3 ms·

The article touches on it. This TEDx talk goes into more depth below the "gradient" level: https://www.youtube.com/watch?v=cF2dhWWUyQ4 https://www.youtube.com/w

by _y5hn 5y ago

The article touches on it. This TEDx talk goes into more depth below the "gradient" level: https://www.youtube.com/watch?v=cF2dhWWUyQ4 https://www.youtube.com/watch?v=cF2dhWWUyQ4 I won't claim to be an expert how it differentiates across "groups" of people. I do care a bit about keeping an open mind and try to see multiple perspectives at the same time though. But yes, that often makes it more general and less powerful (one-sided).

mannykannot 5y ago
The article here is all about arguing against what you have been saying: its author argues that the autism spectrum is not "a typical spectrum [that] is the exact same thing as gradients of frequencies" - or a gradient of anything else, for that matter. This is, in fact, the main point of the article! Carrie Beckwith-Fellows' TED talk does not address this issue at all: she makes an analogy to the color spectrum, but there is no indication that this analogy extends to there being an underlying gradient analogous to frequency. Both author and speaker argue that even the conflation of the somewhat pop-sci gradient of high/low functioning with the spectrum is a misunderstanding! I think the important message from both people here is not that there is no gradient - of "functioning", stress, or something else - but that it is / they are largely orthogonal to the dimension plotted out by the spectrum's categorical values.
- _y5hn 5y ago
  At this point, I think we're just talking beside each other's points. My first post, as a reply to other posts, is just about spectrum and gradients, really being the same, but there's a duality between the continuous frequencies/gradients and the perceptions of colors. I'm being general, whereas you are pointing out problems with too general models (or aggregates for that matter). I don't disagree with that. To attempt to reply though: There doesn't need to be a psychological variable, there just need to be made one or more tests that scores aggregates. However, as the article and TEDx talk says, a number or two would not paint the whole picture of a complex socio-psychological issue.
  - mannykannot 5y ago
    My very first reply to you was to point out that your generality - “a typical spectrum is the exact same thing as gradients of frequencies” - does not appear to apply in this case, and I am glad to see that we now seem to be in agreement on this. Your second paragraph appears to be broadly in agreement with my point (which is merely a paraphrase of the author, believe) that, in this case, the gradients of quantitative test scores are largely orthogonal to the spectrum’s categorical variables. Even if one could combine them to get a meaningful aggregate score (which I suspect you can do, though only to some extent), you cannot tell from that score what aspects of the spectrum describe the subject.
    - _y5hn 5y ago
      For sure. Note, I never disagreed with the author, and even provided links of my own. Not every post is a disagreement, at least not from me ;) From the TEDx-talk, it does seem like the typical spectrum/gradient understanding is lacking in terms of being meaningful for individuals. Maybe because it's a sort of score as seen from the outside, and not even very precise one at that. Also, society also needs to accommodate and improve understanding of diversity, rather than expecting everyone to be clones of each other. The latter doesn't work that well, since we're then not encouraging people's natural strengths and creativity.
      - mannykannot 5y ago
        > Note, I never disagreed with the author… Well, it was certainly hard to see it any other way: Author: The spectrum is not a gradient. From your first reply: A typical spectrum is the exact same thing as gradients of frequencies. As your reply did not explicitly make this claim about the autism spectrum, it nominally did not directly contradict the author, but if it was not about the autism spectrum, then it was utterly pointless to raise it here - and, furthermore, it took several promptings on my part before you addressed this issue. With regard to your second paragraph, I completely agree.
        _y5hn 5y ago

by _y5hn 5y ago

mannykannot 5y ago

The article here is all about arguing against what you have been saying: its author argues that the autism spectrum is not "a typical spectrum [that] is the exact same thing as gradients of frequencies" - or a gradient of anything else, for that matter. This is, in fact, the main point of the article! Carrie Beckwith-Fellows' TED talk does not address this issue at all: she makes an analogy to the color spectrum, but there is no indication that this analogy extends to there being an underlying gradient analogous to frequency. Both author and speaker argue that even the conflation of the somewhat pop-sci gradient of high/low functioning with the spectrum is a misunderstanding! I think the important message from both people here is not that there is no gradient - of "functioning", stress, or something else - but that it is / they are largely orthogonal to the dimension plotted out by the spectrum's categorical values.

_y5hn 5y ago
At this point, I think we're just talking beside each other's points. My first post, as a reply to other posts, is just about spectrum and gradients, really being the same, but there's a duality between the continuous frequencies/gradients and the perceptions of colors. I'm being general, whereas you are pointing out problems with too general models (or aggregates for that matter). I don't disagree with that. To attempt to reply though: There doesn't need to be a psychological variable, there just need to be made one or more tests that scores aggregates. However, as the article and TEDx talk says, a number or two would not paint the whole picture of a complex socio-psychological issue.
- mannykannot 5y ago
  My very first reply to you was to point out that your generality - “a typical spectrum is the exact same thing as gradients of frequencies” - does not appear to apply in this case, and I am glad to see that we now seem to be in agreement on this. Your second paragraph appears to be broadly in agreement with my point (which is merely a paraphrase of the author, believe) that, in this case, the gradients of quantitative test scores are largely orthogonal to the spectrum’s categorical variables. Even if one could combine them to get a meaningful aggregate score (which I suspect you can do, though only to some extent), you cannot tell from that score what aspects of the spectrum describe the subject.
  - _y5hn 5y ago
    For sure. Note, I never disagreed with the author, and even provided links of my own. Not every post is a disagreement, at least not from me ;) From the TEDx-talk, it does seem like the typical spectrum/gradient understanding is lacking in terms of being meaningful for individuals. Maybe because it's a sort of score as seen from the outside, and not even very precise one at that. Also, society also needs to accommodate and improve understanding of diversity, rather than expecting everyone to be clones of each other. The latter doesn't work that well, since we're then not encouraging people's natural strengths and creativity.
    - mannykannot 5y ago
      > Note, I never disagreed with the author… Well, it was certainly hard to see it any other way: Author: The spectrum is not a gradient. From your first reply: A typical spectrum is the exact same thing as gradients of frequencies. As your reply did not explicitly make this claim about the autism spectrum, it nominally did not directly contradict the author, but if it was not about the autism spectrum, then it was utterly pointless to raise it here - and, furthermore, it took several promptings on my part before you addressed this issue. With regard to your second paragraph, I completely agree.
      - _y5hn 5y ago