A blind woman who switched personalities and could suddenly see

DID offers us a rare insight into what's possible. When all your life you saw nothing but hills, the concept of a mountain is hard to wrap your mind around. Imagine your brain being able to shut down your pancreas. Then another part (personality) comes forward and your pancreas is suddenly fine. One minute you're diabetic, the other you're fine. This is an example of a mountain we have a hard time accepting.

Source: my father, a clinical psychologist

See Somatization disorder and Conversion disorder.

Put another way, is our personality a part of our identity, or is it merely an extension of us, in the same way an arm or leg is a part of us? If we cut off an arm, we are no less ourselves. If B.T. lost a few personalities, we would agree that B.T. is no less herself. If we lost our own personality, our "ego", would we be no less ourselves? Eastern philosophy says yes. I see the logic, but I'm having trouble accepting it.

I wonder if there were earlier reports of this kind of thing?

"At first, B.T.’s renewed sight was restricted to recognizing whole words in that one identity. If asked, she couldn’t even see the individual letters that made up the words, just the words themselves. But it gradually expanded, first to higher-order visual processes (like reading), then to lower-level ones (like recognizing patterns) until most of her personalities were able to see most of the time."

The mind is really an amazing thing and this combination could bring to light discoveries about how both vision and various mental disorders affect the brain.

If you hear your voice, AND other voices that do NOT belong to you, in your head, BUT, they can NOT control your body, then it's called Schizophrenia.

And if you hear your voice, and hear other voices that do NOT belong to you, and these voices can sometimes take control your body from you, then it's called Multiple Personalities Disorder.

This begs the question. Doctors keep saying the voices that Schizophrenia hear are not real. Yet, they themselves know that the voices they hear are real. As real as their own voice in their head. As real as the voice you and I hear in our head. The only thing is, the voices they hear, they know does not belong to them.

In other word, Schizo has some kind of brain defects where instead of the brain creating only a single voice, a single entity, the brain creates multiple voices. So those voices Schizo hear are not hallucinations, but rather, real voices like your voice in your head right now. Only problem is, those voices has no control over the body, unlike your voice.

However, if a defect happen, where the voices can temporary take over the body, then maybe that's what Multiple Personality Disorder is.

This also make sense why she becomes blind. She takes control of her body, but they take control of her vision.

I have no doubts at all that one can be truly blind for a variety of non-physical or partially physical reasons, I do not doubt that this woman experienced blindness. But this personality/identity stuff is very questionable indeed.

It's still trying out different paths, hence the different personalities. A new vision-related path is in the works, but not ready for production right now.

His ears can hear his lips speak

All the time the needles flick and rock.

No machine can give the kind of stimulation,

Needed to remove his inner block.

Suppose you're attempting to build up a mental(/scientific) model of how the human brain works. You're trying to come up with a set of abstractions and rules for how those abstractions interact, with which you can intuit, without performing a full simulation, the answers to experimental questions. That is, your model should produce concrete predictions: if I prod the brain at A, B will happen; invisible effect C is correlated with visible effect D; nebulous phenomena E and F are well-defined instances of type G. A good model is more like a type system than a simulation; it gives you names and concepts that match (with some loss of fidelity) those used by reality. It lets you write valid programs (explanations) that match programs (physics) that are being executed by reality.

The report that a person had multiple personalities and some of them were blind, leads to the logical 'program':

Either

(this person has a brain structure that was so bizarre it isn't related to the rest of the species' brains) [assume this is false]

Or

(it is possible for humans to contain multiple personalities which are not equally able to interact with sensory organs)

Therefore =>

(it is possible for human brains to contain multiple personalities)

(sensory organs are wired into personalities)

(personalities can lose their wiring to sensory organs without physical damage to the organs)

That is, a single report cuts out a huge swathe of model-space. With a single example we are obliged to include in EVERY model concepts of personalities (or something, er, homomorphic to them), and a concept of a wiring to the sensory organs, and the ability to express that this wiring can become broken.

A perfect model must explain every example (or leave a place for as-yet-undiscovered mechanisms to explain them). New examples put new burdens on every existing explanation. They cut model-space down by, presumably, percentage chunks ("excise the subspace of model-space that doesn't model personalities and wiring personalities to sensory organs"). If you imagine that there are something like 2^N possible models at a given granularity, then, the cuts are approximately reducing us to 2^(N-1) each time. Our models with N degrees of freedom that has to explain E experiments has more like N-E degrees of freedom (disregarded 'covariance' between experimental results). What this means is that a random model is overwhelmingly likely to be wrong with each successive experiment. (In practice our models have large dark spots where we leave many degrees of freedom open - but even still, it's likely that any explicit modeling choice we make ends up being incorrect, and quickly.)

(This math is pretty made-up. We could do better with something akin to Kolgomorov complexity rather than degrees of freedom, since type systems should be discounted by how complex they are so that systems more complex than reality actually is end up contributing nothing to the total.)

Summary: I enjoy stories about weird neuroscience because they defy my mental model of "how brains work" almost every time, even if I keep revising it.