|
SmithLab
@
SmithLabNeuro
Pittsburgh, PA
|
|
Systems and computational neuroscience in the Steel City.
|
|
|
32
Tweetovi
|
252
Pratim
|
248
Osobe koje vas prate
|
| Tweetovi |
|
SmithLab
@SmithLabNeuro
|
3. velj |
|
Congrats! Also, I feel we should become mortal enemies. Or spoof accounts for each other.
|
||
|
|
||
| SmithLab proslijedio/la je tweet | ||
|
Bob Hainsey
@HainseyB
|
3. velj |
|
Jana Kainerstorfer @cmu_bme continues the great talks @SPIEevents @PhotonicsWest neurophotonics plenary speaking on noninvasive monitoring intracranial pressure.
As a @CarnegieMellon alum I’m happy to see the Tartans on stage! pic.twitter.com/xDO79RIL1K
|
||
|
|
||
|
SmithLab
@SmithLabNeuro
|
26. sij |
|
Glad you find it useful! Link to the github page:
github.com/smithlabvision…
Also try out the built-in spike classifier:
biorxiv.org/content/10.110…
And as you said, utilities to convert between PLX and NEV are included.
|
||
|
|
||
|
SmithLab
@SmithLabNeuro
|
13. sij |
|
We hope you enjoy reading the work as much as we enjoyed doing the science, and we'd be happy to hear your thoughts. /7
|
||
|
|
||
|
SmithLab
@SmithLabNeuro
|
13. sij |
|
We think this work connects to and substantial advances a rich history of single-neuron studies in non-human primate perceptual decision-making, and provides an important link to numerous recent studies in rodents showing large-scale population activity modulation. /6
|
||
|
|
||
|
SmithLab
@SmithLabNeuro
|
13. sij |
|
Slow drift was weakly present in the population average, but strongly evident in mixtures of simultaneously recorded neurons./5
|
||
|
|
||
|
SmithLab
@SmithLabNeuro
|
13. sij |
|
Slow drift was present in both visual cortex (V4) and prefrontal cortex (8Ar). These brain areas, centimeters apart, were tightly locked together to this common process. /4
|
||
|
|
||
|
SmithLab
@SmithLabNeuro
|
13. sij |
|
This signal, which Ben called “slow drift”, could be described as a neural correlate of impulsivity or arousal. It was tightly related to slow fluctuations in the animal’s behavior in terms of hit rate, false alarm rate, reaction time, and pupil size. /3 pic.twitter.com/211Zn1HE0o
|
||
|
|
||
|
SmithLab
@SmithLabNeuro
|
13. sij |
|
Ben leveraged our large-scale electrophysiological recordings in visual and prefrontal cortex to identify a key signal underlying perceptual decisions. /2
|
||
|
|
||
|
SmithLab
@SmithLabNeuro
|
13. sij |
|
I’m very happy to post my first #tweeprint about a new manuscript we’ve just put on @biorxivpreprint:
biorxiv.org/content/10.110…
This is work by Ben Cowley, part of his thesis jointly with Byron Yu’s group and my own (Ben is now a postdoc with @jpillowtime). /1 pic.twitter.com/MytfTTQCzn
|
||
|
|
||
|
SmithLab
@SmithLabNeuro
|
27. ruj |
|
Fair enough - it’s hard to tell if you’ve been led astray by tuning for one feature until you know the bigger picture. If we simply measure tuning for X, find it, and declare this the X area and call it quits, then we’re doomed to never understand the brain.
|
||
|
|
||
|
SmithLab
@SmithLabNeuro
|
26. ruj |
|
Yes! And one could argue motion in MT is one of the great tuning curve stories - single neurons are tuned, stim them & you perceive motion in the tuned direction, lesion & you impair motion perception. Single-neuron tuning itself doesn't lead us astray, overinterpreting it does.
|
||
|
|
||
|
SmithLab
@SmithLabNeuro
|
26. ruj |
|
I think it only invalidates the idea that a single-neuron tuning curve is a simple, direct link to representation. If a neuron is hue-tuned for red, information about hue is in the population containing that neuron, but it is not a red neuron. Context matters & populations matter
|
||
|
|
||
|
SmithLab
@SmithLabNeuro
|
25. ruj |
|
Actually we designed the experiment specifically to relieve my nostalgia for Matlab R2014a (and earlier)! [when jet was default] 😄
|
||
|
|
||
|
SmithLab
@SmithLabNeuro
|
25. ruj |
|
No worries!
|
||
|
|
||
|
SmithLab
@SmithLabNeuro
|
25. ruj |
|
Great work by Ari, Pavan & Hugo on a fun collaboration!
@arisbenjamin @desipoika @jugoguj @KordingLab @SmithLabNeuro twitter.com/arisbenjamin/s…
|
||
|
|
||
|
SmithLab
@SmithLabNeuro
|
22. ruj |
|
This quote is carved into one of the entry doors @TheRoycroftInn in East Aurora, NY. Great advice for running a research lab. pic.twitter.com/HGG5GTvLdW
|
||
|
|
||
|
SmithLab
@SmithLabNeuro
|
30. kol |
|
... but I'm super excited to be working with these outstanding people in the years ahead, and for exciting new directions and collaborations! @CarnegieMellon @cmuneurosci @cmu_bme pic.twitter.com/rMeqaxOZkA
|
||
|
|
||
|
SmithLab
@SmithLabNeuro
|
30. kol |
|
It's been a really great 8+ years - sad to pack up the lab and see it empty after the moving truck has left ... pic.twitter.com/URNnkiqpt3
|
||
|
|
||
|
SmithLab
@SmithLabNeuro
|
21. kol |
|
We've posted a preprint of my (now grad) student Deepa Issar's undergrad project, training a neural network-based spike classifier to improve online decoding: biorxiv.org/content/10.110…
Code will be available soon. pic.twitter.com/kghEwdEG7s
|
||
|
|
||