Alex Danvers
@drdanvers.bsky.social
Social Psychologist, Data Scientist, Baseball Fan, Dog Dad, Improviser
Reposted by Alex Danvers
Want to support science on infants' emotions? Parents of babies (2-24 months) can complete our short, online study: umiami.qualtrics.com/jfe/form/SV_...
January 6, 2026 at 6:49 PM
Want to support science on infants' emotions? Parents of babies (2-24 months) can complete our short, online study: umiami.qualtrics.com/jfe/form/SV_...
Reposted by Alex Danvers
Does social media harm everyone?
No. But it harms *most* adolescents.
However, not all platforms are harmful.
An analysis of 44,211 diaries from 479 adolescents over 100 days finds that 60% of adolescents experienced small, negative effects of social media
link.springer.com/article/10.1...
No. But it harms *most* adolescents.
However, not all platforms are harmful.
An analysis of 44,211 diaries from 479 adolescents over 100 days finds that 60% of adolescents experienced small, negative effects of social media
link.springer.com/article/10.1...
January 6, 2026 at 7:10 PM
Does social media harm everyone?
No. But it harms *most* adolescents.
However, not all platforms are harmful.
An analysis of 44,211 diaries from 479 adolescents over 100 days finds that 60% of adolescents experienced small, negative effects of social media
link.springer.com/article/10.1...
No. But it harms *most* adolescents.
However, not all platforms are harmful.
An analysis of 44,211 diaries from 479 adolescents over 100 days finds that 60% of adolescents experienced small, negative effects of social media
link.springer.com/article/10.1...
Reposted by Alex Danvers
We recently published a theoretical review about how compositional and generative mechanisms in working memory provide a flexible engine for creative perception and imagery.
Pre-print:
osf.io/preprints/ps...
Paper: www.sciencedirect.com/science/arti...
Pre-print:
osf.io/preprints/ps...
Paper: www.sciencedirect.com/science/arti...
January 6, 2026 at 7:04 PM
We recently published a theoretical review about how compositional and generative mechanisms in working memory provide a flexible engine for creative perception and imagery.
Pre-print:
osf.io/preprints/ps...
Paper: www.sciencedirect.com/science/arti...
Pre-print:
osf.io/preprints/ps...
Paper: www.sciencedirect.com/science/arti...
Reposted by Alex Danvers
In summary, this combination of compositionality and task dependence means that, as scientists, we cannot find a “fundamental” working memory format. Any changes we make to an experiment to zero-in on the memory will also change the memory itself! (19)
a brown dog is sitting on a couch holding a gift box
Alt: a brown dog is sitting on a couch and tilting its head in confusion
media.tenor.com
January 6, 2026 at 7:28 PM
In summary, this combination of compositionality and task dependence means that, as scientists, we cannot find a “fundamental” working memory format. Any changes we make to an experiment to zero-in on the memory will also change the memory itself! (19)
Reposted by Alex Danvers
New website is up! Code for it can be found here: github.com/embodied-com...
January 6, 2026 at 12:49 PM
New website is up! Code for it can be found here: github.com/embodied-com...
Reposted by Alex Danvers
I’m not going to play. But who else got this email? Seems like a lot of us
Some researchers have used a preprint of one of my papers, run it through an AI and got it to generate follow-up ideas. Now they would like me to rate the suggestions.
Do I play? I don’t think I want to 🤔
Do I play? I don’t think I want to 🤔
January 6, 2026 at 1:14 PM
I’m not going to play. But who else got this email? Seems like a lot of us
Reposted by Alex Danvers
This is one of my favorite law review articles of all time. You should read it, if you haven't yet. stanfordlawreview.org/wp-content/u...
January 5, 2026 at 9:52 PM
This is one of my favorite law review articles of all time. You should read it, if you haven't yet. stanfordlawreview.org/wp-content/u...
Reposted by Alex Danvers
What is the computational role of dendritic excitations? Byung Hun Lee and team mapped voltage dynamics throughout the dendritic trees of CA1 pyramidal neurons in mice navigating in virtual reality. www.biorxiv.org/content/10.6...
January 4, 2026 at 6:17 AM
What is the computational role of dendritic excitations? Byung Hun Lee and team mapped voltage dynamics throughout the dendritic trees of CA1 pyramidal neurons in mice navigating in virtual reality. www.biorxiv.org/content/10.6...
Reposted by Alex Danvers
Jaana was a distinguished engineer at GitHub and is now a principal engineer at Google.
I expect more testimonials from accomplished engineers about productivity gains of AI in 2026.
Engineers claiming AI tools don’t work well will increasingly look like a skill issue than a problem with AI tools
I expect more testimonials from accomplished engineers about productivity gains of AI in 2026.
Engineers claiming AI tools don’t work well will increasingly look like a skill issue than a problem with AI tools
January 3, 2026 at 6:18 PM
Jaana was a distinguished engineer at GitHub and is now a principal engineer at Google.
I expect more testimonials from accomplished engineers about productivity gains of AI in 2026.
Engineers claiming AI tools don’t work well will increasingly look like a skill issue than a problem with AI tools
I expect more testimonials from accomplished engineers about productivity gains of AI in 2026.
Engineers claiming AI tools don’t work well will increasingly look like a skill issue than a problem with AI tools
LLMs eat Google search, then eat themselves. The way to find reliable information once again becomes curated blog rolls.
The play for OpenAI to make money is to inject ads into ChatGPT and that is going to lead to enshittification* with breathtaking speed.
*the pure meaning of the word— ChatGPT will be shifting who its customers are from users to advertisers once this happens
*the pure meaning of the word— ChatGPT will be shifting who its customers are from users to advertisers once this happens
January 3, 2026 at 2:45 AM
LLMs eat Google search, then eat themselves. The way to find reliable information once again becomes curated blog rolls.
Violence
Asking once again that researchers spend even half as much time thinking about design, sampling, and measurement as they do tinkering with their favorite R package to analyze and present their results.
January 3, 2026 at 2:43 AM
Violence
Reposted by Alex Danvers
This gets at something important that is in danger of being lost not only in the AI conversation but in any efficiency-focused, assembly-line model of academia. A lot of the best scholarship needs wayward paths. It's time-intensive, unpredictable, odd. The process is what produces the knowledge.
One of many things the pro-AI crowd doesn't seem to understand is the very important distinction between academic scholarship and the production of papers. Of course they can be related but they're not necessarily the same thing
December 21, 2025 at 11:56 PM
This gets at something important that is in danger of being lost not only in the AI conversation but in any efficiency-focused, assembly-line model of academia. A lot of the best scholarship needs wayward paths. It's time-intensive, unpredictable, odd. The process is what produces the knowledge.
“When wealth buffers are ample, individuals tend not only to take more risks, but also to learn in more exploratory and experimental ways.”
#psychology #cognitivescience #culturalevolution
phys.org/news/2025-12...
#psychology #cognitivescience #culturalevolution
phys.org/news/2025-12...
Personal risk tolerance has sweeping implications for how societies evolve
In his biography of Elon Musk, historian Walter Isaacson describes a game of Texas Hold "Em poker in which Musk went all in—on every hand.
phys.org
January 2, 2026 at 11:27 PM
“When wealth buffers are ample, individuals tend not only to take more risks, but also to learn in more exploratory and experimental ways.”
#psychology #cognitivescience #culturalevolution
phys.org/news/2025-12...
#psychology #cognitivescience #culturalevolution
phys.org/news/2025-12...
Very nice reading as you contemplate New Year’s resolutions!
Richard Smith condenses Rutger Bregman’s Reith Lectures into a blog.
richardswsmith.wordpress.com/2025/12/30/h...
richardswsmith.wordpress.com/2025/12/30/h...
How small committed groups can change the world and how we need new groups now in a time of crisis
Dutch historian Rutger Bregman’s fine Reith lectures boil down to a core argument: the world is in a mess; we need drastic change; small groups of committed people can make that change happen; it t…
richardswsmith.wordpress.com
January 2, 2026 at 4:51 AM
Very nice reading as you contemplate New Year’s resolutions!
Reposted by Alex Danvers
My 12yo has only ever seen Hugh Jackman in Les Mis and Greatest Showman and she absolutely did not believe me when I explained that he’s Wolverine and it took several clips to convince her
December 31, 2025 at 8:53 PM
My 12yo has only ever seen Hugh Jackman in Les Mis and Greatest Showman and she absolutely did not believe me when I explained that he’s Wolverine and it took several clips to convince her
Probably would have liked this class in college
Belief in "Weird" Things. Old syllabus here: drive.google.com/file/d/1gyoJ...
PSY 318 Syllabus, 2018 Fall 01.pdf
drive.google.com
January 2, 2026 at 2:31 AM
Probably would have liked this class in college
Reposted by Alex Danvers
My students struggled w/in-class exams last semester (much more than pre-pandemic students did). Trying optional re-takes this semester, where very similar exam given the following class mtg (Thur->Tue). Outcome = whichever higher: Add 10% of retake score to take 1, OR use 80% of retake score. 1/2:
December 31, 2025 at 9:15 PM
My students struggled w/in-class exams last semester (much more than pre-pandemic students did). Trying optional re-takes this semester, where very similar exam given the following class mtg (Thur->Tue). Outcome = whichever higher: Add 10% of retake score to take 1, OR use 80% of retake score. 1/2:
Reposted by Alex Danvers
Devastating:
December 31, 2025 at 11:13 PM
Devastating:
Reposted by Alex Danvers
Happy new year to ye all - let’s do this!
December 31, 2025 at 11:18 PM
Happy new year to ye all - let’s do this!
Reposted by Alex Danvers
Things aren’t always black and white. These horses are identically bright.
January 1, 2026 at 2:33 AM
Things aren’t always black and white. These horses are identically bright.
Do years get reverse psychology? Because I don’t think 2026 is going to be my year.
January 2, 2026 at 2:27 AM
Do years get reverse psychology? Because I don’t think 2026 is going to be my year.
Imai to the Astros. Glad it’s not the Dodgers. That said, I’m still salty with the Astros for the sign stealing scandal, so
I don’t love to see this either.
I assume he’s a pretty strong pitcher right out the gate. Then he triggers opt outs.
#MLB #Imai #Astros
www.mlb.com/news/tatsuya...
I don’t love to see this either.
I assume he’s a pretty strong pitcher right out the gate. Then he triggers opt outs.
#MLB #Imai #Astros
www.mlb.com/news/tatsuya...
Japanese righty Imai agrees to deal with Astros (source)
HOUSTON -- The Astros made baseball’s first big splash of 2026 by agreeing Thursday to a three-year, $54 million deal with Japanese right-hander Tatsuya Imai, a source told MLB.com's Mark Feinsand. Th...
www.mlb.com
January 2, 2026 at 2:26 AM
Imai to the Astros. Glad it’s not the Dodgers. That said, I’m still salty with the Astros for the sign stealing scandal, so
I don’t love to see this either.
I assume he’s a pretty strong pitcher right out the gate. Then he triggers opt outs.
#MLB #Imai #Astros
www.mlb.com/news/tatsuya...
I don’t love to see this either.
I assume he’s a pretty strong pitcher right out the gate. Then he triggers opt outs.
#MLB #Imai #Astros
www.mlb.com/news/tatsuya...
Nice advice!
Is one of your goals for 2026 to write more? One question I get all the time is, "How do you make time to write?" So, here is my unsolicited writing advice for the new year:
January 2, 2026 at 1:16 AM
Nice advice!
Reposted by Alex Danvers
2020s: AI "System Prompts" are lengthy, carefully constructed sets of expert rules about a particular domain, created by "prompt engineers".
1980s: AI "Expert Systems" were lengthy, carefully constructed sets of expert rules about a particular domain, created by "knowledge engineers".
1980s: AI "Expert Systems" were lengthy, carefully constructed sets of expert rules about a particular domain, created by "knowledge engineers".
January 1, 2026 at 8:59 PM
2020s: AI "System Prompts" are lengthy, carefully constructed sets of expert rules about a particular domain, created by "prompt engineers".
1980s: AI "Expert Systems" were lengthy, carefully constructed sets of expert rules about a particular domain, created by "knowledge engineers".
1980s: AI "Expert Systems" were lengthy, carefully constructed sets of expert rules about a particular domain, created by "knowledge engineers".