1/ questions raised by linked thread: What are some other significant traits of liberal thinking? One is the tendency to think WEIRDly. Another is the tendency to employ only 1/2 of the evolved psychological mechanisms of social cognition (and of those mostly just care/harm)

2/ Another is that the left-wing emphasis on altruism/empathy/sympathy can be so single-minded, so exclusionary toward other necessary aspects of healthy society, that it becomes pathological.
3/ Another is left’s relative deficiency in the ability to imagine how other people could think differently, and its tendency to therefore conclude that “others” are sick, broken, malformed: https://t.co/sEfuvHbEwO
4/ Another is the left’s greater tendency throughout history to feel not only logically justified, but more, morally obligated, to exclude said sick/malformed/broken “others” from social discourse literally by any means necessary. https://t.co/OBbtApzHyD
5/ These things add up. Patterns emerge. A picture begins to form. More questions arise.

Have we been looking at the partisan divide all wrong all along?
6/ Are we wrong to base our analyses on the beliefs people have? The positions they take? The arguments they make in defense of those beliefs and positions?

Aren’t all of these merely symptoms of deeper causes? Shouldn’t we be looking for THEM?
7/ Are we wrong to categorize and analyze beliefs, positions, and arguments based on unexamined conveniences like what side of the room proponents once sat on,
8/ or based on what amounts to little more than dogmatic tradition, which we never think to question, never mind update based on new information, that one political side is for, say, change and progress while the other is for stability and status quo?
9/ We’ve learned a lot about ourselves in the 200+ years since we started thinking, categorizing, aggregating, in those ways. They’re out of date. Anachronistic. And frankly, based on what we now know, wrong.
10/ What we know now that we didn’t know then is that Hume was right when he said reason is the slave of the passions, or, in more modern terms, intuitions come first, and strategic reasoning follows. Said simply, all reasoning is motivated reasoning.
11/ We also know that the motivations for our reasoning - our passions, intuitions, instincts; our innate tendency to automatically and instantaneously like/dislike, approach/avoid, fight/flee -
12/ come from evolved psychological mechanisms and ways of thinking that were pre-wired into our brains by hundreds of millions of years of natural selection.
13/ What we know now is that all social and political thought and action is downstream from evolutionary psychology.

What we know now is that if the 1st principle of social psychology is “Intuitions come first, strategic reasoning follows,” then
14/ a new principle is needed; one that precedes the first principle. I’ll call it the 0th principle of social psychology it is:

“Psychology (or psychological profile) comes first, intuitions follow.”
15/ So the answer to the questions earlier in this thread is Yes.

We ARE wrong to be thinking, categorizing, analyzing, and concluding based on outdated anachronisms based on which side of the room proponents sat (i.e., left or right),
16/ We ARE wrong to be thinking, categorizing, analyzing, and concluding based on nothing more than status quo intellectual tradition, namely that one group is for change and progress while the other is for stability and social status quo.
17/ Instead, what we SHOULD be basing our thinking, categorizing, analyzing, and concluding on are the evolved psychological mechanisms that motivate everything else.
18/ Instead of looking at symptoms as we have been for the past 200+ years - empirically sloppy, intellectually lazy ones at that - we SHOULD be looking at causes.
19/ We SHOULD be defining, understanding, and analyzing partisan tendencies and groups as what they really are: psychological profiles, consisting mainly of the interactions between two aspects of social cognition; 1) cognitive style, 2) moral matrix. https://t.co/5FwFAECqoy
20/ When we do that, SO MUCH becomes clear. SO MANY of what appear to be paradoxes or internal contradictions not only cease to be come paradoxes or contradictions, but actually make sense.
21/ Not the least of these are possible connections between the psychological profiles and mental illnesses discussed in the thread linked to in the first tweet of THIS thread.
22/ One of the reasons we struggle so mightily in our attempts to understand ourselves and each other is that we’re doing it wrong. We’re fixating on symptoms and we’re blind to causes.
23/ We argue about the waves on the surface of the cultural ocean, even as we’re oblivious to the Gulf Stream of evolutionary psychology that carries all of us along.
24/ The situation is not unlike our befuddlement with human reasoning BEFORE Sperber and Mercier came up with their Argumentative Theory of reasoning.
25/ Recall that before Sperber and Mercier it was assumed that our capability of conscious reasoning evolved to help us make better decisions and find truth. But hundreds if not thousands of studies showed that we’re really bad at both.
26/ Scientists were befuddled. If reason evolved to help us make better decisions and find truth then why are we so bad at it? What’s wrong with our ability to reason? Where’s the bug in the code?
27/ The befuddlement disappeared like fog on a warm morning when Sperber and Mercier suggested that maybe it’s not our reasoning that’s flawed, maybe it’s our assumption as to its purpose.
28/ Instead, they proposed, what if we changed our assumption as to the evolved purpose of reason.

What if, they asked, reason DIDN’T evolve to help us find truth?

What if, instead, it DID evolve to help us WIN ARGUMENTS; persuade others that OUR intuitions are the RIGHT ones?
29/ As soon as they did that the befuddlement with reason cleared up. The supposed “flaws” of reasoning are not flaws at all, they’re features, “designed” by natural selection to help reason do exactly what it is supposed to do, persuade, convince, WIN.
30/ I suggest that a similar epiphany awaits us if we switch our assumptions about partisan animosities away from anachronistic categories like left and right, and unquestioned intellectual traditions like change/progress vs stability/status quo, and
31/ TOWARD a science- and evidence- based understanding of root causes. Pull the intellectual and investigated thread all the way back to where it ends; at natural selection, and the evolved psychological mechanisms of social cognition. https://t.co/kgnl9OQwKA
32/ If we do that, then, I propose, 1) findings about the correlation between mental illness and social cognition (i.e., political thinking and beliefs) will make sense, and 2)
33/ we’ll be able to develop *effective* policies and approaches that will actually help shrink the size of the partisan divide and reduce the amount of animosity that flows back and forth arcoss it.

More from All

How can we use language supervision to learn better visual representations for robotics?

Introducing Voltron: Language-Driven Representation Learning for Robotics!

Paper: https://t.co/gIsRPtSjKz
Models: https://t.co/NOB3cpATYG
Evaluation: https://t.co/aOzQu95J8z

🧵👇(1 / 12)


Videos of humans performing everyday tasks (Something-Something-v2, Ego4D) offer a rich and diverse resource for learning representations for robotic manipulation.

Yet, an underused part of these datasets are the rich, natural language annotations accompanying each video. (2/12)

The Voltron framework offers a simple way to use language supervision to shape representation learning, building off of prior work in representations for robotics like MVP (
https://t.co/Pb0mk9hb4i) and R3M (https://t.co/o2Fkc3fP0e).

The secret is *balance* (3/12)

Starting with a masked autoencoder over frames from these video clips, make a choice:

1) Condition on language and improve our ability to reconstruct the scene.

2) Generate language given the visual representation and improve our ability to describe what's happening. (4/12)

By trading off *conditioning* and *generation* we show that we can learn 1) better representations than prior methods, and 2) explicitly shape the balance of low and high-level features captured.

Why is the ability to shape this balance important? (5/12)

You May Also Like