial body of research has accumulated in its support. Multiple studies have shown that parents do, in fact, vary investment based on the sex of the offspring and the overall condition. For example, in a survey of breastfeeding mothers in Kenya, women who were financially secure produced richer milk (higher in fat content) for sons than for daughters. This relationship was reversed in lower socioeconomic mothers, who provided their daughters with better milk (Fujita et al., 2012).
It is important to note that the Trivers-Willard hypothesis is not limited to postnatal parental investment. A mother can vary the amount of parental investment provided in utero which has a direct effect on the success of the pregnancy. If a woman decides to withdraw her investment from the fetus, on an unconscious biological level of course, this may result in the abortion of the fetus. A decrease in the mother’s condition can lead to sex-differentiated abortions which increases the relative odds of producing female offspring. A higher frequency of male offspring is observed with an increase in optimality in the mother’s condition due to sex-differentiated abortions that preferentially terminate female fetuses. This assumes that the mother has some unconscious knowledge about the sex of the fetus, even before it is born. Through this elaborate mechanism, a mother’s overall condition can exert some degree of influence over her offspring’s sex-ratio at birth (SRB). As such, the Trivers-Willard hypothesis can be reformulated as predicting the variation in SRB based on the overall environmental condition of the parents.
Major Markers of Parental Condition
An important question then arises: what should could be used as a marker for the general variable of “parental condition”? Numerous empirical studies have examined the effects of various elements on the SRB in human samples. It is important to note that experimentally, the SRB is defined as the ratio of male-to-female births. As such, an increased SRB means more male than female offspring are being produced, and vice versa. Also, SRB does not take into account variations in parental investment after birth which may result in differential survival for male and female offspring. Some of the major proposed cues to parental condition are discussed below.
Research investigating the relationship between SRB and ambient temperature has yielded inconsistent results. Some studies have shown that an increase in SRB is associated with warmer temperatures (Helle, Helama, & Lertola, 2009; Catalano, Bruckner, & Smith, 2008). However, these results are not replicated consistently. A recent study of the Australian population showed that SRB could not be predicted from records of ambient temperature (Barnaby, Haywood, Lester, & Ormsby, 2013). Due to these contradictory results, we cannot yet describe this relationship with certainty and further research into the matter is needed. Of all the studied predictors of SRB, ambient temperature has the least conclusive evidence.
Studies have shown that wealth is positively correlated to SRB in the human population. A recent study by Cameron and Dalerum (2009) examined a sample of 397 billionaires from all over the world. They discovered that the billionaires produced more male offspring (60%) than the rest of the general population (51%). This difference in SRB was independent of the origin of wealth (inherited wealth vs. earned wealth vs. growing inheritance) and was more pronounced for male billionaires than female billionaires (Cameron & Dalerum, 2009). A second study, using more complete data, found that a statistically significant increase in SRB only occurred in the sub-group of billionaires who had inherited their wealth at a young age (Sebastian, 2013). In this subpopulation, wealth would have influenced the entirety of their reproductive history. The same study found an increased SRB in the general billionaire population, however this increase was not statistically significant (Sebastian, 2013). Taken together, these results provide support for the hypothesis that wealth is a predictor of high SRB in the population of billionaires who acquired their wealth before the conception of all their children.
Latitude is also thought to play a role in determining the SRB. Although the link between the two is unclear, substantial empirical evidence has shown that geographical latitude and SRB are positively correlated. Studies have shown that as latitude increases, so does SRB. In other words, as you move further away from the equator, the ratio of male-to-female births increases (Navara, 2009). These results have been replicated several times (e.g. Grech, Savona-Ventura, & Vassallo-Agius, 2002). However, some attempts not only failed to replicate these results but also found the opposite: male births declining with increasing latitude (Grech, Vassallo-Aguis, & Savona-Ventura, 2000). Although the evidence is not unanimous, most of the research suggests that male births do indeed increase with increasing latitude. Further studies are needed to solidify this claim.
Violence of the Father
Although it sounds very counterintuitive, one study has shown that how violent a man is significantly predicts the male-to-female ratio of his offspring. Empirical evidence indicates that the more violent a man is, the more likely he is to father sons instead of daughters (Kanazawa, 2006). A review of the literature could find no other studies that attempted to replicate these findings. Although, the validity of these findings is called into question by the dearth of supportive replication, these findings cannot be written off as having no empirical merit.
Another bizarre finding is that the beauty of the parents also has a role to play in the SRB. Kanazawa (2011) found that in a sample of 17,419 British individuals, children who were rated as being attractive at the age of seven by teachers were 23% more likely to have a daughter at a 40 year follow-up. Conversely, children who were rated as being unattractive at the age of seven by teachers were 25% more likely to have a son at the 40 year follow up (Kanazawa, 2011). These results were in agreement with previous research that showed that parents labelled as very attractive had 26% lower odds of having a son compared to the rest of the population (Kanazawa, 2007).
Big and Tall Parents
Finally, it has been shown that parents who are bigger and taller tend to have a higher male-to-female offspring ratios. Kanazawa (2005) found that in a sample that initially included more than 17,000 individuals, both the height and weight of the parents were positively correlated with the number of male offspring produced. This correlation remained statistically significant even after controlling for the number of female fetuses, to rule out the effect of overall increased fertility (Kanazawa, 2005).
There is substantial evidence indicating that the number of male offspring produced decreases significantly in the wake of some form of environmental crisis or natural disaster. For example, the SRB of the Japanese population affected by the Kobe earthquake decreased significantly after the disaster (Fukuda, Fukuda, Shimizu, & Moller, 1998). Similarly, the rates of male births declined significantly in the wake of a short war in Slovenia (Zorn, Sucur, Stare, & Meden-Vrtovec, 2002). Although both of these studies explained the variation in SRB in terms of a decrease of sperm motility, this is an invalid justification. A decrease in sperm motility should affect the overall fertility of the population (number of births) while maintaining the SRB (male-to-female ration). The authors did not attempt to explain these results based on fetus abortion, an alternative that appears more consistent with the available data.
A relatively recent study by (Dama, 2012) found some very important results. He found several factors which were predictive of SRB, one of which was parasite stress: the number of disease-causing parasites encountered by the average individual. Parasite stress was found to be negatively correlated to SRB. Not only that, but it was also statistically shown to be the strongest predictor by far of SRB among all the other predictors detected (e.g. fertility, polygyny, son-preference, and latitude; Dama, 2012). The other predictors are all in line with previous research mentioned above.
Piecing Together the Evolutionary Puzzle
As discussed above, there is a substantial body of research investigating potential cues of parental condition and how they relate to SRB. Time and time again, the variation in male-to-female offspring occurs in the direction predicted by the Trivers-Willard hypothesis. The ability of this hypothesis to repeatedly generate valid predictions has earned it significant empirical support. However, the Trivers-Willard hypothesis does not provide a mechanism that details the psychological process which translates environmental cues of condition into sex-differential abortions.
One theory has been set forth by Kanazawa (2005) who postulates a generalized Trivers-Willard hypothesis (gTWH) that states:
"Parents who possess any heritable trait which increases the male reproductive success at a greater rate…than female reproductive success…will have a higher-than-expected offspring sex ratio…Parents who possess any heritable trait which increases the female reproductive success at a greater rate…than male reproductive success…will have a lower-than-expected offspring sex ratio…" (Kanazawa, 2005, p.585)
The gTWH attempts to bring all the different variables linked to SRB and summarize them with a simple heuristic that parents will produce more offspring of the sex that will yield the greatest benefit from the provided genes. Based on this logic, violent men tend to produce more sons due to the fact that male offspring would benefit from possessing a “violent gene” more than female offspring (Kanazawa, 2006). Similarly, attractive parents tend to produce more daughters than sons as girls benefit from physical attractiveness more than boys (Kanazawa, 2007).
This hypothesis not only applies to genetically inherited traits but also to culturally inherited traits such as wealth (Cameron & Dalerum, 2009). In fact, it is fair to claim that gTWH is nothing more than an analysis of the benefit of inheritance. However, this theory fails to explain how factors such as ambient temperature, latitude, parasite stress, and environmental crises can influence SRB. Based on this shortcoming, it seems safer to assume that it is variations in the potential pay-off of parental investment, and not the value of genes passed on, that determines the SRB. In other words, as long as the additional costs of investing in male offspring culminates in substantial evolutionary returns, the SRB is likely to tip in favor of male offspring.
How can parents, more specifically the mother, assess whether or not it would pay-off to go the extra mile and invest in male offspring? What cue, or cues, does the mother use to determine her own condition? If there are in fact multiple cues, how are they integrated? Although these questions remain unanswered, I propose a model in which the mother only employs information from one cue: parasite stress.
Dama (2012) found that parasite stress was by far the strongest predictor of SRB variation. As parasite stress increased, SRB decreased significantly. However, this alone does not indicate that parasite stress is the only cue used in determining maternal condition and therefore SRB. We cannot ignore the fact that all the aforementioned variables do in fact predict SRB. However, upon further inspection, it becomes apparent that all these other variables vary as a function of parasite stress.
Wealth, for example, tends to show a negative correlation with parasite stress. For example, Sonko et al. (2014) found that rates of malaria were much higher in individuals from the poorer quintiles of the population. To the astute reader, this finding may be intuitive. Greater wealth leads to greater education, better hygiene and living quarters, better nutrition, and better health care. Therefore, people from higher socioeconomic strata will be exposed to less parasites. This is also in line with the finding that wealth only predicts SRB if the individual had acquired the wealth prior to the conception and birth of all his or her children (Sebastian, 2013).
Similarly, there is a plethora of evidence showing that geographic latitude and parasite stress are strongly negatively correlated. For all major parasites, parasite density decreases with increasing latitude (Dama, 2012). The fact that bigger and taller parents tend to produce more sons can also be explained in a similar fashion. Individuals who are reared in environments that are low in parasites tend to approximate their maximum potential height. As such, height and weight may be presumed to be signs of low parasite density. Also, after environmental crises such as natural disasters and wars, access to health care decreases. Similarly, sanitation decreases which allows for an increased incidence of parasitic infections. This in turn would result in the observed drop in SRB. Although this link has not been empirically substantiated, it is a logical link between crises and parasite stress.
Research into the link between ambient temperature and SRB has produced contradictory findings. However, the parasite stress theory of SRB predicts that the relationship between the two should vary with geography. Studies that find that more sons are produced in warm temperatures are normally carried out in Northern European countries where the climate is generally cold (Catalano et al., 2008; Helle et al., 2009). In these environments, parasites would be adapted to living in a colder climate. As such, when the temperature increases, the efficiency of the parasites is compromised resulting in a decrease of parasitic infection and an increase in SRB. On the other hand, studies carried out on an Australian sample, where the climate is generally warm, have found no relationship between temperature and SRB (Barnaby et al., 2013).
Of all the above mentioned cues to parental condition, two do not conform to the predictions of this parasite-stress theory: degree of paternal violence and the overall beauty of the parents. Beauty is an indicator of low parasite density. This in turn would predict that more male offspring would be born. This, however, is not the case. There is evidence that violence also increases with increasing parasite density, especially in men (Thornhill & Fincher, 2011). Further research into these two variables is needed as there may be other forces in play when it comes to their relationship to SRB.
Combining all the evidence, a good argument can be made that the only cue mothers really use is parasite stress (which can easily be measured as frequency of infection, symptomatic or otherwise). All other variables do not affect SRB directly but rather cause subtle changes in parasite stress. Any variable that increases parasite stress, or is itself caused by an increase in parasite stress, will be negatively correlated to SRB. Conversely, any variable that decreases parasite stress, or is itself caused by a decrease in parasite stress, will be positively correlated to SRB. All changes to SRB occur via a change in the rate of sex-differential fetal abortions.
Future research is needed to test the validity of the parasite stress model of the Trivers-Willard hypothesis. For example, a longitudinal study that follows newly engaged couples could track the frequency of illness in both partners. Over the course of the next decade, as the couples begin to produce offspring, the SRB could be measured. We could then statistically test the claim that couples who get sick more often would, ex hypothesi, have a lower SRB on average than more healthy couples. Alternatively, studies investigating the relationship between other variables could control from parasite stress and see if the results remain significant.
The parasite stress model is in no way an attempt to replace the Trivers-Willard hypothesis, only to enhance our understanding of the functional mechanism underlying it. Although future research may either support or refute it, it provides a logical explanation for a well-established phenomenon. It unifies all the predictors of the sex ratio at birth and combines physiological, social, and geographic cues into one overarching theory. As our knowledge on the subject increases, it may revolutionize our understanding of pregnancy, parental investment, and the ability to manipulate the sex-ratio of entire populations.
Barnaby, J. D., Haywood, J., Lester, P. J., & Ormsby, D. K. (2013). Feeling the heat? Substantial variation in temperatures does not affect the proportion of males born in Australia. Human Biology, 85(5), 757-767. doi: 10.3378/027.085.0506
Cameron, E. Z., & Dalerum, F. (2009). A Trivers-Willard effect in contemporary humans: Male-biased sex ratios among billionaires. PLoS ONE, 4(1), 1-4. doi: 10.1371/journal.pone.0004195
Catalano, R., Bruckner, T., & Smith, K. R. (2008). Ambient temperature predicts sex ratios and male longevity. Proceedings of the National Academy of Sciences of the United States of America, 105(6), 2244-2247. doi: 10.1073/pnas.0710711104
Dama, M. S. (2012). Parasite stress predicts offspring sex ratio. PLOS ONE, 7(9), doi: 10.1371/journal.pone.0046169
Fujita, M., Roth, E., Lo, Y. J., Hurst, C., Vollner, J., & Kendell, A. (2012). In poor families, mothers' milk is richer for daughters than sons: a test of Trivers-Willard hypothesis in agropastoral settlements in Northern Kenya. American Journal of Physical Anthropology, 149(1), 52-59. doi: 10.1002/ajpa.22092
Fukuda, M., Fukuda, K., Shimizu, T., & Moller, H. (1998). Decline in sex ratio at birth after Kobe earthquake. Human Reproduction, 13(8), 2321-2322. Retrieved from http://www.oxfordjournals.org/en/
Grech, V., Savona-Ventura, C., & Vassallo-Agius, P. (2002). Unexplained differences in the sex ratio at birth in Europe and North America. British Medical Journal, 324, 1010. doi: 10.1136/bmj.324.7344.1010
Grech, V., Vassallo-Aguis, P., & Savona-Ventura, C. (2000). Declining male births with increasing geographical latitude in Europe. Journal of Epidemiology and Community Health, 54(4), 244-246. doi: 10.1136/jech.54.4.244
Helle, S., Helama, S., & Lertola, K. (2009). Evolutionary ecology of human birth sex ratio under the compound influence of climate change, famine, economic crises and wars. Journal of Animal Ecology, 78(6), 1226-1233. doi: 10.1111/j.1365-2656.2009.01598.x
Kanazawa, S. (2005). Big and tall parents have more sons:Further generalizations of the Trivers–Willard hypothesis. Journal of Theoretical Biology , 235(4), 583–590. doi: 10.1016/j.jtbi.2005.02.010
Kanazawa, S. (2006). Violent men have more sons: Further evidence for the generalized Trivers–Willard hypothesis (gTWH). Journal of Theoretical Biology , 239(4), 450–459. doi: doi:10.1016/j.jtbi.2005.08.010
Kanazawa, S. (2007). Beautiful parents have more daughters: A further implication of the generalized Trivers–Willard hypothesis (gTWH). Journal of Theoretical Biology, 244(1), 133-140. doi: 10.1016/j.jtbi.2006.07.017
Kanazawa, S. (2011). Beautiful British parents have more daughters. Reproductive Sciences, 18(4), 353-358. doi: 10.1177/1933719110393031
Navara, K. J. (2009). Humans at tropical latitudes produce more females. Biology Letters, 5(4). doi: 10.1098/rsbl.2009.0069
Sebastian, S. (2013). Revisiting a sample of U.S. billionaires: How sample selection and timing of maternal condition influence findings on the Trivers-Willard effect. PLoS ONE, 8(2), 1-6. doi: 10.1371/journal.pone.0057446
Sonko, S. T., Jaiteh, M., Jafali, J., Jarju, L. B., D'Alessandro, U., Camara, A., . . . Saho, A. (2014). Does socio-economic status explain the differentials in malaria parasite prevalence? Evidence from The Gambia. Malaria Journal, 13, 449. doi: 10.1186/1475-2875-13-449
Trivers, R. L. (1972). In B. Campbell, Sexual Selection and the Descent of Man (pp. 136-179). Chicago: Aldine-Atherton
Trivers, R. L., & Willard, D. E. (1973). Natural selection of parental ability to vary the sex ratio of offspring. Science, 179(4068), 90-92. Retrieved from http://www.jstor.org/stable/1734960
Zorn, B., Sucur, V., Stare, J., & Meden-Vrtovec, H. (2002). Decline in sex ratio at birth after 10-day war in Slovenia: Brief communication. Human Reproduction, 17(12), 3173-3177. Retrieved from http://www.oxfordjournals.org/en/