Complex natural processes such as the control of food intake might inspire the design of new Artificial Intelligence algorithms.
Specifically, research in developmental programming reveals how the nervous system reprograms itself even before birth.
The traditional view of the nervous system, prior to the discovery of epigenetics, described it as having a fixed initial design, as specified by genes (a design of the neural architecture that would later be the basis for learning processes). However, as we know now, the interaction of genes with the environment triggers functional changes in the nervous system during pregnancy. These design changes are adaptations to the specific ecosystem in which the fetus is thriving.
If we make the analogy with the control system of a robot, it would be something like this: when a brand new robot leaves the factory it comes with a standard battery management program (as pre-programmed in its “genetics code”), when it arrives at its destination and discovers that there is no easy access to recharging stations (“famine” as the “early interaction with the environment”), a meta-program ( the “epigenetic” code) would be responsible for modifying the “innate” recharging and energy saving behaviors (“basic instincts adapted to the environment” ).
Malnutrition in early stages of development, followed by improved nutrition, causes accelerated growth in many species of animals and plants. This stage of rapid growth can be seen as a compensatory mechanism of the nutritional deficiency experienced in the past. Although this mechanism can be advantageous from the point of view of short-term adaptation, it also implies an important cost for the organism, since it has been proven that it reduces the longevity of the animals that experience it.
According to the Predictive Adaptive Response (PAR) hypothesis (Gluckman and Hanson, 2004), the organism adapts its physiology to adapt to the environment present in the critical perinatal period, assuming (predicting) that this is the ecological context in which it will have to to live. The problem is that this prediction of the characteristics of the environment usually becomes erroneous in the developed world, where individuals are exposed to high-calorie food and the energy expenditure derived from physical exercise is minimal.
In general, we know that the child’s food intake behavior and metabolism are affected by the interaction with the environment that occurs in early stages of development. For this reason, fetal and maternal care is so important, because during pregnancy there are changes in the nervous system that can have consequences in later stages of development, including adulthood (Arrabales, 2017).
Therefore, to think that obesity and overweight are the result of the lack of will or a sedentary lifestyle is simply an unscientific reductionism. Weight regulation and food intake behavior systems depend on multiple factors, including the ability of the nervous system to alter its own genetic programming during development.
Overweight, obesity and other metabolic disorders are often associated with the western lifestyle. Although it is true that the lack of physical exercise, a sedentary lifestyle and a high-fat high-calorie diet are factors that contribute to obesity in adults, specialized literature on endocrinology and associated studies show that considering these as the only causal factors is incorrect (Ross and Desai, 2014).
There is no doubt that lifestyle has an important impact on the development of these diseases, but an effective approach to the problem necessarily requires understanding what other key factors contribute to the regulation of appetite and metabolism.
Keeping the focus on the perinatal context, we know that malnutrition or overnutrition in this period of development can unleash major changes with health consequences that manifest in later stages of life (heart disease, insulin resistance, hypertension, etc.). A detailed knowledge of the mechanisms that take place during the perinatal period and the understanding of the interaction between the nutritional environment and the genotype may offer new perspectives for the treatment of obesity-related disorders (Vickers and Sloboda, 2012). In addition, from the point of view of bio-inspiration, understanding how the nervous system programs itself to adapt to the environment (developmental programming), could inspire the design of new algorithms for artificial adaptive systems.
Body weight and food intake regulation are largely controlled by the hypothalamus. Recent research has shown that there is a process of metabolic imprinting, through which the postnatal development of the hypothalamus is affected by the nutritional environment to which the individual is exposed (Bouret and Simerly, 2006). The levels of various hormones represent the conditions existing in the environment of the fetus or newborn, and in turn also directly influence the central nervous system, regulating its activity and also its very development.
The arcuate nucleus (ARC) of the hypothalamus has been identified as one of the key centers in the metabolic programming of development. Its projections to other hypothalamic nuclei, induced during development, such as the paraventricular nucleus (NPV), determine the future normal regulation of body weight and glucose homeostasis (Taylor and Poston, 2007).
In the specific case of food intake behavior, there is evidence that certain perinatal nutritional patterns (fetal and maternal) induce changes in the postnatal metabolic phenotype related to appetite and satiety signals (Ross and Desai, 2014). In other words, a “bad programming” of the appetite and satiety signals will therefore be responsible for the increased risk of obesity and associated disorders.
Epidemiological evidence indicates that the children of mothers exposed to famine during the first two trimesters of pregnancy have a higher incidence of obesity than the reference population (Ravelli, van Der Meulen, Osmond, Barker, & Bleker, 1999). The underlying mechanism in these contexts would be hyperphagia and programmed adipogenesis during perinatal malnutrition, in conjunction with subsequent exposure to caloric excess and consequent compensatory growth (of course, this doesn’t mean that all overweight or obesity comes from perinatal malnutrition).
Another source of evidence of programmed hyperphagia comes from animal models, predominantly rodents, on which maternal and neonatal nutrition is manipulated, reproducing results analogous to those observed in epidemiological evidence in humans.
The results obtained in studies with animal models confirm the tendency to obesity due to conditions of low birth weight and subsequent accelerated compensatory growth, as well as that altered insulin / glucose homeostasis is programmed in utero (Armitage, Khan, Taylor , Nathanielsz, & Poston, 2004).
References
Armitage, J. A., Khan, I. Y., Taylor, P. D., Nathanielsz, P. W. y Poston, L. (2004). Developmental programming of the metabolic syndrome by maternal nutritional imbalance: How strong is the evidence from experimental models in mammals? The Journal of Physiology, 561(2), 355-377
Arrabales, R. (2017) “Estrategias para la Reversión y la Modulación de la Hiperfagia Programada Perinatalmente”. UNED.
Gluckman, P. D. y Hanson, M. A. (2004). The fetal matrix: Evolution, development and disease Cambridge University Press.
Ravelli, A. C., van Der Meulen, J. H., Osmond, C., Barker, D. J. y Bleker, O. P. (1999). Obesity at the age of 50 y in men and women exposed to famine prenatally. The American Journal of Clinical Nutrition, 70(5), 811-816.
Ross, M. G. y Desai, M. (2014). Developmental programming of appetite/satiety. Annals of Nutrition & Metabolism, 64 Suppl 1, 36-44.
Vickers, M. H. y Sloboda, D. M. (2012). Strategies for reversing the effects of metabolic disorders induced as a consequence of developmental programming. Frontiers in Physiology, 3, 242.
