“let me hear you make decisions / without your television / let me hear you speaking / just for me”

[larryhammer]

Three links:

Beyond Bob Ross, a collection of links to relaxing painting videos.

A well-edited AMV of Studio Ghibli scenes set to Loreena McKennitt’s “Mummers’ Dance”. Largely Miyazaki movies, but also some others including at least two of Takahata’s and a shot from Future Boy Conan. (via) (Bonus MV: dancers from classic musicals set to “Uptown Funk”.)

The Dunning-Kruger Effect is probably not real, but a statistical artifact. (via)

(Why, yes, I did bury the lede. Why do you ask?)

---L.

Subject quote from Stripped, Depeche Mode.

Comments

[asakiyume]

The thing about the Dunning-Kruger Effect is one more example of why I'm suspicious of most experiments that attempt to get sciency with anything relating to human nature or behavior. The more the premises and results of experiments are analyzed, the more one finds problems with the experiment and problems interpreting the data, due in large part (but not entirely) to unrecognized biases and assumptions.

Beyond that, the thing is that the lessons learned from the it-turns-out-to-be-poorly-constructed-or-interpreted experiments are *still valid* without the experiment. In other words, the lesson "Be careful when you try to estimate your skill; it's harder than you think; you may assume you know more than you do, or you may tend to underestimate yourself" is a good one just in principle, regardless of an experiment to support it.

I get really uncomfortable when people use science to support moral principles, because it makes me wonder what they'd do if the science didn't support the principle.

[thistleingrey]

the lesson is a good one just in principle

Agreed. Not only moral principles but cultural cross-currents, I think: has someone been taught since early childhood to sound self-deprecating or to sound bold, regardless of how they feel? How does this interact with responses to a researcher?

[asakiyume]

Yes, there are just so many factors to confound experiments.

[larryhammer]

Indeed.

[sovay]

I get really uncomfortable when people use science to support moral principles, because it makes me wonder what they'd do if the science didn't support the principle.

If they are any good at being a scientist, they alter or discard the theory. If they hang on to the theory in the face of the facts saying otherwise, they are doing science badly (and become relevant to my family's research on cognitive dissonance!). I know it's not that simple in practice, but having been raised by scientists if not professionally one myself, I reserve the right to judge.

it's long....

[asakiyume]

I expressed myself really badly because I was free associating, and that last paragraph didn't have much of a logical link to what I was saying before.

I agree that a good scientist should change their theory based on the evidence they obtain from experiments. What I was trying to say in that last paragraph, though, is that I think some things are axiomatic--things relating to moral/ethical principles is mainly what I'm thinking of--and not up for proving scientifically. For instance, "The survival of humanity is a good thing." --That's something I believe axiomatically. I don't want someone to try to prove the statement scientifically, not only because pragmatically speaking I don't think it's possible to design a good experiment to address the question, but also because I wouldn't want support for human existence to depend on the results of an experiment. The truth of the statement isn't something that depends on proof.

Or take education: I want education to be open to people regardless of what science says about their abilities. Science nowadays (unlike in the past...) says that markers like sex and skin melanin content really have nothing to do with our ability to learn at all. And that's great! But I would want people's access to education to be unchanged even if that *wasn't* what scientific evidence said.

I see the dangers in my outlook--namely, that you get someone with really awful moral/ethical beliefs who also doesn't want to let scientific evidence sway their opinions. Someone who, for instance, didn't want women *ever* to have access education, no matter what. But that's where I think one needs to get absolutist about people's moral beliefs and say some are better than others. Which is super duper dangerous also, as history has shown--but I guess I feel that science isn't the way we can avoid those dangers. For some things, yes! Like whether a vaccine works, or whether human activity contributes to climate change. Those things I think we can establish through science, and we should change our opinions depending on what science says. But not things relating to fundamental morals.

[oracne]

Thanks for the AMV link!

[larryhammer]

Innit cool?

[sovay]

The Dunning-Kruger Effect is probably not real, but a statistical artifact.

Interesting. My mother's research in the '70's was not a direct forerunner of the Dunning–Kruger experiments, but it was not totally dissimilar in that she was studying how people's self-perceptions interact with their objectively assessed skills, specifically from the perspective of cognitive dissonance. The prevailing assumption had been that when people who believed themselves to be bad at, say, math were shown that they had actually done quite well on a math test, they would accept the new data and do just as well if not better when given the opportunity to re-take the test. Instead, it turned out that in cases where I am bad at math was an essential part of the subject's self-image, they resolved the dissonance by doing worse on the test than they had the first time—failing questions they had originally aced. They couldn't twist reality around to make it match the inside of their heads, so they sabotaged the inside of their heads to make them match reality. A percentage of subjects did just realize they were better at math than they had believed and did not effectively falsify their answers the second time around: there was no dissonance in play. But where it was, it was a whammy.

[thistleingrey]

That is interesting as well!

[sovay]

That is interesting as well!

I think so!

I went looking for citations online and discovered that my grandfather actually did some similar work in the '40's, observing the selective recall of completed/unfinished tasks when viewed within a framework of success or failure, i.e., "No sweat, I zipped through that in no time!" vs. "I suck at this! It took forever!"

I am beginning to feel that Tiny Wittgenstein is some kind of family tradition, only expressed as a personification rather than a paper since I am not a psychologist.

[thistleingrey]

Tiny Wittgenstein is some kind of family tradition

I had that thought while reading your earlier comment.... Well, as long as Tiny Wittgenstein can take a seat sometimes.

[larryhammer]

Well, as long as Tiny Wittgenstein can take a seat sometimes.

That.

[sovay]

Well, as long as Tiny Wittgenstein can take a seat sometimes.

I have never failed a test because I thought I should!

[larryhammer]

I am ... not relieved, but at the least glad to hear that.

[sovay]

I am ... not relieved, but at the least glad to hear that.

I don't know how to read this comment.

[larryhammer]

I was reading your comment as indicating that Tiny Wittgenstein does indeed sometimes take a seat, and was glad to hear that. "Glad," though, somewhat understated the reaction but "relieved" was overstated it (and sounded condescending, to my ear).

[sovay]

"Glad," though, somewhat understated the reaction but "relieved" was overstated it (and sounded condescending, to my ear).

Thank you for the explanation. We must have different values for adjectives, because it sounded to me as though you did not quite believe me, and I was not looking forward to having to defend my mental health, much less against a skeptical interlocutor.

[larryhammer]

I was not looking forward to having to defend my mental health, much less against a skeptical interlocutor

Oh dear. Indeed no, not something to look forward to. Apologies for my clumsy reaction.

[larryhammer]

they resolved the dissonance by doing worse on the test than they had the first time

... ouch.