Breaking the Bird Barrier: Scientist Decodes Zebra Finch Language

Join the discussion

• Hacker News
• Once again, Dr. Dolittle’s seminal work is ignored.
  by satisfice
• Another win for machine learning:

  >She then applied machine learning to analyse how information was encoded in the calls before testing her findings through behavioural experiments.

  by AndrewKemendo
• Does anyone have a link to a decoded message?
  by esafak
• No doubts the crows are all having a good laugh at our expense.
  by ChuckMcM
• I started feeding a local murder of crows earlier this year.

  They seem to have distinctive two-caw sounds for “human sighted, food is about to be thrown”, a “food is here, no sign of the human” and finally, a “human, we’re here, you coming out to throw us some scraps?”

  They make plenty of other calls too, but I haven’t been able to associate those with anything I’ve observed them doing.

  I may just be over anthropomorphizing, but I’ve recorded them and those three dual-caw calls sound subtly different (to my untrained ears). I’m certain actual ornithologists have looked into this. Chances are, though, I may be suffering from “I want to believe” syndrome.

  by BuyMyBitcoins
• I remember hearing about an interesting paper; it argued that Zebra finch songs were as complex as recursively enumerable languages on the Chomsky hierarchy. I wanted to see if I could find it but came across another paper arguing that their embedded context sensitivity can be explained by simpler rules.

      https://www.pnas.org/doi/10.1073/pnas.0908113106
  

  Just the same, these little fellows are some of the cutest on our planet.

  Left this comment as another computer science connection.

  by bluechair
• A lot of interesting information here, but this one paragraph blew my mind:

  > Although the birds occasionally made mistakes, they more often confused calls with similar meanings rather than similar sounds. “Their responses indicated they have a mental imagery of the meaning of their vocalisations,” Elie said. “In other words, that they understand the meaning of their call types.”

  by ghurtado
• Yes, it rather begs the omniscience of the researchers, doesn't it?
  by Exoristos
• I work in the field. The main takeaway is zebra finches (maybe other songbirds) can discriminate vocalizations based on function, even if they sound similar.

  Generally, birds can tell apart categories of sounds e.g. vocalizations from different individual birds, male vs female, call vs song, conspecific vs heterospecific etc. The question is if birds can do it for specific function e.g. agonistic calls vs non-agonistic calls. Simple question but way harder to test because of associated contextual info. with vocalizations.

  The paper is culmination of last decade of work (includes many of the past works) but this is the new result.

  by pks016
• Interesting! Could you break this down further? Maybe specific examples of what you're mentioning might help - for instance, what "similar-sounding calls with different functions" are zebra finches discriminating between? What exactly are different situations of these functions? How exactly (and through what experiments) did she determine this?

  And I read the article (but not her research paper), and have a vague idea, but it's tough to get a full grasp - and if this is just too difficult to fully explain, no worries!

  by computerdork
• From https://www.jeremycollerfoundation.org/news-and-insights/pre... it sounds like the main takeaway is that they seem to conceptualise the sounds:

  > the birds mistook long-distance contact calls and short-distance contact calls, which are acoustically very different. But, they never mistook a short contact call with a short alarm call, which have very similar acoustic patterns but entirely distinct meanings. As such, the study reveals that call perception elicits a mental imagery of the meaning of call-types, rather than triggering a reflexive response.

  How do you see that?

  by Aachen
• Very interesting topic, but a strange choice of source. I'd recommend these instead:

  Coller foundation press release: https://www.jeremycollerfoundation.org/news-and-insights/pre...

  The actual publication in Science: https://www.science.org/doi/10.1126/science.ads8482

  by ileonichwiesz
• Thank you, I was also looking for a better source and a link to the paper. I thought the paper was this older one https://www.nature.com/articles/s41467-018-06394-9 from 2018.
  by kiproping
• It does have the title going for it.
  by MathMonkeyMan
• Here's a non-paywalled version of the research.

  https://www.biorxiv.org/content/10.1101/2024.11.14.623689v1....

  by HarHarVeryFunny