Sometimes the best way to learn how the brain works is to watch what happens when it goes awry. When one part—a clump of neurons or a brain-building gene—doesn’t do what it is supposed to, the brain may fail in an illuminating way. Its failure may even expose some of the hidden foundations of the mind.

Neuroscientists have recently become fascinated with a particularly telling pair of rare brain disorders. One was identified in 1965 by English physician Harry Angelman, who was struck by the faces of three children he treated. These children were always smiling and often laughing. This disorder, now known as Angelman syndrome, affects around 1 in 20,000 children. Along with the smiles and laughs come other symptoms, some of which overlap those of severe autism. Many children with Angelman syndrome never learn to speak or read. They also keep their bodies in motion, often flapping their hands. When they nurse they suckle desperately, thrusting out their tongue.

Another similarly rare condition, called Prader-Willi syndrome, produces a different set of symptoms. Babies with Prader-Willi nurse very little—so little that they often have to be tube fed. However, once Prader-Willi children get to be a few years old, they develop an insatiable appetite. They will try to get around any obstacle put between them and food. Their fierce hunger is driven by a malfunctioning hypothalamus, a region deep in the brain that governs hunger and growth. Instead of autism, many people with Prader-Willi syndrome develop schizophrenia by adulthood, hearing voices and generating paranoid delusions.

Despite their differences, Prader-Willi and Angelman are two sides of the same coin. Scientists have searched for the genetic basis of the two syndromes and have tracked most cases of both to defects on the same spot of the human genome, a stretch of DNA on chromosome 15. Which disease a child gets depends on which parent’s chromosome 15 carries the defect (every person’s cells contain two genetic copies, one from the mother and one from the father). Prader-Willi syndrome is caused by a mutation in a father’s genes that deletes a chunk of DNA on chromosome 15. Angelman syndrome is associated with a mutation on the mother’s chromosome 15.

If you think back to the genetics you learned in school, this pattern makes no sense. A gene is a gene is a gene. Two identical stretches of DNA ought to have the same effect on a child, regardless of which parent it comes from. But sometimes our genes break the rules of high-school genetics. The effects of dozens, perhaps hundreds, of genes depend on whether you inherited them from your mother or your father.

Dissimilarities arise because not all genes are actively expressed in our cells. Some of the genes get switched off, or silenced. Each time a cell divides and makes a new copy of its DNA, special enzymes attach caps at certain spots along the copy’s length. Those caps make it impossible for a cell to read the specific genes they are attached to. As a result, those genes can’t make the corresponding proteins. In some cases the caps are attached to only one parent’s copy of a gene. The other parent’s copy remains uncapped, free to produce proteins.

This parental silencing is known as gene imprinting, and it is turning out to be important to our health. Mutations that change the pattern of imprinting can have a big effect on our bodies. If a cell fails to imprint one of the two parents’ genes, for example, it will have two genes producing a protein instead of just one. The cell will make twice as many copies of the protein.

The discovery of gene imprinting in 1984 raised a big question: Why should genes from one parent be silenced in the first place? In 1999, David Haig of Harvard University offered a startling hypothesis. Gene imprinting, he proposed, is the result of the evolutionary struggle between mothers and fathers for reproductive success. This fight doesn’t take place between mothers and fathers themselves but between the genes that they pass down to their offspring. Natural selection favors genes that can make more copies of themselves. But the best strategy for the genes of fathers is not the same as the one that’s best for the genes carried by mothers.

For millions of years, mothers have had to invest a huge amount of time and effort in their children. The investment starts in the womb as mothers supply growing fetuses with nutrients.

Haig noted that the demands of reproduction have forced mothers into an evolutionary trade-off. If they invest a lot in one child, he or she gets bigger and healthier and more likely to survive to adulthood. But investing too much in one child can undermine the well-being of any siblings and can put a mother’s own health at risk. The most successful solution is to invest a lot—but not too much—in each child.

Fathers do not face this trade-off. As a result, natural selection should favor a different strategy for their genes. From a dad’s perspective, the more nutrients his child can get from the mother, the more likely it is that the child will grow up healthy and pass on his or her father’s genes.