LightNews
koenfucius.bsky.social
koenfucius
@koenfucius.bsky.social
2.1K followers 230 following 6.8K posts
Accidental behavioural economist koenfucius.substack.com
Posts Media Videos Starter Packs
Pinned
koenfucius.bsky.social
koenfucius @koenfucius.bsky.social · 13d
Blogged: A change of opinion may not be what it seems

Our surprise at the ease with which people change opinion - even make U-turns - is misplaced, and based a failure to understand how opinions arise and what they (don't) mean:

buff.ly/Q7ugwO6
1 3
koenfucius.bsky.social
koenfucius @koenfucius.bsky.social · 1h
AI will take our jobs—and that’s (mostly) OK with us, US survey by Friis & Riley finds.

When AI is described as equally effective as humans at lower cost, there is support for automation for 58% of occupations, with
12% of jobs categorically off limits:

buff.ly/82HlaFX
3
koenfucius.bsky.social
koenfucius @koenfucius.bsky.social · 4h
Study on survey data from 13k women from 180 countries suggests the qualities women prioritize in an ideal romantic partner (from politics and ethnicity to religion, financial security and height) are linked to their political views:

buff.ly/XHI99mc

via @psypost.bsky.social
2
koenfucius.bsky.social
koenfucius @koenfucius.bsky.social · 4h
Exactly 3 years ago, astronomers detected a gamma ray burst from the brightest explosion ever seen by humans, occurring 2.4 million light years from here.

It, and its afterglow, are still preoccupying them:

buff.ly/SkF1D59
2
koenfucius.bsky.social
koenfucius @koenfucius.bsky.social · 13h
For an argument to be valid, it must be absolutely impossible for the conclusion to be false if the premises are true—but it doesn’t matter *whether* the premises are true.

Weird? Perhaps, but it matters… a lot. Find out more:

buff.ly/AaGRO42
1
koenfucius.bsky.social
koenfucius @koenfucius.bsky.social · 15h
Hotelling’s Law and two ice cream vendors on a beach explain why moderate mainstream political parties’ drifting towards the extremes doesn’t necessarily mean the centrist voter moved first.

ICYMI, last call for my post, What happened to the middle?

buff.ly/KeljCQz
1
koenfucius.bsky.social
koenfucius @koenfucius.bsky.social · 16h
What happens when a chemical engineer sets out to establish the ideal level of income inequality?

Brian Gallagher talks to Venkat Venkatasubramanian who did just that:

buff.ly/rLrFQkM
1
koenfucius.bsky.social
koenfucius @koenfucius.bsky.social · 18h
People do not need to know much about science to trust it, argues @janpfa.bsky.social in his PhD thesis—we see ability to discover and understand what is hard to know as a sign of competence, and consensus about something as a sign of its trustworthiness:

buff.ly/dhEMqie
1 4
koenfucius.bsky.social
koenfucius @koenfucius.bsky.social · 22h
The common left and right narratives about gun violence—‘bad hombres’ needing a big stick; poor people needing more carrots—miss the point, says Jens Ludwig.

Most gun violence is impulsive, normal folk who in a very difficult situation make a bad call:

buff.ly/Xsxl1WF
1 2
koenfucius.bsky.social
koenfucius @koenfucius.bsky.social · 23h
How Lizardmen can make a nonsense of the correlation between rare characteristics.

@spencrgreenberg.bsky.social highlights a concerning weakness in survey-based research:

buff.ly/Ox74mgP
1 1
koenfucius.bsky.social
koenfucius @koenfucius.bsky.social · 1d
Timing is money, especially when you’re being judged by experts.

Research by Ginsburg & van Ours suggests the—random—order of appearance in a piano competition influences expert judgment, in turn predicting future success, regardless of musical ability:

buff.ly/ggNaQMX
2
koenfucius.bsky.social
koenfucius @koenfucius.bsky.social · 1d
Our brains can divide time into meaningful sections or “chapters”, and we can harness this ability to boost motivation, give meaning to mundane tasks, and navigate energy peaks and lows:

buff.ly/qXCYEo4
2
koenfucius.bsky.social
koenfucius @koenfucius.bsky.social · 1d
Is ‘colour’ objective or subjective?

Either answer may seem obvious to you, but it’s not that simple. Colour illustrates the tension and ambiguity that exists between the two:

buff.ly/9XekB3S
1
koenfucius.bsky.social
koenfucius @koenfucius.bsky.social · 1d
I’m sure those were submitted, but the jury must have determined them in poor taste…
1
koenfucius.bsky.social
koenfucius @koenfucius.bsky.social · 1d
”Why should the left be against stock market declines? […] Prime facie all arguments are that the left should be in favour.”
@brankomilan.bsky.social is puzzled by the left’s preoccupation with the stock market—and then unpuzzled.

Enlightening post:.

buff.ly/MusH19b
1 1 2
koenfucius.bsky.social
koenfucius @koenfucius.bsky.social · 1d
Do LLMs like ChatGPT, consummate as they are in having, er, meaningful conversations, have a conception of ‘meaning’?

It is widely believed they don’t (because of how they are built and work), but it’s not that clear cut, argues @tommyblanchard.bsky.social:

buff.ly/Smaypaw
1 3
koenfucius.bsky.social
koenfucius @koenfucius.bsky.social · 1d
Even detailed, continuous feedback is not a sure fire remedy for overconfidence, research with chess players suggests:

buff.ly/Py0dklM
1
koenfucius.bsky.social
koenfucius @koenfucius.bsky.social · 1d
I used to be a utilitarian, but I drifted away, without ever clearly pinpointing what I found was wrong with it.

David Pinsof is a fellow lapsed utilitarian, and he sets out with great clarity why—utilitarianism is bullshit.

Indeed, it is:

buff.ly/QgaCAoV
1 3
koenfucius.bsky.social
koenfucius @koenfucius.bsky.social · 1d
Another replication finds no evidence for stereotype threat—the fear of being judged based on negative stereotypes about the performance (eg maths) of a certain group one identifies with (eg women)—like here, in 5 countries, by Stoevenbelt et al:
buff.ly/PahTYGE
2
koenfucius.bsky.social
koenfucius @koenfucius.bsky.social · 1d
Human General Intelligence requires an ability to learn, which is closely linked to a goal.

Thus, artificial general intelligence will need a goal, too.

While we get ours from evolution, what goals will we give machines?

Good piece by @lionelpage.bsky.social:

buff.ly/nQ8XxAl
1 2 5
koenfucius.bsky.social
koenfucius @koenfucius.bsky.social · 1d
Are stereotypes inaccurate? This has long been the prevailing narrative—without much data in support, and still maintained in the face of strong contradicting evidence, this chapter by Jussim et al argues:

buff.ly/Hb2jsRs
1 2
koenfucius.bsky.social
koenfucius @koenfucius.bsky.social · 2d
Why would two ice cream vendors, having figured out the best location is right in the middle of the beach, start moving towards the extreme ends?

Not necessarily because that’s where the customers went.

A metaphor for centrifugal mainstream politics:

buff.ly/KeljCQz
2
koenfucius.bsky.social
koenfucius @koenfucius.bsky.social · 2d
Blair en Cameron loodsten hun partijen naar het centrum; ook in andere westerse democratieën regeerden gematigde, centristische partijen.

Is het echt de kiezer die hen nu wegdrijft van het midden?

Mijn @apache.be stukje, Waar is het midden naartoe?

buff.ly/xZ9BhpS
koenfucius.bsky.social
koenfucius @koenfucius.bsky.social · 2d
Tonsils, HUH! What are they good for?

Certainly not absolutely nothing.

So why have they been removed for centuries? Why less so more recently? And why is a tonsillectomy still one of the most prevalent childhood surgeries?

@laurenleffer.bsky.social explains:

buff.ly/GMBiuHp
1
koenfucius.bsky.social
koenfucius @koenfucius.bsky.social · 2d
Whatever you believe about when, or even whether, we’ll have AGI—it’s worth hypothesizing what would happen to us the day after.

@louisrosenberg.bsky.social argues the impact of the philosophical identity crisis might overshadow all else:

buff.ly/LorGSVc
1
koenfucius.bsky.social
koenfucius @koenfucius.bsky.social · 2d
Which do workers value more—pay or perks?

New research looks into how people trade-off money versus other job benefits:

buff.ly/D82VQde
1