Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Researchers translate a bird's brain activity into song

eurekalert.org

Post reply on HN
Print view View original

1–9 of 9 posts

Re: Researchers translate a bird's brain activity into song

#2

Post by subroutine »

I TA'd for Tim Gentner at UCSD. He tried to recruit me to his lab; interesting research but I really hated corralling birds. Each quarter a different grad student is responsible for 'the pigeon lab'. It's 10% showing undergrads how reinforcement learning works, 90% chasing down loose pigeons through the basement of McGill & Mandler Hall. I could barely get the pigeons, so I was always in aw of how his lab members gently snatched finches in mid flight.

Re: Researchers translate a bird's brain activity into song

#3

Post by sci_prog »

For a second I was wondering why don't they just apply the same technique on humans and train their model that way instead... that's when I remembered that it involves sticking electrodes into one's brain to get the necessary data. What are the chances that we'll be able to read the brain activity noninvasively (thorough the scull) and at detail great enough to achieve similar things?

Re: Researchers translate a bird's brain activity into song

#6

Post by raducu »

sci_prog wrote: post #3

For a second I was wondering why don't they just apply the same technique on humans and train their model that way instead... that's when I remembered that it involves sticking electrodes into one's brain to get the necessary data. What are the chances that we'll be able to read the brain activity noninvasively (thorough the scull) and at detail great enough to achieve similar things?

I always wonder when I hear about neuralink -- since we plan on sticking wires into the brain anyway -- why not go a step further, and stick some wires into the amygdala and some other areas and cure things like BPD, sociopathy, ADHD and some other.

It would have been so nice to do that in my younger years when I struggled with BPD.

Re: Researchers translate a bird's brain activity into song

#7

Post by wombat23 »

otikik wrote: post #5
otikik wrote: post #4

Paper available here: https://www.cell.com/current-biology/pdfExtended/S0960-9822(...

I have not been able to locate any audible samples of the "song" that is produced.

I found them here just before the references:

https://www.sciencedirect.com/science/article/pii/S096098222...

Re: Researchers translate a bird's brain activity into song

#8

Post by captainmuon »

sci_prog wrote: post #3

For a second I was wondering why don't they just apply the same technique on humans and train their model that way instead... that's when I remembered that it involves sticking electrodes into one's brain to get the necessary data. What are the chances that we'll be able to read the brain activity noninvasively (thorough the scull) and at detail great enough to achieve similar things?

Wasn't there a study some years ago where they could reconstruct subvocalizations (the "inner voice") by placing electrodes on the throat or jaw? The realisation was that when you say something in you head, you send a very weak signal to the acutal muscles that can be picked up. I remember a bit of hysteria that someone would be able to do this remotely, in order to scan people at airports, but that fortunately never materialized.

I think this was the story: https://www.sciencedaily.com/releases/2018/04/180405133040.h...

Re: Researchers translate a bird's brain activity into song

#9

Post by throwaway88974 »

captainmuon wrote: post #8
sci_prog wrote: post #3

For a second I was wondering why don't they just apply the same technique on humans and train their model that way instead... that's when I remembered that it involves sticking electrodes into one's brain to get the necessary data. What are the chances that we'll be able to read the brain activity noninvasively (thorough the scull) and at detail great enough to achieve similar things?

Wasn't there a study some years ago where they could reconstruct subvocalizations (the "inner voice") by placing electrodes on the throat or jaw? The realisation was that when you say something in you head, you send a very weak signal to the acutal muscles that can be picked up. I remember a bit of hysteria that someone would be able to do this remotely, in order to scan people at airports, but that fortunately never…

It is already possible to do this remotely. There's tech that allows to point some kind of sonic device at your head from a fair distance and it's possible to hear your inner voice. I am the victim of this device testing here. For the last year a group of people used sonic device to test interaction with my brain, without my permission, illegally. I have experienced similar symptoms to the ones reported by victims of "Havana Syndrome". I could provide more details, if interested.
Post reply on HN