A train is moving at high speed and will run over five workers on the track. Would you push someone to be run over by the train and save five other people? This is one of the 60 dilemmas whose resolution has been investigated in the neural circuits that process decisions of a group of volunteers. The experiments, published by Joshua Green in the scientific journal Neuron, track people's brain activity by scanning their brains as they decide what to do in extreme situations. Most decide, quickly, not to push someone into the road. Neuroimaging techniques detect intense activation in areas of the right hemisphere that process the emotions underlying decisions that affect others.

A new experiment asks volunteers to prevent the five people from being run over if they manipulate the switches to divert the train onto a track where only one person is present. This action would cause possible indirect harm and directly prevent a higher evil. The majority opt to move the switches. In this case, it takes two more seconds to decide, whether the answer is yes or no to move the switches. It is then observed that the activation of areas of the brain that perform cognitive functions is more intense than in the dilemma of pushing someone. In contrast, activity in areas that process emotions is reduced.

In both experiments, the two types of intelligence through which human beings know appear: cognitive and emotional, each with greater activity in areas of one of the hemispheres of the brain. The left frontal brain processes in a more analytical, systematic, impersonal and slower way. For example, a reflection, even a brief one, may or may not move us to a solidarity financial aid to unknown victims of catastrophes in distant countries. The right hemisphere is more intuitive, global, staff and fast. For example, we feel urged ipso facto to help someone in grave danger. Except for pathologies, both systems are connected and act harmoniously.

These analyses allow us to better understand that the moral judgment that decides not to cause direct harm to a person involves a strong emotional component. Thanks to dopamine, the happiness hormone, the innate emotion of refusing to harm, or of liking to help, is converted into compassion in the engine of the brain's feelings. The person thus knows what is good or bad in himself.

In the case of the train, the volunteers choose, in five seconds and with an intense feeling of compassion, not to push anyone. Those who decide to do so spend seven seconds, two seconds more necessary to overcome the emotional barrier and guide their behavior by other motivations.

This scientific evidence shows that the natural human aversion to harm - expressed in the universal principle of don't do to others what you would not want done to you -emerges from two intimately connected brain systems: an emotional and a cognitive one. The rational facet, slower, financial aid when the immediate natural shortcut of feelings is not enough, but it is necessary to deliberate and calculate.

Also revealing is the research of Antonio Damasio's team published in Nature. They study how people with brain damage in the region that connects the emotional and the analytical solve ethical dilemmas. These patients follow an unusual utilitarian patron saint and quickly decide to kill -push one person down the road- to save five. However, in a more impersonal context, such as pushing needles, their behavior is normal. Because of this brain lesion, these people lack the innate guide alarm of emotion in moral judgment, although the deliberative system is maintained. Unpleasant feelings, the aversion to do harm that constitutes a signal of caution, leave them unperturbed.

If there is a contradiction between both components of human rationality, how does the analytical system prevail? The case of the train is again illustrative. When the dilemmas of pushing someone or changing the switches are presented to utilitarian volunteers - trained in the risk/benefit calculus as a behavioral rule - they resolve to both push and change the switches in the same time. In such cases they use the two seconds most needed in this mental activity to rationally adjust the cost/benefit, and thus avoid following the emotional, intuitive and natural shortcut to the right thing to do.

Animals are never wrong about what is good or bad for them: their instinct only allows them to be right. However, people, freed from the confinement of biological automatism, are faced with dilemmas and are open to making mistakes when deciding.

Codes of conduct provide a hierarchical scale of values that are considered relevant to qualify something as good or bad. They are not biologically determined, and therefore differ in normative aspects from one culture to another. As social regulations, they humanize when the legal and the ethical converge to reward the good (helping, curing) and penalize the bad (killing, not lending attendance in an accident). For this reason, there is a social schizophrenia when laws and ethics diverge.

Even with the cautions inherent to research on something as complex as the human mind, neuroscience today points to the way in which the natural and therefore universal principle of not doing to others what I do not want for myself is registered in the brain. The innate emotional shortcut in the face of dilemmas involving human lives is a natural predisposition to do the right thing.