It's the best kind of surreal to announce that THE WAR FOR KINDNESS is out today! I truly hope this book can be useful, illuminating, and empowering to readers, and can't wait to hear your reactions. Much more at: warforkindness.com ✊🌻❤️

I appreciate this paper but also have many reactions (personal, not on behalf of CogSci Society). Most primarily: I have found cog sci to be the right community for me to develop in, intellectually and professionally. So for me, something about the field has been successful. /1

Testing out my image classifier. Works pretty well! #machinelearningparent #muffins #machinelearning

Exciting #rlang changes are coming in v0.4 Using {{}} for quasiquotation/interpolation looks like a quantum leap in readability (and teachability). #rstats

Interested in human-AI collaboration and sketching? Check out our latest work (w/ Monica Dinculescu, hardmaru, Google Magenta Project) exploring real-time collaboration with sketchRNN, which we’re excited to present at this year’s ACM Creativity & Cognition (Creativity & Cognition)!

In the age of large scale digital data, the challenge is becoming less, “What does my phenomenon look like in the wild?” and more, “How do I build a causal theory from wild data?.” My blog post on a new paper by Seth Frey et al:

My paper with Tal Linzen will be at #emnlp2019! We compare LSTMs, BERT, and GPT: syntactic agreement hits a performance ceiling with surprisingly little data (~10m tokens) and small models. Obvious to some, but: Big data isn't going to magically solve linguistics, folks.

At the conclusion of a successful hackathon week! We have been developing peekbank.stanford.edu - an open repository of word recognition eye-tracking data. Contributions welcome!

PhD student Natalia Vélez (natvelali.bsky.social @ 🦋) has earned a name for herself as "The Science Sketcher." Want her to draw you and your research? DM us a photo and a line about your work, and she'll choose a handful of submissions to put on the page. We'll share the results here. #scicomm

The #Paepsy2019 Grand finale, the Wilhelm-Wundt-Dialogue between Dr Caro Rowland & Michael C. Frank on the puzzle of #languageacquisition is about to start! #paepsy_2019

This is a really nice demo+cheatseet for thinking about how to analyze reaction times. It's also a wonderful example of communicating complex concepts visually. Thanks to Jonas K. Lindeløv for creating it!

Consider UC Irvine Language Science for your PhD. We are taking our first round of graduate applications this fall. We are a new kind of department for the 21st century. Below, some FAQs, and why you should apply. Thread below 👇

The first ManyBabies project, "Quantifying sources of variability in infancy research using the infant-directed speech preference," has been accepted at AMPPS! psyarxiv.com/s98ab This 67 lab (!) paper shows the power of collaboration to advance methods and theory!

Are you “not unhappy” with the changing seasons? Read this fresh preprint with Michael Franke about how people interpret double negatives psyarxiv.com/tqjr2

I’m no epidemiologist, but I am a #dataviz specialist, so here are some thoughts on coronavirus and log scales: 1) In the initial outbreak phase, a virus like this spreads exponentially not arithmetically, i.e a log scale is the natural way to track the spread

ManyBabies 1 now out! Quantifying sources of variability in infancy research using the infant directed speech preference. Babies prefer babytalk the world around - but preference strength was moderated by age, language, and method. doi.org/10.1177/251524… psyarxiv.com/s98ab/

As my father explained during his lifetime, a riot is the language of the unheard.

One presidential candidate is saying we should pass legislation that would ban chokeholds and make it more likely for police to be held accountable. His opponent is telling the military and the police to shoot and gas people who protest this violence. That’s a clear contrast.