El misterio de la conciencia: lo que sabemos y lo que ignoramos

So when neuroscience really starts taking consciousness seriously as a research program, that's relatively recent.

Francis Crick, co-discoverer of DNA, spends the last decades of his life on this.

He teams up with Christof Koch and they start looking for what they call the neural correlates of consciousness.

Mira, Crick y Koch hicieron algo crucial: decidieron dejar de intentar resolver el problema filosófico completo y centrarse en algo más modesto pero más manejable.

Look, Crick and Koch did something crucial: they decided to stop trying to solve the full philosophical problem and focus on something more modest but more manageable.

¿Qué actividad neuronal específica corresponde a una experiencia consciente específica?

What specific neural activity corresponds to a specific conscious experience?

Si estoy viendo el color rojo, ¿qué está pasando exactamente en mi cerebro en ese preciso momento?

If I'm seeing the color red, what is happening exactly in my brain at that precise moment?

No responde el problema difícil, pero construye una base científica sólida.

It doesn't answer the hard problem, but it builds a solid scientific foundation.

The binding problem, trying to understand how all these separate bits of processing, color, shape, motion, get stitched together into one unified experience.

I'm told Koch spent years staring at single neurons in the visual system trying to find the ones that 'decide' what you're consciously seeing.

La verdad es que fue una línea de investigación enormemente fructífera.

It was an enormously fruitful line of research.

Pero a medida que avanzaba, quedaba claro que la conciencia no residía en ningún lugar específico.

But as it progressed, it became clear that consciousness didn't reside in any specific place.

No había una 'neurona de la conciencia'.

There was no 'consciousness neuron'.

Lo que empezó a emerger era algo mucho más interesante: la conciencia parecía depender de patrones de comunicación global entre distintas regiones del cerebro, no de una zona concreta.

What began to emerge was something far more interesting: consciousness seemed to depend on global patterns of communication between different brain regions, not on one specific area.

Which is where the theories start to diverge and get genuinely fascinating.

There are now two main competing scientific theories, and they make very different predictions about what consciousness actually is.

Walk me through them.

A ver, la primera es la Teoría del Espacio de Trabajo Global, propuesta originalmente por Bernard Baars y desarrollada científicamente por Stanislas Dehaene.

So, the first is Global Workspace Theory, originally proposed by Bernard Baars and developed scientifically by Stanislas Dehaene.

La idea central es que la conciencia surge cuando la información se transmite de forma masiva por todo el cerebro, cuando pasa de ser local a ser global.

The central idea is that consciousness arises when information is broadcast massively across the whole brain, when it goes from being local to being global.

Imagina que el cerebro tiene muchos procesadores especializados trabajando en paralelo, y la conciencia es el momento en que uno de esos procesos gana la competición y se emite a todas las demás regiones.

Imagine the brain has many specialized processors working in parallel, and consciousness is the moment when one of those processes wins the competition and gets broadcast to all the other regions.

That has a kind of intuitive appeal.

It maps onto the experience of something suddenly coming into focus, clicking into place.

And the second theory?

La segunda es la Teoría de la Información Integrada, de Giulio Tononi, y es mucho más radical y, diría yo, mucho más perturbadora en sus implicaciones.

The second is Integrated Information Theory, by Giulio Tononi, and it is far more radical and, I would say, far more disturbing in its implications.

Tononi propone que la conciencia no es un tipo de proceso sino una propiedad intrínseca de cualquier sistema que integre información de cierta manera.

Tononi proposes that consciousness is not a type of process but an intrinsic property of any system that integrates information in a certain way.

Y la cuantifica con una medida matemática que llama phi, la letra griega.

And he quantifies it with a mathematical measure he calls phi, the Greek letter.

So phi is essentially a measure of how much a system is more than the sum of its parts, how much information is generated by the whole that can't be reduced to the parts separately.

High phi equals high consciousness.

Phi equals zero, you're not conscious at all.

Exactamente.

Exactly.

Y aquí viene lo que hace que la teoría sea tan incómoda para mucha gente: si phi es lo que define la conciencia, entonces potencialmente cualquier sistema con suficiente integración de información podría ser consciente en algún grado.

And here is what makes the theory so uncomfortable for many people: if phi is what defines consciousness, then potentially any system with sufficient information integration could be conscious, to some degree.

No solo los humanos.

Not just humans.

No solo los animales.

Not just animals.

Teóricamente, ciertas redes podrían tener algo parecido a la experiencia subjetiva.

Theoretically, certain networks could have something resembling subjective experience.

That's panpsychism.

The idea that consciousness is a fundamental feature of the universe, like mass or charge, present to some degree in everything.

Which sounds mystical but is actually taken seriously by serious philosophers.

Bueno, el panpsiquismo tiene una historia filosófica respetable, desde Leibniz hasta Whitehead, y hay filósofos como Galen Strawson o Philip Goff hoy en día que lo defienden con argumentos rigurosos.

Well, panpsychism has a respectable philosophical history, from Leibniz to Whitehead, and philosophers like Galen Strawson and Philip Goff today defend it with rigorous arguments.

No es misticismo, es una posición filosófica.

It's not mysticism, it's a philosophical position.

Pero también genera objeciones enormes, sobre todo el llamado 'problema de la combinación': si las partículas individuales tienen micro-experiencias, ¿cómo se combinan para formar la experiencia unificada de un ser humano?

But it also generates enormous objections, especially the so-called 'combination problem': if individual particles have micro-experiences, how do they combine to form the unified experience of a human being?

The extraordinary thing is that these two main theories, Global Workspace and IIT, ended up in something almost unprecedented in neuroscience.

A formal, pre-registered adversarial collaboration.

Scientists on both sides agreed in advance on the experiments and what results would count as evidence for each theory.

Sí, y fue fascinante y un poco frustrante a la vez.

Yes, and it was fascinating and a little frustrating at the same time.

Los resultados, publicados en 2023, fueron ambiguos.

The results, published in 2023, were ambiguous.

Ninguna teoría salió claramente ganadora.

Neither theory emerged as a clear winner.

Encontraron evidencia mixta y, lo que es más revelador, la colaboración terminó con desacuerdo sobre cómo interpretar los propios datos.

They found mixed evidence, and more tellingly, the collaboration ended with disagreement about how to interpret the data itself.

Lo cual, en un sentido, es el proceso científico funcionando como debe, pero también demuestra lo lejos que estamos de resolver esto.

Which, in one sense, is the scientific process working as it should, but it also shows how far we are from resolving this.

There's a third framework I want to bring in, predictive processing, Karl Friston's work, because it approaches consciousness from a completely different angle.

Es que Friston propone algo que me parece elegante aunque brutalmente difícil de comprender en detalle: que el cerebro no es principalmente un receptor pasivo de información sino una máquina de hacer predicciones.

Friston proposes something I find elegant though brutally difficult to understand in detail: that the brain is not primarily a passive receiver of information but a prediction machine.

Constantemente genera modelos del mundo y solo procesa los errores de predicción, las discrepancias entre lo que esperaba y lo que recibe.

It constantly generates models of the world and only processes prediction errors, the discrepancies between what it expected and what it receives.

Y la conciencia, en este marco, es algo así como el modelo que el cerebro construye de sí mismo.

And consciousness, in this framework, is something like the model the brain builds of itself.

So consciousness as a kind of controlled hallucination.

The brain's best guess at what's out there, not a direct readout of reality.

Which is humbling.

I spent twenty-five years believing I was reporting on what was actually happening.

La verdad es que todos lo creemos.

We all believe it.

[laughs] Pero hay implicaciones prácticas muy serias que emergen de toda esta investigación teórica.

[laughs] But there are very serious practical implications that emerge from all this theoretical research.

Una de las más importantes tiene que ver con los animales.

One of the most important concerns animals.

En 2012, un grupo de neurocientíficos prominentes firmó la Declaración de Cambridge sobre la Conciencia, declarando formalmente que los mamíferos, las aves y muchos otros animales poseen los sustratos neurológicos necesarios para estados conscientes.

In 2012, a group of prominent neuroscientists signed the Cambridge Declaration on Consciousness, formally declaring that mammals, birds, and many other animals possess the neurological substrates necessary for conscious states.

Which, on the one hand, seems obvious to anyone who's ever had a dog.

But having it stated formally by scientists, signed at a conference with Stephen Hawking present, that's different.

That has ethical weight.

Tiene un peso enorme, sí.

It carries enormous weight, yes.

Y la investigación reciente sobre los pulpos me parece especialmente fascinante porque su sistema nervioso es radicalmente distinto al nuestro.

And recent research on octopuses I find especially fascinating because their nervous system is radically different from ours.

La mayoría de sus neuronas están en los tentáculos, no en el cerebro central.

Most of their neurons are in their tentacles, not in the central brain.

Si son conscientes, y hay evidencia bastante convincente de que tienen algo parecido a la experiencia subjetiva, entonces la conciencia ha evolucionado de forma completamente independiente al menos dos veces.

If they are conscious, and there is fairly compelling evidence they have something resembling subjective experience, then consciousness has evolved completely independently at least twice.

Lo cual sugiere que es una solución que la naturaleza encuentra una y otra vez.

Which suggests it is a solution that nature finds again and again.

And then the obvious question, the one everyone is asking right now, is AI.

Is a large language model conscious?

Is there something it's like to be ChatGPT?

And I know what you're about to say.

Mira, antes de responder lo que crees que voy a responder, déjame ser preciso.

Look, before I answer what you think I'm going to say, let me be precise.

La pregunta no es si los modelos actuales son conscientes, porque casi con certeza no lo son.

The question is not whether current models are conscious, because they almost certainly are not.

La pregunta es si existe en principio alguna razón por la que un sistema de procesamiento de información suficientemente complejo no podría serlo.

The question is whether there is in principle any reason why a sufficiently complex information processing system could not be.

Y ahí la respuesta honesta es: no lo sabemos.

And there the honest answer is: we don't know.

Lo que sí sabemos es que el hecho de que algo parezca consciente, de que hable de sus 'sentimientos', no es evidencia de nada.

What we do know is that the fact that something seems conscious, that it talks about its 'feelings', is not evidence of anything.

The philosophical zombie problem.

A being that behaves exactly like a conscious creature but has no inner experience whatsoever.

You can't tell from the outside.

Which is also, by the way, the problem with other humans.

You can't actually verify that anyone else is conscious.

Es que es uno de los argumentos filosóficos más perturbadores que existen: el solipsismo no se puede refutar de forma estrictamente lógica.

It's one of the most disturbing philosophical arguments that exists: solipsism cannot be strictly logically refuted.

Solo tienes acceso directo a tu propia conciencia.

You only have direct access to your own consciousness.

Y esto tiene implicaciones clínicas devastadoras que me parecen las más urgentes de toda esta investigación.

And this has devastating clinical implications that I think are the most urgent of all this research.

Estoy pensando en el trabajo de Adrian Owen con pacientes en estado vegetativo.

I'm thinking of Adrian Owen's work with patients in vegetative states.

This stopped me completely when I first read about it.

Owen, a British neuroscientist, in 2006 takes a patient who's been diagnosed as vegetative, completely unresponsive, and puts her in an fMRI scanner and asks her to imagine playing tennis.

And her brain lights up exactly the same as a healthy person's brain.

She is in there.

She can hear.

She just can't move.

Y después Owen desarrolló una técnica para que pacientes en ese estado pudieran responder preguntas de sí o no simplemente imaginando dos cosas distintas.

And then Owen developed a technique for patients in that state to answer yes or no questions simply by imagining two different things.

Imaginar tenis era 'sí', imaginar caminar por tu casa era 'no'.

Imagining tennis was 'yes', imagining walking through your house was 'no'.

Y cuando le preguntaron a un paciente si sentía dolor, respondió que no.

And when one patient was asked if he was in pain, he responded that he was not.

Piensa en lo que eso significa.

Think about what that means.

Estábamos tomando decisiones médicas, a veces decisiones sobre si continuar el tratamiento, sin saber que había alguien consciente ahí dentro.

We were making medical decisions, sometimes decisions about whether to continue treatment, without knowing there was someone conscious inside.

That's not a philosophical puzzle anymore.

That's a person.

The history of medicine is full of things we now know were catastrophically wrong, and consciousness is the frontier where the stakes are the absolute highest.

Completamente de acuerdo.

I completely agree.

Y por eso me parece que la pregunta no es solo científica ni filosófica sino profundamente ética y política.

And that's why I think the question is not only scientific or philosophical but deeply ethical and political.

Cuanto más entendemos sobre la conciencia, más obligaciones adquirimos: con los pacientes, con los animales y, eventualmente, si la inteligencia artificial sigue avanzando, con cualquier sistema que pudiera albergar algo parecido a una experiencia interior.

The more we understand about consciousness, the more obligations we acquire: toward patients, toward animals, and eventually, if artificial intelligence keeps advancing, toward any system that might harbor something resembling an inner experience.

Ignorar estas preguntas porque son difíciles sería una irresponsabilidad enorme.

Ignoring these questions because they are difficult would be an enormous irresponsibility.

So where does that leave us.

I mean, we've gone from Descartes' pineal gland to fMRI scanners and adversarial collaborations and patients answering questions by imagining tennis.

And the honest answer is still that we don't know what consciousness is.

We're better at measuring the edges of it than understanding the center.

Hay una idea del filósofo Colin McGinn que me parece muy honesta: puede que la conciencia sea un misterio que la mente humana no tenga la capacidad de resolver, de la misma manera que un perro no puede resolver una ecuación diferencial.

There's an idea from the philosopher Colin McGinn that I find very honest: consciousness may be a mystery that the human mind lacks the capacity to solve, in the same way that a dog cannot solve a differential equation.

No es que la respuesta no exista;

It's not that the answer doesn't exist;

es que quizás nuestros cerebros evolucionaron para sobrevivir en la sabana africana, no para comprender su propia naturaleza más profunda.

it's that perhaps our brains evolved to survive on the African savanna, not to understand their own deepest nature.

Y si eso te parece frustrante o te parece hermoso depende enteramente de quién seas.

And whether you find that frustrating or beautiful depends entirely on who you are.