Colossal. The progress made in recent years by the artificial intelligence are, without any doubt. But he still has to do with a rival of size : the human brain and the incredible plasticity of his neurons. It is this that explains the neuroscientist Stanislas Dehaene in the exclusive interview he gives to The Express on the occasion of the release of his new book to Learn ! The talents of the brain, the challenge of the machines*.

However, this is not because the man is already a great “mechanical” that we should be here to stay. The president of the scientific council of the national Education warned against certain errors of the school which, unwittingly, have sometimes a tendency to kill the natural curiosity of the child to inhibit his thirst to learn, or even to confuse the idea of the test (required to fix the knowledge) with that of the sanction (stressful and paralyzing). In the passage, the expert also criticises the current obsession of the opponents of the screens that would not necessarily be so dangerous to the brains of our children. According to him, the main challenges of public health are not where you expect them.

In your book, you mention the learning abilities of the human brain from the very young age. Your findings go against the idea, long taught, that the brain of the new-born would be similar to a blank page ?

The research shows, instead, that the baby is a sort of supercomputer, a statistician of genius, a budding scientist who spends his time doing experiments. Thus, around 10 or 12 months, when he throws from his table of objects, it is in the process of testing the law of gravity, seek to understand why some of them fall, while others remain stable. This theory of the brain, a statistician can resolve the conflict of traditional between the innate and the acquired. The brain of a toddler has a vast space of theories or assumptions possible – that is the part of the innate. Very quickly, he will face to the outside world so as to eliminate those that do not operate and maintain the most adequate – this is the part of the acquis.

are you one of those who were pushing for a compulsory education from the age of 3, which shall enter into force in the new school year 2019 ?

I said that the amazing faculties of the babies unfold well before this age, where the fundamental role of the environment and of the family in the enrichment of its knowledge. But the fact that the kindergarten will become compulsory from 3 years old – this was already the case for 97 % of children – may help to go even further, by proposing activities that parents would not make spontaneously, due to lack of time or equipment. Provided, of course, that the structure of the home is not a simple daycare. We need to be very vigilant on this point.

What counts is not so much the compulsory nature of the tuition and the development of the children. I think a lot, for example, the activities of proto-mathematics, that are to teach them to measure, to sort, to build, to create forms in two or three dimensions.

expose you to lose this scientific capital, we have the birth, if it is not operated ?

Without stimulation, it might wither away or, in any case, not develop as fast. And the early years are sensitive periods for learning. In early life, synaptic plasticity is enormous : several million synapses will make or break every second. But this plasticity disappears quickly. It is obvious in the domain of language : in the course of the first year, we put in place of the phonemes, that is to say, the consonants and vowels of our language. After this period, a French child will not hear all the sounds of the chinese, while a small Japanese will not make the difference between an “l” and “r”.

“At the beginning of life, the synaptic plasticity is enormous.”


In my book, I explain that the brain plasticity freezes the meaning very literal. Neurons are cells extremely arborescent. On these trees, dendrite, there are small spines that won’t stop moving. We can see it clearly under the microscope : they retract, go away, come back over the learning… until the time where they find themselves surrounded by a sort of net stiff enough that the rest. For the learning of a foreign language, the plasticity collapse around puberty.

Why do you not, in your book, the question of screens, today strongly pointing the finger ? They are not dangerous for the brains of toddlers ?

in my opinion, this topic is not part of the basic health, nutrition and the enrichment of its environment. It would be better to highlight other risks, such as the consumption of alcohol by the pregnant woman. In France, 1% of babies in utero are affected, which is both considerable and entirely preventable. Same thing for nutrition : how to ensure that children from disadvantaged backgrounds have access to higher quality food ?

In my book, I recount the dramatic story of these children of Israel who have suffered from a deficiency of thiamine. A few weeks of deprivation, because the milk they received was no longer enriched in vitamin B1, and the brain remained permanently altered. But how many dramatic situations like this are not detected ? These variables fortes on which we can intervene. It is of paramount importance.

Much more than the screens, so…

This obsession is a little absurd. What’s the story there exactly ? Television, of course, engenders a certain passivity and its effects are far from positive. As far as software is concerned, it is different. There are excellent educational games, including for small. From the moment the child interacts, there is learning. The “danger” of the screens – if danger there was – is mainly related to time management. This is the case for those adolescents who are not able to win their video games. Parents exercise their authority to stop a possible use frantic. Any excess is to be monitored, but a lot of studies that show that video games can increase alertness or concentration.

“The ‘danger’ screens – if danger there was – is mainly related to time management.”


You say that you are a scientist and not a policy. However, some of your pronouncements, for example, the defence of the method, syllabic versus the global method, you class inevitably in a camp…

I am only on the research field. What I hold to heart, is to pass on my knowledge to the teachers as to parents. With regard to learning reading, the approach in kindergarten has changed a lot these past fifteen years. According to the statistics of the Directorate for evaluation, forward studies and performance of national Education, the children at the end of grande section, are more alert in phonology, which is one of the keys to the entry in the reading. The scientific studies that other researchers and myself have put forward have contributed a lot. The resistors that remain are resistors ideological.

Some people find it difficult to accept that one has objective data. These weigh more than the opinion of a particular teacher. In the field of reading, the data are even more convincing that they are international. All countries that use the alphabet are the same conclusion : start learning with a method of sound is going faster, including as regards the understanding of the texts, that start with a global or semi-global.

In your book, you criticize some of the “errors” of the education system. What are they ?

I see, for example, that school can kill this hungry curiosity, abounding, amazing, so clean the babies of the human species. Their brain is continually exploring new activity states : it is enough to see them change activity or expression every five or ten seconds. Very quickly, in growing, they start to point the finger at the object of their curiosity. No other primate possesses the faculty of attracting the attention of the other to question him. Alas, when the children enter the school, this curiosity fades often. We can ask ourselves if there is not a causal relationship between the two.

A chapter of my book dissects the brain mechanisms of the curiosity stems from the fact that every discovery we make activates the reward circuitry, the dopamine. But this same circuit, the school can turn it off in rabrouant children and / or distributing bad grades. The school does not always stimulations appropriate to the level of each student. Consequently, some get bored because it is too easy for them, while others are discouraged because they are no longer able to follow.

It should therefore have an education more “a la carte” in the classes ?

of course. The Montessori method, with which my children were born when we were living in the United States, has this wonderful thing that they choose, with the agreement of the teacher, activities in line with their level and their taste. Only the curiosity drives them, even if the teacher guides, in reality, their choice to exercise stimulants.

experiments show that a teaching is too explicit can also help to kill the curiosity : the student concludes that the teacher knows everything better than him and that he only has to wait for him that he chews knowledge. The lectures are really not the best way to proceed. It is necessary, on the contrary, solicit the attention of the child, asking him to participate, disrupt the class to ask questions, make presentations… there’s a thin line to keep from him everything to say, at the risk of losing his curiosity, and not tell him anything, at the risk of falling into another pitfall that I denounce in the book : the pedagogy of discovery.

In this passage on the critical pedagogies of discovery, you cite also that of Montessori that you are defending at the moment….

I speak of this current of thought from Rousseau and which actually runs through Dewey, Decroly, Freinet, Montessori, or Piaget, and that would be that the child discovers by himself what he will learn. There is here a confusion of two ideas : of course, the student should be active, involved – and a great deal of research has shown -, but it would be impossible to discover, only, that mankind has taken centuries to invent. It is therefore necessary that the school provide him with a structured environment.

This has been able to ascertain, in particular, in the field of computer science. Seymour Papert, the pope of the educational informatics, had claimed that it was enough to give a computer to children so that they learn to use it. This idea was extended by Nicholas Negroponte, another computer scientist who has argued that we could revolutionize education by parachuting in a country computers $ 100. Uruguay has taken hold, giving all the children a personal computer. However, the school’s results have rather declined during this period. Why ? Because, without specific pedagogy, the computer is no longer a source of distraction. This pedagogy explicitly computing is sorely lacking in France.

Another pitfall French : we would have too much trend, you say, to confuse a test and a punishment ?

a Lot of research in cognitive psychology showed that test allowed the child to improve its performance. On a beach two hours of work, it is better to devote more than half the time to revise and pass the other half to test it to get a “return on error”. I recommend the system of “flash cards” that are found in the United States, these maps, on which you have the question on one side and the answer on the other. The child asks seriously, and then returns the card and see if it is wrong. Software are also doing this very well. The important thing is to receive an error signal fast, accurate, indispensable to the brain to correct itself.

The notes seem to go in this direction. Why do you think they are often unnecessary ?

A simple 15/20, in the absence of other information, does not say why you are wrong. The notes are often also delayed, a few weeks after the control, and you don’t remember very well why you made such a mistake. I’m not saying that it is useless, but it is not very effective. To this is added an emotional aspect that should be avoided at all costs.

studies in animals have clearly demonstrated that stress and anxiety can completely block the learning process. The neural network is frozen and no longer teach. A stream of psychological research has also proven the dangers of the mindset of so-called “fixiste”, which is to explain to a child : “If you have bad grades in math, it is that you are not gifted for that matter.” A teacher should, rather, adopt the mindset of a progressive, in a restaurant with the error status useful, which is that a signal is informative : “You’ve had a bad mark in mathematics, it means that you have not yet learned this, but you’re going to integrate it gradually.”

You spend a part of your book to sleep. Why is it fundamental for learning ?

I rely on recent findings that show that, while we sleep, the brain is hyperactive and running a learning algorithm. This is the time when our neurons reactivate everything that has been learned in the day. This is crucial, even more so in toddlers. Take the period of learning words : a baby is going to first perceive that the term “dog”, learned in a particular context, refers to that specific animal. Sleep will allow him to secure the information and generalize. The next day, he will realize more easily that this word applies to all dogs.

“While we sleep, the brain is hyperactive.”


Some experiments have shown that it could manipulate the sleep to learn better…

trying, for example, synchronize a sound waves with slow waves that the brain emits during sleep, it is possible to increase the depth. The next morning, the learning process is better consolidated. You can also send during the night smells or sounds which evoke moments of the day devoted to this or that learning. The smell of rose, that students have felt during their course of japanese, would bias the processing of information during sleep to that particular content.

Some cases of hyperactivity may also be due to a simple lack of sleep ?

You must be very careful, because this is only for some children – but the research has shown that these children, victims of sleep apnea, could develop disorders that look like hyperactivity or inattention. By improving their sleep, we could do away with these disorders. Other measures may also be taken. I think in particular to adolescents who are not able to get up early in the morning. This is not the bad will of their own, it is only related to their sleep cycle that shifts with the onset of puberty. Some of the pioneer schools have pushed back an hour to the start of classes, only for this age class. Result : the school performance has improved.

Your title refers to the “challenge” machines. The progress of artificial intelligence are revolutionizing the methods of learning ?

My book explains the advanced phenomenal of artificial intelligence, but also the way in which it differs from the algorithms employed by our brains. When the software AlphaGo has beaten the world champion of the game of go, I saw in it a historic day. Nevertheless, the human brain is still far to be imitated by machines. What they lack is the ability to formulate scientific theories, as we have seen, the tiny babies. The machines remain highly specialized, they are not capable of thinking in multiple domains and to gather their knowledge in a symbolic form.

But things are changing very quickly and I would not be surprised that in ten or fifteen years of age, machines far more advanced see the light of day. All the pillars of human learning that j’épart of vogue in my book (attention, active engagement, return on error, the consolidation) are being modeled by the artificial intelligence.

Some people, very critical of the cognitive sciences, you are failing to take account of social factors. Their answer-you ?

I reject totally the criticism. My personal concern and our main concern, the scientific Council of the ministry of national Education, which I chair, it is the disadvantaged children. Many research show that they are the ones who benefit in the first place of the research and recommendations of interventions inspired by cognitive science, because they are the ones who suffer most from errors of teaching and learning. While children from backgrounds more privileged have the chance to have parents that compensate for the flaws of the system.

Our aim is to fight against inequalities. The scientific Council has, in particular, worked extensively with the national Education to develop assessment tools capable of detecting in CP and in CE1, the needs and progress of each child. These small tests will also provide teachers the means of pedagogical intervention adequate and individualized.

The role of the scientific Council is also helping to move research and discoveries from the lab to the classroom. How are you going to do ?

We only have an advisory role to the ministry of Education. We have, for example, been consulted in the development of the guidelines on reading for teachers. It is not we who have written it, but we’ve brought a critical eye. The training of teachers is also a key point on which we can intervene. We help to design online courses, in the framework of continuous training, or courses which the Schools of higher education of the teaching profession and education could build on. All this without encroaching on the pedagogical freedom of teachers.

The general principles that I explain in my book are compatible with a variety of methods. Once teachers have understood these key principles, which are closely related to the functioning of the brain, they can deploy all their creative teaching. I often say that, for fully exercising his freedom, he must be aware of the consequences of his choices. Otherwise it is not freedom, it is to play heads or tails. Do you want to play heads or tails with the future of our children ? A. H. and A. A.

* Learn ! The talents of the brain, the challenge of machines, by Stanislas Dehaene. Ed. Odile Jacob, 384 p., 22,90 €.


Please enter your comment!
Please enter your name here