Most answers here seem to be about the neuroscience and psychology of color perception and measurable behavioral differences (across gender or language for instance).

This is not the question as it is framed in the philosophy of mind.

To be more precise, the inverted spectrum thought experiment has no relation to neuroscience or psychology. It is a metaphysical question about whether subjective experience is ontologically distinct from the physical correlates of that experiences.

Think of it this way: two people, A and B, both look at blue and red objects and identify them the same way -- as blue and red. This does not prove that their subjective experience is the same.

They merely have the same input/output behavior (or, if you stick 'em in an MRI machine, the same observable correlate of internal state which is not the same as internal state as-experienced-by-subject). A may actually be subjectively perceiving what he is calling "blue" the way the other person perceives "red" and vice versa. Their answers and fMRI patterns match because they've learned different names for their different subjective perceptions of the same object, triggering roughly the same neural firing patterns. A's name for his "red" perception is "blue." B's name for his blue perception is "blue." It's a double negative kind of logic.

• A: (object X+fMRI pattern X)("blue" subjective perception)(blue perception is mapped to word "red"): "This is red"
• B: (object X+fMRI pattern X)("red" subjective perception)(red perception is mapped to word "red"): "This is red."

Yes, this sounds complicated, but is not really. It depends on whether or not you consider "what it feels like" to be ontologically separate from the electrical signals shooting around etc.

The reason this is not a neuroscience question is that there is no way for you to get inside my head and see what "blue" feels like to me. You can talk to me, you can shove all sorts of instruments into my head, you can do anything from the outside. But you cannot experience what it feels like for me to be me.

This "what it feels like to me" abstraction is often called a quale (singular of qualia). Some people think qualia are nonsense, others think they are the most real things there are. In fact the only things. The only difference between a robot with a light sensor that can detect and compute with "blue" and me is that there seem to be blue qualia going on inside my head. There is no reason to suspect that there are blue qualia for the robot, because there is no reason to suspect that the robot even has an "I."

http://en.wikipedia.org/wiki/Qualia

This is the anti-Strong-AI position.

I am not attempting to resolve this question of course. It is one of the trickiest things to think about in the philosophy of mind, and heavy-hitter minds like Schrodinger and Dennett have squared off about it and reached no consensus.

I just wanted to point out that there is a view on this question within which the neuroscience and psychology are basically irrelevant.

For practical purposes I do believe that you and I probably see "blue" roughly the same way (modulo some cultural, gender and individual differences). I don't see "red" where you see "blue." But this is an assumption based on Ockham's razor. NOT the scientific Ockham's razor, but the metaphysical one, since this is not about the simplest explanation of objective data. It is about the simplest explanation of subjective experience.

It is in the same class of beliefs as atheism. The inverted spectrum conjecture cannot be proven wrong, just like God/Flying Spaghetti Monster, so the only reason to believe everyone sees "blue" roughly the same way is because there is no reason to posit anything more complicated, other than to demonstrate that there are gaps in strong AI reasoning having to do with conflating subjective experience with observable correlates of that experience.