Peterdjones comments on Seeing Red: Dissolving Mary's Room and Qualia - Less Wrong

38 Post author: orthonormal 26 May 2011 05:47PM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (152)

You are viewing a single comment's thread. Show more comments above.

Comment author: Peterdjones 31 May 2011 04:18:54PM *  1 point [-]

The Mary’s Room thought experiment explicitly claims that Mary knows every physical fact about the given phenomenon but does at the same time implicitly suggest that some information is missing.

Not exactly. There is a sense in which seeing are red tomato conveys the same information as being told "there is a red tomato here". Nonetheless,there appears to be a difference between the two cases It is not clear whether the difference consists of some missing information,or something else.

Mary was merely able to to dissolve part of human nature by incorporating an algorithmic understanding of it. Mary wasn't able to evoke the dynamic state sequence from the human machine by computing the algorithm

There is nothing to stop Mary computing any algorithms. Humans can run through algorithms in their heads or with pencil and paper. Mary should have no problem since she is stipulated to be a super scientist. What she can't do is run the algorithm on the same hardware. She can only run it in the conscious/verbal part. not the automatic, unconscious, perceptual part.

Understanding something means to assimilate a model of what is to be understood. Understanding something completely means to be able to compute its algorithm, it means to incorporate not just a model of something, its static description, it means to become the algorithm entirely. To understand something completely means to remove its logical uncertainty by computing it.

Understanding something completely is only equivalent to understanding it's algorithm if a) it can be decomposed into a software description and a hardware platform. and b) all the significant work is being done by the software, with the hardware being just a neutral platform. Those conditions are not fulfilled in all cases.

We have seen that Mary actually can run through any algorithm, and if your expectation is that she still doesn't understand what colours look like, that would be a case where the hardware is making a difference.

You might object that Mary won't learn anything from experiencing the algorithm, but if Mary does indeed know everything about a given phenomenon then by definition she also knows how the algorithm feels from the inside.

She doesn't know everything by definition; she knows everything physical by definition. That doesn't tell us whether she would be able to figure out how things seem, phenomenally, from the inside. If we had instances of being able to successfully predict qualia from physics or neurology or whatever, there would be no need for an intuition-based parable like Mary's room. As it is. how much Mary would be able to figure out about her qualia is not something we know in advance: instead, our reaction to the story tells us what we think the answer is.

Comment author: XiXiDu 31 May 2011 06:34:03PM *  0 points [-]

It is not clear whether the difference consists of some missing information,or something else.

I am still at the very beginning of learning math, so maybe I am completely confused here. I do not see how there can be a difference without any difference of information content. How the information are interpreted has a bearing on the information content, because humans are not partly software and partly hardware. Brains are physical, chemical systems. Any difference in the processing of sensory information would have a bearing on the measure of the brains Kolmogorov complexity. Therefore, even if the difference between Mary before and after her retina operation is not due to new sensory information, any difference in how previous information are being processed is equivalent to a difference of the neurological makeup of her brain, and therefore its algorithmic complexity.

Comment author: Peterdjones 02 June 2011 01:34:53AM 1 point [-]

I am still at the very beginning of learning math, so maybe I am completely confused here. I do not see how there can be a difference without any difference of information content.

There can't. But that doesn't mean you can work back from information to medium.

It can hardly be disputed that qualia convey or encapsulate information. Yet the Mary story suggests something rather strange — that, although she has all the (physical) information about how colour works, she gains some extra information when she sees colours for the first time.

Information is something that is copied and transferred from place to place. That being the case, something is always left behing, namely the original basis (or format or medium or physical instantiation) of the information. Consider an epic poem in an oral tradition, that is then written down in manuscript, the text of which is then used in a printed book, which is then transferred to microfilm, which is then made into a CDROM. There is no way someone who has access to the CDROM could work backwards to the previous incarnations.

What is left behind is not just more information. Even if we had complete information about the original manuscript of our poem, we wouldn't have the book itself.