Realized I am struggling so much with this chapter in my book (on evidence) because essentially it feels like the world is currently utterly hostile to the idea of evidence/data/measuring anything, BOTH on the left and the right tbh

Cannot believe that tech people will with a straight face try to convince me that hypervigilantly monitoring ten thousand microfeuds and the mental health statuses of OSS Big Deal Guys is an acceptable way to live

also the playbook for this is like: - politicians fight every step and refuse structurally impactful solutions and discard good plans and force the environmental advocates to make a bazillion compromises which creates imperfect process - then the advocates/scientists/etc get blamed for it ???????

Literally twice as many words in this chapter as there should be, why is my brain like this

More human production happening regardless of machine abilities 😃

 As you might recall I love to sew my own clothes. I also love clothes IN GENERAL, their features, their details. In my efforts to shift to make>buy sometimes I force myself to sketch a piece I really love, to break down WHAT I love

I mean doing situated text analysis BEFORE these models was already really interesting, and so is reading back through your old journals, I do not get why people are dunking on this, it shows understanding & curiosity. Bsky has turned into "we hate all models" as a default

This is something that I have ALWAYS wondered when seeing these estimates because it seemed intuitively obvious to me that folks overestimate small proportions! Neat thread on this!

Literally every single thing I have ever heard about this conference is bad

I think it takes a lot of chutzpah to say "this category grouping we came up with does not predict political choice reliably, and you know what that means? It's clearly a great predictor" I hate election group difference astrology so much

No one else around me thinks this but the best cognitive bias is pluralistic ignorance

Life has risks. Medicine has risks. Going to school also has risks btw, as do most things we value, like having relationships. The key is to appropriately understand the situation and the risks. In fact this is absolutely CENTRAL if you don't want to get taken in by grifts in the ultimate grift era

😬😬😬😬😬😬😬

On this, there's a lot of talk about cutting down on "toil" in technical jobs right now but even separate from the AI of it all, do we know how to do that? Do we have a vision for what non-grind knowledge work would look like?

Me on the way to the airport from one of my last keynote trips for the year: clearly I'm getting less filtered because I said 'bullshit' up there My friend: Hate to break it to you but you said bullshit TWICE

You're a social learner! We are good at transferring info to/from one another. We're "flexibly creative imitators." I love that phrase! It creates "cumulative culture."

Successful teams have: A learning culture (learning being celebrated), sense of agency, (belief that when you do something, a result will happen and not be blocked) self-efficacy (positive belief in yourself) & motivation, and sense of belonging, @grimalkina.bsky.social tells #Positconf2025

Streaming at #PositConf2025 is not working! I miss the last part of @grimalkina.bsky.social keynote 😿😭 Thank you, Dr. Cat, for everything you presented (I will finish when they share the video). I always follow what you do with great interest. It is always very enlightening to hear you speak.

Successful code is copied via conventional human social learning, @grimalkina.bsky.social tells #Positconf2025 . Coding solutions are improved by the group, via both ‘tweaks’ and ‘leaps’. Most leaps fail, she notes, but the wins are big advancement. All are part of the same process!

What makes developers feel more threatened by AI? If orgs think you need to be brilliant to succeed, and there’s a lot of public negative measurement of people’s perceived skills. - @grimalkina.bsky.social at #Positconf2025 Feel less threatened? Learning is valued.

Really enjoyable talk by @grimalkina.bsky.social on psychology for technical teams - and a lot applies to surgical teams and surgeon/scientists teams, have been listening and learning. thankyou #positConf2025 (also - love the suit!)

Good morning Posit, excited to share my keynote in which we will explore: - thinking traps - a gif that includes my dog (if you can spot him!) - data about your well-being you might REALLY want to send your boss #positconf2025

We have dropped an 🥰 anniversary special 🥰 for Change, Technically A year of podcasting with @analog-ashley.bsky.social and @daniloc.xyz has been more lovely than I ever expected, full of learning. Never let them take your swagger 😎 www.changetechnically.fyi/2396236/epis...

Find me at posit::conf to get -> - a very tiny sticker (the smallest at the conf?? we'll have to see) - a keynote about believing in open science so hard you create the missing evidence about technology work + thriving - a research-backed code review anxiety workbook #positconf2025

Pretty excited to be speaking at #positconf2025 also potentially the only conference I've spoken at this year where I think there will actually be bsky activity?! Stay tuned for some psychology for technology teams!

Becoming a parent has radicalized me in a lot of ways, but maybe nothing more than how much happily telling people “We’re having another girl!” has been taken as an open invitation for misogyny. Apparently one girl is a blessing, but two girls is a catastrophe. It’s gross.

Sitting at the airport and this family is talking to another family about they have all boys. This woman pipes up, "oh they're soooo much easier than girls!" Then they all shit on girls for a while, including the two women who have daughters. Just like complete spontaneous misogyny is so common.

I've been told a couple of times in tech to not use the term "empathy" because some groups dislike it which is hilarious because that's asking me to....have empathy.....

Cognitive Scaffolding > Cognitive Cherrypicking, I will die on this hill In other words: tying complex work outcomes to overly isolated, taken-out-of-context cognitive abilities and evidence about them will fail us youtu.be/wYhX3QxskxQ?...

A deep dive into the destruction of US cancer research by @jonathanmahler.bsky.social “It’s an absolutely unmitigated disaster,” a former top official at NIH told him. “It will take decades to recover from this, if we ever do.” Gift link: nyti.ms/48iH3Cr

theme improvements mean a lot to me, emotionally! Congrats team!!!!

TODAY! I am hosting our free "developer science" office hour, an experiment from Catharsis in platforming interesting experts for lively informal discussion on technical topics and bridging interesting research with the real world 🥰 www.eventbrite.com/e/empirical-...

"cultures are not neutral because they have human-made assumptions built into their designs" - Hamedani, M. G., Markus, H. R., Hetey, R. C., & Eberhardt, J. L. (2024). We built this culture (so we can change it): Seven principles for intentional culture change. American Psychologist, 79(3), 384.

This science story has everything: - a superheroic scientist who has the condition she is battling to drive work on - a riveting scientific breakthrough story (I was gasping) - an absolute gutpunch about whether her work will be defunded www.science.org/content/arti...

People keep telling me they can't afford to attend #PositConf2025 next week, so I need to make a public service announcement: attending virtually is very affordable (or free!) & is amazing! Get registered & hangout with me on the Discord server!! #databs #rstats #python

*raises hand* childcare literally helps all of us. The future of our entire world is this world's children. It always feels kind of weird to say this as a queer childfree person but ?!?!?!?!??!?! commonsense

When you are about to interview an empirical security researcher you have to do your homework We've still got a little room! Join us this week at an open "dev science" office hour with Dr. Ariana Mirian at: www.eventbrite.com/e/empirical-...

We just finished Outer Wilds and I can't stop thinking about it

Cat presents a really empathetic view of how people come to develop anti-vaxx views, especially when they're in a medical system that treats them with disdain AND doesn't seem to actually help them.

These tech men using their finite time on planet earth to joke that I am so untechnical I couldn't use a computer mouse if it wasn't on a mousepad have the exact same stench as the men of the church of my childhood who joked that I got a good SAT score because I accidentally pattern matched...

Speaking outside the academics here, so what do I know. But I honestly think a lot of unexamined issues here come from the fact that people are not willing to face what an enormous completely self-sustaining ecosystem, pseudoscience has become. They CAN and WILL replicate themselves.

This event on Empirical Security that I teased yesterday is now live for registration! 😎 Me & @arianamirian.bsky.social dive into the path to empirical security, why it's hard, why it's important. It's a zoom science office hour where YOU can ask questions too www.eventbrite.com/e/empirical-...

Nooooooo @arianamirian.bsky.social yet another one for our "emotional sabotage" security files (I collect examples like this thank you)

“…our results suggest that anti-phishing training programs, in their current and commonly deployed forms, are unlikely to offer significant practical value in reducing phishing risks.” —“Obvious” solutions often fail. *Especially* training. This is why it’s important to actually test interventions.

This event on Empirical Security that I teased yesterday is now live for registration! 😎 Me & @arianamirian.bsky.social dive into the path to empirical security, why it's hard, why it's important. It's a zoom science office hour where YOU can ask questions too www.eventbrite.com/e/empirical-...

If the human brain is so impressive, why does my human brain currently want to write every single idea in the world AS LONG AS IT ISN'T THE END OF MY BOOK

I had good ideas in public and forgot that this is mostly a path to additional labor for women. Womp womp.

(whispers: this is a pre-announcement but an event is coming together with an INCREDIBLE researcher next week & it is FREE & OPEN & VIRTUAL it's on EMPIRICAL SECURITY THAT IS COMPASSIONATE ABOUT HUMAN BEHAVIOR I'm PSYCHED & you should stay tuned to my feed so you can grab a limited slot 🥰 )

tangential but universal, this is a CORE pet peeve of mine in applied work

yes, certainly scientific models can be good, I'm a cognitive modeller 🥰 but they can also be very bad, maybe useful to see this too: bsky.app/profile/oliv...

A challenging paper I'm looking forward to reading. I believe more strongly than I can say in the need to create a positive vision for models and how people use models that is a part of *science* rather than "product." As people in the tech industry we have a responsibility for this.