Neuroscience PhD candidate @ZuckermanBrain@Columbia | @BerkeleyBioE alum | Studying RL mechanisms for vocal learning in songbirds ( fan of animalsππ¦π¦π¦)
We know dopamine guides reinforcement learning in externally rewarded behaviorsβthink a mouse learning to press a lever for food or juice. But what about skills like speech or athletics, where thereβs no explicit external reward, just an internal goal to match? π§΅ (1/7)
They should make a map so precise that it becomes the same exact size as the geography that itβs representing and then in fact later on comes to replace the geography itself.
this map literally does not correspond to geographic reality and the only reason they made it is to cater to stupid people with sub-80 spatial IQ who donβt even pay for the subway
A result I love that didnt make the thread: song basal ganglia isnt a simple "frequency" or "duration" knob
Suppressing sBG could β¬οΈβ¬οΈ or have no effect on an acoustic parameter depending on whether that parameter was relevant for learning. Thatβs exciting beyond birdsong (1/2)
ALT Shutting off the sBG always prevents the expression of learning (syllable maturity, top row). But affects frequency only if it is related to maturity. And if it *is* related, sBG suppression pushes frequency against the learning trajectory. Thus, when frequency is positively correlated with learning (right column), sBG suppression reduces frequency, but when it is negatively correlated (left column), suppression increases frequency.
Where, exactly, does learning happen in the brain?
Out now @Nature , we identify a synaptic locus of birdsong learning and show that the circuit can be tuned to make birds learn faster, but at a cost.π§΅ #neurosciencenature.com/articles/s41586-0β¦
Open link: rdcu.be/fiyrS
Where, exactly, does learning happen in the brain?
Out now @Nature , we identify a synaptic locus of birdsong learning and show that the circuit can be tuned to make birds learn faster, but at a cost.π§΅ #neurosciencenature.com/articles/s41586-0β¦
Open link: rdcu.be/fiyrS
Did you know that the first known whale, Pakicetus, walked on land? Learn more and trace the echoes through time, charting humanityβs understanding of whales, starting 50 million years ago with Pakicetus at projectceti.org/listen
Photo: @AMNH /Carl Buell
Now in PRE: "Transient dynamics of associative memory models."
I argue that the "blackout catastrophe" (the famous Ξ±β0.14 transition) is not catastrophic when viewed from an out-of-equilibrium, dynamical perspective.
Journal: journals.aps.org/pre/abstracβ¦
PDF: dclark.io/media/clark-pre-20β¦
bored want to draw but don't know what to draw so help me out here... first 5 ppl to comment any bird species of their choice will get a doodle of it sometime today or tmrw βοΈ