Mary's Room: Has the Answer Been in Her Textbook All Along?

Thought Experiments [Part 5]

Mary is a brilliant scientist.

(You’ve probably heard this one before.)

Mary (poor Mary) has spent her entire life trapped in a black-and-white room. But despite this limitation, she has (apparently) learned all the physical information there is to know about colour. She gained all this knowledge by watching black-and-white television and reading black-and-white books. She knows ALL the physical facts about colour vision — all of it! From the wavelengths of light to the precise neural activity patterns they trigger.

But Mary has never seen colour.

The question — the one that philosophers have been quibbling over for decades — and the one that we’ll try to answer in this article is:

‘What will happen when Mary is released from her black and white room or is given a colour television monitor? Will she learn anything or not?’

Let’s keep this question in mind.

The Mary’s Room thought experiment was created by Frank Jackson. Early in his career, Jackson used this thought experiment to argue that physicalism — the idea that consciousness can be fully explained by the physical — is false.

Jackson answers his question by claiming that:

‘it seems just obvious that she will learn something about the world and our visual experience of it’.

With the question answered — Mary would learn something new — he continues with his argument:

‘But then it is inescapable that her previous knowledge was incomplete. But she had all the physical information. Ergo there is more to have than that, and Physicalism is false.’

Jackson named his argument The Knowledge Argument.

Interestingly, later in his career, Jackson rejected his own argument. He now embraces representationalism, a form of physicalism.

But for many, the intuition that Mary will learn something new is strong. We imagine Mary finally stepping out of her black-and-white room and seeing a red flower or a red stop sign and saying something like, ‘Oh! That’s what red looks like.’

Fitting with this intuition, much of the debate about Mary’s Room centres around what it is (exactly) that Mary learns. And whether what she learns can tell us anything interesting about consciousness.

But in all this philosophical back-and-forth about the importance (or irrelevance) of what Mary learns, have we missed something crucial? On the question of what Mary will learn, does science already have the answer?

To find out, let’s first examine the philosophical arguments that have dominated this debate for decades and then turn our attention to what science might tell us about Mary’s fate.

To do that, we’ll ask three key questions:

1. Why has Mary’s Room been so convincing?

2. What are the main objections to the Knowledge Argument? and

3. What does science tell us will happen to Mary?

Q1: Why Has Mary’s Room Been So Convincing?

The knowledge argument seems deceptively simple.

But to understand why this argument has been so influential, we need to look more carefully at how it’s constructed. The argument rests on two key moves and two key assumptions.

The first move Jackson makes is to define physicalism in terms of information. Defining it this way means that if physicalism is true, then knowing all the physical information means knowing everything there is to know. A completed science would tell us about the causal and relational facts and the consequences of those facts.

The second move is to distinguish physical information from phenomenal information. Phenomenal information, according to early Jackson, is information about what experiences feel like from the inside — the subjective, qualitative aspects of experience that can’t be described as physical information.

The crucial claim is that these two types of information are independent — gaining all the physical information will not give you phenomenal information. No amount of knowledge about wavelengths, neural activity, or behavioural responses will tell you what it’s like to see red if you’ve never experienced it.

The first key assumption that Jackson has us imagine is that it is possible that Mary could know everything there is to know about colour. And the second assumption is that despite this complete knowledge, Mary will experience an aha! moment immediately upon seeing colour for the first time.

Once we accept Jackson’s conception of information, his distinction between physical and phenomenal information, the assumption that Mary could know everything and the assumption that she will have an aha! moment, the knowledge argument follows straightforwardly. Mary gains new (phenomenal) information when she sees colour despite having all the physical information. Therefore, there must be more to know than just physical information. So, physicalism must be false.

The argument’s force comes from how reasonable these premises can seem. After all, isn’t there something about the experience of seeing red that can’t be captured in even the most detailed physical description? Isn’t it obvious that Mary will learn something new? But as we’ll see in the next question, not everyone accepts these seemingly intuitive moves and assumptions.

Q2: What Are the Main Objections to the Knowledge Argument?

There are four common objections to Jackson’s Knowledge Argument. These objections fall into two main camps — those who accept Mary will learn something new (the first three objections), and those who deny she will learn anything new (the fourth objection).

Let’s briefly review.

1. The Physical Information Objection

The first objection challenges Jackson’s definition of information. While these critics might accept that Mary could learn all the information that can be expressed in objective, scientific language, they deny that this covers all physical information. There is simply more physical information than that which can be gained through a television monitor or expressed in purely objective terms.

If you want to read more about this argument, I recommend the book Consciousness and the Limits of Objectivity: The Case for Subjective Physicalism by Robert Howell and the paper Jackson on Physical Information and Qualia by Terence Horgan.

2. The Ability Argument

Developed by David Lewis and Laurence Nemirow (and later embraced by Jackson himself), this objection distinguishes between knowledge-that (propositional knowledge) and knowledge-how (practical knowledge). They argue that when Mary leaves the room, she gains new abilities — like the ability to imagine, recognise, and remember colour experiences — rather than new factual knowledge. According to this view, knowing what it’s like to see red is not about acquiring new information — it’s about gaining new abilities.

I discussed this objection a little in my article on Nagel.

3. The Phenomenal Concepts Response

Philosophers like Brian Loar and Barbara Montero argue that Mary gains new concepts rather than new information when she leaves the room. Known as the phenomenal concept strategy, this view holds that seeing colour provides Mary with new conceptual resources but not new facts about the world.

4. Dennett's Skepticism

Daniel Dennett directly challenges the intuition that Mary learns anything new when she leaves the room. He argues that if Mary truly knew all the physical facts, she would know exactly what seeing colour would be like. Our failure to accept that she would know everything simply highlights how difficult it is for us to grasp what it means to ‘know everything’.

To illustrate this, he presents a thought experiment where Mary is shown a blue banana. As a brilliant scientist, she would not be fooled into thinking it was yellow, as she would understand the wavelengths of light being reflected and their effects on the human visual system.

---

Perhaps unsurprisingly, none of the above arguments have been universally accepted as knock-down proof against Jackson’s argument. (We do like to debate these things, don’t we!?)

So, to solve this riddle, do we need to find a newborn to lock up in a black-and-white room? (If only ethics would allow it!)

Of course, even if we could lock up a baby, we’d still face the challenge of making our subject as brilliant as Mary — which seems like an impossible task. What would all the physical information look like, anyhow? And could a human brain, even a brilliant one, really contain all of it?

While we can’t exactly replicate Mary’s case, we can look to science to shed some light (or colour) on what Mary would learn when she is released from her room.

Q3: What does Science tell us About Mary?

Before we jump into the scientific evidence, we need to ask an important question: is empirical evidence even relevant to this thought experiment?

You might be wondering — why look at scientific evidence for a thought experiment? After all, thought experiments shouldn’t be taken literally — they’re meant to be imagined scenarios that don’t exist in our world.

But thought experiments can serve different purposes.

Some thought experiments are purely about logic. Consider the classic philosophical puzzle: Could an all-powerful being create a stone so heavy they couldn’t lift it? The question exposes a logical contradiction in the concept of an all-powerful being. If the all-powerful being can’t create such a stone, it’s not an all-powerful being, but if it can’t lift it, it’s not all-powerful either. We don’t need to know all the empirical evidence about stones or power to see the logical problem.

Other thought experiments, like Mary’s Room, make claims about what is possible in our actual world. When a thought experiment makes claims about our world, it needs to be both logically possible and physically possible in our world. For these types of thought experiments, empirical evidence matters greatly.

So, let’s review the evidence.

While we don’t have evidence of babies being raised in black-and-white rooms, we do have evidence of animals being raised in similar sensory-deprived situations.

In 2004, a paper published in Current Biology described an experiment where baby monkeys were raised in a room without colour. It wasn’t technically a black-and-white room — but it was similar enough to Mary’s situation to be interesting. To ensure each cell type in the monkey’s retina was stimulated, the room was illuminated by monochromatic lights that changed wavelength randomly every minute.

While in their colour-deprived room, these ‘Mary’ monkeys learned to play a game. They were shown three squares — two in the top row and one in the bottom row. The monkey’s task was to find the square in the top row that matched the brightness of the square in the bottom row. The monkeys got very good at playing this game.

Schematic illustration of stimuli. The monkey’s task was to choose a square in the top row that matched the square in the bottom row.

After their first birthday, the monkeys were moved from their colour-deprivation room to a room with natural light. There, the researchers tested them with a similar matching game. But instead of matching brightness like before, now they had to match the hue. To make sure they were truly matching hue (and not just brightness or saturation), the researchers kept brightness and saturation constant — only the hue changed.

So, did these Mary monkeys learn anything or not?

Well, they did eventually learn to match the hue correctly. But it took them over 31 days and fifty-five thousand trials to do it like a normal monkey.

Average correct responses were plotted against training days. Open circles represent the performance of normally reared monkeys, while closed circles represent that of color-deprived monkeys. 1 denotes the day when monkeys began the color-matching task.

Sadly, even though the Mary monkeys eventually learned to perform this task, other tasks showed they didn’t ever develop colour vision like a normal monkey.

Normal colour vision is not about seeing the raw wavelengths of light. We see a red rose as red, whether it’s in sunlight, fluorescent light, or shadow. The Mary monkeys never gained this ability — to them, the rose would change colour with different lighting because they perceived raw wavelengths rather than stable object colours.

And when asked to group similar colours, the Mary monkeys created entirely different groupings than normal monkeys or humans would. These weren’t temporary problems — months later, they still saw colour in this unusual way.

These findings suggest something important about how brains develop. Just like a soccer player’s body changes with practice — developing stronger muscles, faster reflexes, and more precise motor control — the brain also physically changes through experience. The physical brain structures needed for normal colour vision develop through experience.

Without the right training, a soccer player will never develop the body required to play soccer. Likewise, without the right visual input during development, the brain will not develop the brain structures required to see colour.

While we can’t ethically conduct colour deprivation studies in humans, we do have some relevant evidence from natural experiments in humans.

For example, research from MIT looked at what happens to children’s vision when they are born blind due to cataracts and have those cataracts removed later in childhood or adolescence.

Just like the Mary monkeys, these children did not immediately gain normal vision when their cataracts were removed. The development of visual abilities requires time and experience. Some aspects of vision, like recognising faces, perceiving depth, and tracking moving objects, were never learned.

Even in children with normal vision, colour development is gradual — while basic colour vision develops by age 7, the ability to detect subtle differences in colour saturation does not reach adult levels until late adolescence.

Scientists have discovered something surprising about colour vision — it might be uniquely resilient compared to other visual abilities. When children who were born blind have their sight restored, they struggle with many aspects of vision, but colour vision might be special. Some evidence suggests that if treatment happens before adulthood, the brain may be capable of developing normal colour perception.

But in Mary’s case, things could be different. She’s an adult whose eyes have never been exposed to colour; she may have missed important brain development required for colour vision.

Despite the uncertainty of Mary’s fate — the key point remains: normal colour vision must be learned through experience.

Mary wouldn’t have that immediate aha! moment that Jackson’s thought experiment assumes. She may never experience colour — but if she did, her understanding of colour would develop gradually through experience. This experience would cause physical changes, just like any other physical skill.

So, what does this evidence mean for Jackson’s thought experiment?

The irony is that Mary herself — who knows everything there is to know about colour vision — would know all of this. She understands that when she leaves her room, she will not experience colour like a normal adult. She knows about critical periods, about how the brain develops through experience, about the gradual nature of learning to see.

But this raises an interesting question about Mary’s situation.

If Mary’s brain requires specific neural structures to process colour normally, and those structures can only be developed through experience, could she really have all the physical information while lacking these physical structures?

Would we claim Mary could know everything there is to know about colour if she were colour-blind? What if she had damage to the brain areas that are normally active during colour perception?

Mary’s Room was intended to expose what science can’t explain. But maybe the more interesting thing it shows is how complex learning and knowledge is. Can we know something without experiencing it? Can we learn something without our brains changing? And what exactly is information anyway — how do we get it, and how do we keep it?

Next week…

Let’s explore different types of learning and memory. What’s the difference between understanding something intellectually, experiencing it firsthand, and learning it through practice?

Thank you.

I want to take a small moment to thank the lovely folks who have reached out to say hello and joined the conversation here on Substack.

If you’d like to do that, too, you can leave a comment, email me, or send me a direct message. I’d love to hear from you. If reaching out is not your thing, I completely understand. Of course, liking the article and subscribing to the newsletter also help the newsletter grow.

If you would like to support my work in more tangible ways, you can do that in two ways:

• You can become a paid subscriber

• or you can support my coffee addiction through the “buy me a coffee” platform.

I want to personally thank those of you who have decided to financially support my work. Your support means the world to me. It’s supporters like you who make my work possible. So thank you.