Trevor Day (PhD)
banner
trevorkm.day
Trevor Day (PhD)
@trevorkm.day
Developmental psychologist and linguist
WA | MN | DC
T32 postdoc @ Georgetown with Elissa Newport and Peter Turkeltaub
🍁🥏♾️
show_rows = TRUE
November 6, 2025 at 8:44 PM
🤡
November 2, 2025 at 2:41 PM
Waiting for next season’s prize task “something that came from a museum” with an obvious 5 point winner.
October 24, 2025 at 2:57 PM
As a child (6?), my pediatrician once asked me if I had any more questions (about the visit). I asked him why the alphabet is the way that it is. He gave me a very nice explanation.
October 24, 2025 at 11:56 AM
I replied to you, lmao
October 11, 2025 at 9:33 PM
“Everyone uses AI” is a crazy take
October 11, 2025 at 9:21 PM
Academia man does whatever an academy does
October 1, 2025 at 12:32 PM
What does this mean?
September 10, 2025 at 12:01 PM
... this does pretty well, predicting scores within 20 words (out of up to 680), which generates more consistency in the first analysis, when kids had a mix of instruments.
June 16, 2025 at 3:14 PM
But, in our specific dataset (the Baby Connectome project), some kids were sampled using an instrument that isn't normed for their age range (they were too old), so we also created a set of models to predict vocabulary score from an undersampled score with a possible ceiling effect) ...
June 16, 2025 at 3:14 PM
... this gives summary values (peak growth rate, and time at peak) that are related to one another, but normally distributed.

This lets you compare across individuals whose vocabulary was sampled at {8, 11, 14} and {16, 19, 21}, for example.
June 16, 2025 at 3:14 PM
You say that like god would ever have been found on twitter
May 29, 2025 at 1:06 PM