In this new paper led by @drcharlotte.bsky.social and myself, we explored phonological representations in monolingual and bilingual neural networks trained on speech recognition: doi.org/10.1073/pnas...
This is extremely rad. Here's developmental psychology:
I've got a couple bingos on here for sure haha
"Terrible things are happening outside. Poor helpless people are being dragged out of their homes. Families are torn apart. Men, women, and children are separated. Children come home from school to find that their parents have disappeared." Diary of Anne Frank January 13, 1943
Minute cryptic has really helped me break into understanding how the clues work! It's a single daily clue with an explainer video each day
Reading Bill Labov's papers in undergrad is one of the things that ignited my love for studying language. He meant a lot even to people like me who never met him. May his legacy continue
This response from @sbearbergman.bsky.social to @hildur.bsky.social’s post has changed my week— it is so wholesome, and human, and so the energy we all need right now. And Kerstin Langenberger, the naturalist who posted “I am on my way with the necessary equipment”? This woman is the hero we need.
You there - what day is it?
Why sir, it's stylish but illegal ikea monkey day
Stylish but illegal ikea monkey day! Then I haven't missed it!!
This is so embarrassing of Concordia, how wrong-side-of-history can one university get?
A report by Israeli newspaper Yedioth Ahronoth says that several universities in the US and Canada have entered into agreements with Israel-linked security companies to suppress Palestine solidarity protests occurring on their campuses.
🔴 LIVE updates: aje.io/es3
@kbyers.bsky.social
@asander-science.bsky.social
You can read all the details in our Open Access paper here, published in Infancy: onlinelibrary.wiley.com/doi/10.1111/... This is Melanie's first first-author publication!
On the whole, we found that item-level responses were capturing something about infants’ word production, but they weren’t giving us anything beyond global measures like age and vocabulary, and item-level responses weren’t helpful at all for comprehension.
This makes sense psychometrically: parents only get one binary chance to accurately report on each individual word, but can accurately capture their child’s vocabulary across the hundreds of words on the CDI.
The upshot: predictors that tell you how good a kid might be at knowing words in general (i.e., older kids know more words, and kids with bigger vocabularies know more words) are telling you as much about word-level performance as word-level metrics (whether the parent thinks that child knows “dog”)
This pattern didn’t hold in a more complex analysis with moderators: we added kids’ age and their total vocabulary score to a mixed-effects model. When these were taken into account, word-level knowledge no longer predicted performance.
Violin plots showing proportion target looking as a function of whether the child was reported to understand (top) or produce (bottom) a word. Black pointranges show the group means. Individual coloured points are trial performance.
Data from 126 kids (14-31mos) showed that without any moderators, kids looked more at target words they reportedly knew, but this only held for production:
🚨New paper alert! We estimate babies’ vocabs w/ CDI all the time. Can it also estimate the specific words kids know? We collected CDIs and gaze data to see if kids looked more at words they reportedly knew on the CDI. Here’s what we found: 🧵https://doi.org/10.1111/infa.12641
The Manybabies4 paper is out! Infants' Social Evaluation of Helpers and Hinderers: A Large-Scale, Multi-Lab, Coordinated Replication Study onlinelibrary.wiley.com/doi/abs/10.1... 1000 babies tested in 37 labs; "Overall, 49.34% of infants preferred Helpers over Hinderers in the social condition"
