In their article, Nettle et al. propose an evolution-based hypothesis to explain why obesity is more prevalent in some human populations than in others. According to their insurance hypothesis (IH), individuals should respond to food insecurity cues by experiencing psychological and behavioral changes that promote increased fat storage. Such changes are posited to be driven by the increased survival afforded by fat stores in buffering against energy shortfalls during times of food scarcity.
We agree with the authors’ general supposition that food insecurity should promote increased energy storage, as it can buffer the organism from dropping below levels necessary for survival and/or reproduction if food becomes temporarily unavailable. Indeed, one of the primary selective forces that has granted organisms the ability to store energy in the first place has been the advantages it provides organisms when food is scarce, variable, or both. However, a good theory proves its mettle by parsimoniously accounting for existing patterns of data and generating new predictions, and the IH falls short in both regards. In particular, the IH seems to be missing important parameters that would allow it to account for (and make novel predictions about) (a) sex differences in the relationship between food insecurity and adiposity observed in some species, such as humans, but not in other species, such as birds, and (b) observed differences between developed and developing nations in the degree to which food insecurity predicts adiposity. We elaborate on these points below.
To evaluate their hypothesis, Nettle et al. present a theoretical model of optimal eating behavior under varying conditions of food security. The authors then reviewed evidence suggesting that food insecurity causes weight gain among species of birds. This is followed by a meta-analysis of research examining the association between food security and body weight in human populations. The relatively straightforward association between food insecurity and adiposity predicted by the authors’ model accounted well for the pattern of results obtained in birds. Male and female birds demonstrated increased storage of adipose tissue when in resource-scarce conditions compared to when in resource-abundant conditions. In humans, however, the association between food insecurity and body weight was highly conditional. Specifically, the predicted association was only observed for women in high-income countries and did not emerge until after the post-pubertal transition.
One of the strengths of theoretical models in evolutionary biology is that they can be used to make nuanced predictions about differences one should expect to observe between species, between the sexes, and even between individuals within a species by considering the target organism's life history, mating system, ecology, obligatory investment in offspring, and so on. Unfortunately, the IH does not capitalize on these theoretical strengths. Instead, the authors describe an overly simplified model in which optimal eating behavior is determined largely by food security or availability, current fat reserves, and the probability of death by starvation or other causes. However, there are at least two factors known to play key roles in determining adiposity that have been excluded from the main theoretical model that, if added, would go a long way in accounting for the observed differences between men and women, humans and birds, and people living in high-income and low-income countries. In particular, we recommend that the authors more deliberately consider (1) the importance of body fat for reproduction, and (2) environmental and social factors impacting access to food among the poor.
Bearing on the first of these points, the authors need to consider the relative importance of body fat for successful reproduction in males and females. In the eyes of evolution, reproduction is at least as important as survival. Accordingly, an organism's energy regulation mechanisms should be finely tuned to ensure, specifically, that the organism is able to meet the energy requirements needed for reproduction. This requirement is likely to vary considerably depending on the sex and species that is being considered. For an internally gestating human female, for example, the energy demands of reproduction are far greater than those for the non-gestating, non-lactating, human male. Indeed, research finds that women's ovarian function, gestation length, offspring birthweight, and milk production are each sensitive to women's energy balance and are optimized when energy is readily available. The quantity and viability of men's sperm, on the other hand, are relatively unaffected by men's energy status unless they approach the starvation threshold (Ellison Reference Ellison2003; Fontana & Torre Reference Fontana and Torre2016).
Similarly, because successful reproduction imposes on internally gestating mammals vastly different demands than those that are imposed on an egg-laying avian species, we should expect cross-species differences in fat regulation mechanisms that reflect these differences. Perhaps reproduction is optimized among birds when both males and females possess a certain amount of body fat due to the role played by each in egg incubation. Incorporating the energy demands of reproduction into the IH model would undoubtedly go a long way in accounting for why the predicted effects apply selectively to human women, but generalize across male and female birds. It would also likely generate novel predictions about how males and females of other species adjust their fat stores in response to food insecurity, as well as make predictions about the species in which sex differences are expected and those in which they are not.
As a more minor point, we also encourage the authors to consider adding to their model a society-level parameter capturing access to calories among the poor. Access to food among the poor is higher in high-income countries than in middle- and low-income countries (e.g., see Levine Reference Levine2011). Accordingly, mechanisms that increase fat storage in response to resource scarcity are more likely to promote obesity in these high-income countries because their poor have the opportunity to overeat. Among those living in lower- and middle-income countries, these same mechanisms would not promote obesity because they would be operating in an environment similar to their selective context (Prentice Reference Prentice2001). Adding this parameter would likely lend additional flexibility to the authors’ model, making it more predictive of observed obesity patterns.
In their article, Nettle et al. propose an evolution-based hypothesis to explain why obesity is more prevalent in some human populations than in others. According to their insurance hypothesis (IH), individuals should respond to food insecurity cues by experiencing psychological and behavioral changes that promote increased fat storage. Such changes are posited to be driven by the increased survival afforded by fat stores in buffering against energy shortfalls during times of food scarcity.
We agree with the authors’ general supposition that food insecurity should promote increased energy storage, as it can buffer the organism from dropping below levels necessary for survival and/or reproduction if food becomes temporarily unavailable. Indeed, one of the primary selective forces that has granted organisms the ability to store energy in the first place has been the advantages it provides organisms when food is scarce, variable, or both. However, a good theory proves its mettle by parsimoniously accounting for existing patterns of data and generating new predictions, and the IH falls short in both regards. In particular, the IH seems to be missing important parameters that would allow it to account for (and make novel predictions about) (a) sex differences in the relationship between food insecurity and adiposity observed in some species, such as humans, but not in other species, such as birds, and (b) observed differences between developed and developing nations in the degree to which food insecurity predicts adiposity. We elaborate on these points below.
To evaluate their hypothesis, Nettle et al. present a theoretical model of optimal eating behavior under varying conditions of food security. The authors then reviewed evidence suggesting that food insecurity causes weight gain among species of birds. This is followed by a meta-analysis of research examining the association between food security and body weight in human populations. The relatively straightforward association between food insecurity and adiposity predicted by the authors’ model accounted well for the pattern of results obtained in birds. Male and female birds demonstrated increased storage of adipose tissue when in resource-scarce conditions compared to when in resource-abundant conditions. In humans, however, the association between food insecurity and body weight was highly conditional. Specifically, the predicted association was only observed for women in high-income countries and did not emerge until after the post-pubertal transition.
One of the strengths of theoretical models in evolutionary biology is that they can be used to make nuanced predictions about differences one should expect to observe between species, between the sexes, and even between individuals within a species by considering the target organism's life history, mating system, ecology, obligatory investment in offspring, and so on. Unfortunately, the IH does not capitalize on these theoretical strengths. Instead, the authors describe an overly simplified model in which optimal eating behavior is determined largely by food security or availability, current fat reserves, and the probability of death by starvation or other causes. However, there are at least two factors known to play key roles in determining adiposity that have been excluded from the main theoretical model that, if added, would go a long way in accounting for the observed differences between men and women, humans and birds, and people living in high-income and low-income countries. In particular, we recommend that the authors more deliberately consider (1) the importance of body fat for reproduction, and (2) environmental and social factors impacting access to food among the poor.
Bearing on the first of these points, the authors need to consider the relative importance of body fat for successful reproduction in males and females. In the eyes of evolution, reproduction is at least as important as survival. Accordingly, an organism's energy regulation mechanisms should be finely tuned to ensure, specifically, that the organism is able to meet the energy requirements needed for reproduction. This requirement is likely to vary considerably depending on the sex and species that is being considered. For an internally gestating human female, for example, the energy demands of reproduction are far greater than those for the non-gestating, non-lactating, human male. Indeed, research finds that women's ovarian function, gestation length, offspring birthweight, and milk production are each sensitive to women's energy balance and are optimized when energy is readily available. The quantity and viability of men's sperm, on the other hand, are relatively unaffected by men's energy status unless they approach the starvation threshold (Ellison Reference Ellison2003; Fontana & Torre Reference Fontana and Torre2016).
Similarly, because successful reproduction imposes on internally gestating mammals vastly different demands than those that are imposed on an egg-laying avian species, we should expect cross-species differences in fat regulation mechanisms that reflect these differences. Perhaps reproduction is optimized among birds when both males and females possess a certain amount of body fat due to the role played by each in egg incubation. Incorporating the energy demands of reproduction into the IH model would undoubtedly go a long way in accounting for why the predicted effects apply selectively to human women, but generalize across male and female birds. It would also likely generate novel predictions about how males and females of other species adjust their fat stores in response to food insecurity, as well as make predictions about the species in which sex differences are expected and those in which they are not.
As a more minor point, we also encourage the authors to consider adding to their model a society-level parameter capturing access to calories among the poor. Access to food among the poor is higher in high-income countries than in middle- and low-income countries (e.g., see Levine Reference Levine2011). Accordingly, mechanisms that increase fat storage in response to resource scarcity are more likely to promote obesity in these high-income countries because their poor have the opportunity to overeat. Among those living in lower- and middle-income countries, these same mechanisms would not promote obesity because they would be operating in an environment similar to their selective context (Prentice Reference Prentice2001). Adding this parameter would likely lend additional flexibility to the authors’ model, making it more predictive of observed obesity patterns.