‘Ego depletion’—willpower as a muscle that can get fatigued—does not survive scrutiny.
A better way to consider willpower, argues Alberto De Luca, is ‘metacontrol theory’—not mental *resources* but mental *modes*: persistence, and flexibility.
buff.ly/iCSIwkv
Grandparents caring for their grandchildren scored higher on memory and verbal fluency tests, compared with those who didn’t—after adjusting for factors like age and health, research on nearly 3000 English grandparents 2016-2022 finds:
buff.ly/5UgLeF4
Adding “…including women and children” when referring to victims of violence boosts outrage, research by Grigoreva Crean et al finds,
The ‘women’ reference contributes independent of ‘children’ and of associations with motherhood:
buff.ly/1c5d49G
More evidence that conscious perception emerges from a constant brain-body dialogue—new research suggests people’s heart rate drops just before they notice they make a mistake in a visual task:
buff.ly/tWn3j0M
via @psypost.bsky.social
The shrimp that packs a punch also carries a shield for defence in one-on-one combat.
Mantis shrimp can whack like few others—prey and predator, but also rivals over shelter, and that shock-absorbing shield comes in handy, as high-speed footage shows:
www.popsci.com/environment/...
Social media’s effect on public opinion was to shift influence away from ‘elite’ experts to ordinary Jacks and Jills.
But, argues @danwphilosophy.bsky.social , AI will make the pendulum swing back the other way to some degree, and reshape public opinion:
buff.ly/pQY7K6Z
Parents of a baby cannot avoid regular exposure to, um, bodily effluvia of various kinds.
Might that lower their general disgust for disgusting stuff?
Yes, but… only once their baby is being weaned and starting to eat solids, new research suggests:
buff.ly/D5B4IdU
Longitudinal research by Grossman et al on global welfare forecast accuracy suggests that intellectual humility—recognizing the limits of one’s own knowledge— is a consistent predictor of forecast accuracy:
buff.ly/0hYUrly
“if you think signaling explains anything less than 60% of the human condition, then I’m afraid you are the crazy one.”
@davidpinsof doesn’t mince his words 💪
We can debate about how to estimate the %, but the argument that it is a *lot* is compelling:
buff.ly/8lJgnnn
Organic molecules—the building blocks of life—come in left- or right-handed variants, but all life on Earth is homochiral, only one variant of each molecule occurs.
Mirror life could be created, but what are the risks?
@kateadamala et al investigate:
buff.ly/pmCo9Ho
How do singers in a (Barbershop) choir decide what note to sing next?
Like most decision making, it’s not pure, raw memorizing—they rely on pattern recognition to learn *and* reproduce the right notes.
And if there’s no discernible pattern? 😱😱 :
buff.ly/Kb1xvwD
De evolutie gaf ons een krachtig vermogen om de complexe wereld te snappen: we kunnen patronen herkennen, leren en toepassen— onschatbaar bij het leren en het nemen van beslissingen.
Behalve wanneer die patronen onvindbaar zijn.
Mijn @apache_be stukje:
buff.ly/DQG3Gls
Chemical warfare is almost as old as life itself, both to kill or stun prey and to fend off predators—but while being toxic can be an effective defence, it’s not perfect.
Some animals eat poisons and don’t die. How do they do it?
buff.ly/vzAZGB0
Goodness knows the peer review process needs a lot of fixing.
But what it needs like a hole in the head is an “I did not use AI, honestly!” checkbox for reviewers, argues Matt Grawitch.
Instead, AI should be deliberately integrated in the process:
buff.ly/GJiZXka
How to Do Great Work?
Answering this substantively and not with glib waffle is not a modest challenge.
Paul Graham gives it a go in this long read to eclipse most long reads, densely packed with finely honed ideas and advice to get the best out of yourself:
buff.ly/IhplWo0
We evolved an immensely powerful capacity to make sense of a complex world—our ability to recognize, learn and deploy patterns is invaluable for learning and making decisions.
Except when we’re dealing with *absence* of patterns, as a choir discovered:
buff.ly/Kb1xvwD
TL;DR:🧵 by coauthor jvoth.bsky.social buff.ly/EpqegPW
2/2
Research by Bursztyn et al, one of the coolest ever studies on incentives to motivate behaviour—the Luftwaffe’s “rat race” of hierarchical, tiered, expanding status-based incentives for German WWII pilots.
The performance profiles say it all:
buff.ly/o1kye9h
1/2
Een ongemeen moeilijke zangpartituur die de baspartij maandenlang parten speelt legt bloot hoezeer we zijn afgestemd op patroonherkenning, zowel bij het uitvoeren van taken als bij het leren
Mijn @apache.be stukje, Het onleerbare lied:
buff.ly/DQG3Gls
A remarkable externality?
The release of a new album by big artists like Taylor Swift leads to a spike in streaming (+40%)… and a peak in traffic fatalities, involving mainly young drivers who are alone in their vehicle:
buff.ly/FgS99Df
Why would AI assistants behave like they’re human? In part because they’re trained to do so.
But there is more to it—even without training, they do: it’s their default.
This @anthropic post sets out a theory to explain—the Persona Selection Model :
buff.ly/PhK1iS9
Committing a faux pas is often experienced as a reputational threat.
Research by Goksel et al suggests that, contrary to the common inclination, laughing at yourself is a more effective impression-management strategy than signalling embarrassment:
buff.ly/7G8ABwu
The therapist, not the therapy
New research suggests people reading transcripts of psychotherapy interventions rate AI interventions higher for empathy, but only when the source is hidden or misrepresented as human:
buff.ly/v6hCN6s
via @psypost.bsky.social
“Our brains are identifiable as uniquely human through their common anatomical features, yet there is a tremendous amount of variability among human brains.”
Think the brain is utterly fascinating?
Think again. It’s *even more so*.
buff.ly/4OfwqQI
In 2040, people under 31 will not be able to legally buy tobacco products in the UK (legislation incoming next year, with the age increasing year on year).
Will it lead to the first smoke free generation?
buff.ly/VBCJtvu
Big5 personality profile predicts preference for different types of wine, research suggests—people high in openness or agreeableness lean towards higher alcohol content wines, those high in extraversion or neuroticism prefer low-alcohol variants:
buff.ly/7Tt8AL7
David Hume as an Art teacher? Hmm…
buff.ly/GvmyPgT
Be as offensive as you can.
Not *quite* the takeaway message from this week’s @davetrott piece, but…
If you want people to pay attention to you, don’t blend in with the wallpaper—be different, be interesting, yes even controversial or offensive:
buff.ly/DH9mGNH
Research by Corazzini et al suggests that people tend to believe they can influence uncertain outcomes and act accordingly.
Given unambiguous evidence disproving this illusion, they do adjust their beliefs, but only slowly adjust their behaviour:
buff.ly/0SmS4DM
Both learning and decision making are rooted in pattern recognition and interpretation—including in performing barbershop harmony songs.
How do I know?
My fellow basses and I ran into The Unlearnable Song:
buff.ly/Kb1xvwD
