For each chick they do 24 trials divided into 4 blocks with retraining on the ambiguous shape and actual rewards after each block. During the actual tests they didn't give rewards. In figure 1 they show the data bucketed by trial index. It's a bit surprising it doesn't show any apparent effect vs trial number, e.g. the first trial after retraining being slightly different.
I have to admit I'm super skeptical there's not some stupid mistake here. Definitely thought provoking. But I wish they'd kept iteratively removing elements until the correlation stopped happening, so they could nail down causation more precisely.
Objects that have sharp edges generate higher frequency harmonics when agitated, because lower-size features resonate on higher frequencies (like shorter strings ring on higher pitch). Objects that are round resonate on low frequencies only. The "kiki" sound has more high frequency content than the "bouba" sound, and it's no mystery why the brain associates one with the other.
> But I guess it's about why so we associate those with spiky shapes, though surely it's because they represent sharp immediate changes in frequency?
Sure, but it's a very abstract connection between objects being sharp in vision and frequencies changing sharply in hearing. There's no guarantee any given organism would make the connection.
In the book „the design of everyday things“ it is mentioned that „natural mappings“ exist. Moving the knob of a vertical slider to the upper end universally means „brighter“ or „louder“, not „less bright“ or „more silent“.
I don't think it's abstract at all. Rub something sharp (anything from a stick to a phonograph needle) on an object and you'll directly transcribe its spatial frequency spectrum into an audio frequency spectrum.
>But I guess it's about why so we associate those with spiky shapes
I think the why just got a lot tricker than we imagined. Because we failed to replicate this experiment on other primates, we couldn't avoid a semantic suspicion about those associations. Now we probably have to set semantics aside or let it get a lot weirder, because we can replicate across ~300My.
>surely it's because they represent sharp immediate changes in frequency?
Maybe, and I think "multi-sensory signal processing" is the best framing, but the representation could also carry harder to think about things like "harm".
It's also super cool because the bouba-kiki effect framing was chosen due to methodological convenience for linguists and cultural anthropologists and their experimental bounds, not neuroscientists or signal processing folks. We could potentially find other experiments quickly, since chicks are a model organism and the mechanism is clear.
This is just one micro-instance of a much larger thing. Brain encodes structural similarity across modalities. Corollary: language is far from arbitrary labels for things.
No, language is still pretty close to arbitrary labels. The handful of tenuous common threads like the bouba-kiki effect don't change the overall picture that much. The simple fact that language varies as much as it does is sufficient to prove that it's only loosely bound to anything universal.
I think this is a misunderstanding of the arbitrariness of the sign. Arbitrary doesn't mean "random" or "uniformly sampled." The fact there are systematic tendencies among languages in how things are called doesn't negate the arbitrariness of the sign, they could have been called other things. We can also decide to refer to things by another name and we can use any arbitrary name we like! There is no limits on what names we can use (besides silly physiological constraints like having a word with 50 000 consonants). But, of course, there's much more to language than just labels!
For me, the interesting thing in this paper vis-à-vis language is that it shows how much innate structure in cognition must shape our language.
Arbitrariness of the sign is a principle that requires so many epicycles to present as "true" that it's more of a warning against overgeneralization than an insight with any significant predictive power in its own right.
I'm not entirely sold by this discovery. For example when you learn to train dogs, you learn about the 3 voices. Encouraging voice, atta boy, negative voice, more stern, and the big "NO!".
To some degree these words type sounding language are doing the same thing. Some sounds will irk, some will soothe, and it would affect this 'evidence' found.
I think the researchers agree with your premise. The “evidence” is not that chicks have more language understanding than previously understood, but rather that the source of the universality of bouba/kiki is due to something more primitive than built in human language hardware.
I think it’s natural to think of this in terms of frequencies so the kiki shape has a higher visual frequency. As does the word have a higher audio frequencies within in than bouba so that is naturally associated with the lower frequency undulating line of that shape.
The published one repeated the experiment w/ day old chicks and IIRC the same number w/ the same results, so it's got a little more N than the preprint.
This phrase is a direct quote from the 1955 play (and 1960 film) Inherit the Wind, spoken by the character Henry Drummond (based on Clarence Darrow) regarding the teaching of evolution. It frames scientific education and intellectual freedom as the ultimate, pure progress of human civilization, contrasting with dogmatic resistance.
Context: The line refers to the 1925 Scopes "Monkey" Trial, which debated the legality of teaching evolution in Tennessee schools.
Significance: It serves as a dramatic defense of modernism, science, and freedom of thought against traditionalist views.
Cultural Impact: While based on historical events, the play uses this line to argue that intellectual inquiry is the cornerstone of advancement.
For each chick they do 24 trials divided into 4 blocks with retraining on the ambiguous shape and actual rewards after each block. During the actual tests they didn't give rewards. In figure 1 they show the data bucketed by trial index. It's a bit surprising it doesn't show any apparent effect vs trial number, e.g. the first trial after retraining being slightly different.
I have to admit I'm super skeptical there's not some stupid mistake here. Definitely thought provoking. But I wish they'd kept iteratively removing elements until the correlation stopped happening, so they could nail down causation more precisely.
Objects that have sharp edges generate higher frequency harmonics when agitated, because lower-size features resonate on higher frequencies (like shorter strings ring on higher pitch). Objects that are round resonate on low frequencies only. The "kiki" sound has more high frequency content than the "bouba" sound, and it's no mystery why the brain associates one with the other.
Is this not reducible to whether a speech sound contains fricatives and stops or not? They produce spiky sounds
But I guess it's about why so we associate those with spiky shapes, though surely it's because they represent sharp immediate changes in frequency?
I'd be interested on results of shapes imagined when you take the source as musical or other non speech sounds.
> But I guess it's about why so we associate those with spiky shapes, though surely it's because they represent sharp immediate changes in frequency?
Sure, but it's a very abstract connection between objects being sharp in vision and frequencies changing sharply in hearing. There's no guarantee any given organism would make the connection.
In the book „the design of everyday things“ it is mentioned that „natural mappings“ exist. Moving the knob of a vertical slider to the upper end universally means „brighter“ or „louder“, not „less bright“ or „more silent“.
I don't think it's abstract at all. Rub something sharp (anything from a stick to a phonograph needle) on an object and you'll directly transcribe its spatial frequency spectrum into an audio frequency spectrum.
Do you think it's obvious that a chick would understand that connection?
>But I guess it's about why so we associate those with spiky shapes
I think the why just got a lot tricker than we imagined. Because we failed to replicate this experiment on other primates, we couldn't avoid a semantic suspicion about those associations. Now we probably have to set semantics aside or let it get a lot weirder, because we can replicate across ~300My.
>surely it's because they represent sharp immediate changes in frequency?
Maybe, and I think "multi-sensory signal processing" is the best framing, but the representation could also carry harder to think about things like "harm".
It's also super cool because the bouba-kiki effect framing was chosen due to methodological convenience for linguists and cultural anthropologists and their experimental bounds, not neuroscientists or signal processing folks. We could potentially find other experiments quickly, since chicks are a model organism and the mechanism is clear.
Things could move fast here.
This is just one micro-instance of a much larger thing. Brain encodes structural similarity across modalities. Corollary: language is far from arbitrary labels for things.
No, language is still pretty close to arbitrary labels. The handful of tenuous common threads like the bouba-kiki effect don't change the overall picture that much. The simple fact that language varies as much as it does is sufficient to prove that it's only loosely bound to anything universal.
https://en.wikipedia.org/wiki/True_name
>language is far from arbitrary labels for things
I think this is a misunderstanding of the arbitrariness of the sign. Arbitrary doesn't mean "random" or "uniformly sampled." The fact there are systematic tendencies among languages in how things are called doesn't negate the arbitrariness of the sign, they could have been called other things. We can also decide to refer to things by another name and we can use any arbitrary name we like! There is no limits on what names we can use (besides silly physiological constraints like having a word with 50 000 consonants). But, of course, there's much more to language than just labels!
For me, the interesting thing in this paper vis-à-vis language is that it shows how much innate structure in cognition must shape our language.
Arbitrariness of the sign is a principle that requires so many epicycles to present as "true" that it's more of a warning against overgeneralization than an insight with any significant predictive power in its own right.
I'm not entirely sold by this discovery. For example when you learn to train dogs, you learn about the 3 voices. Encouraging voice, atta boy, negative voice, more stern, and the big "NO!".
To some degree these words type sounding language are doing the same thing. Some sounds will irk, some will soothe, and it would affect this 'evidence' found.
I think the researchers agree with your premise. The “evidence” is not that chicks have more language understanding than previously understood, but rather that the source of the universality of bouba/kiki is due to something more primitive than built in human language hardware.
Preprint: https://www.biorxiv.org/content/10.1101/2024.05.17.594640v1....
I think it’s natural to think of this in terms of frequencies so the kiki shape has a higher visual frequency. As does the word have a higher audio frequencies within in than bouba so that is naturally associated with the lower frequency undulating line of that shape.
https://news.ycombinator.com/item?id=44280619 it's a good analogy.
I'm very intrigued by this, but I'll be much more interested when this is replicated on non-domesticated animals...!
It must take some strange things to survive co-evolution with humans for several thousands years
What's the N value of this study
I don’t know, but it really should be in units of N dozen.
From the preprint linked above:
> We tested a total of 42 subjects, 17 of which were females.
The published one repeated the experiment w/ day old chicks and IIRC the same number w/ the same results, so it's got a little more N than the preprint.
baba is keke
baba is you
All the universal translators in fiction make more sense now lol.
Believe it or not: This is pure and unadulterated advancement of civilization.
Please elaborate.
I looked it up, according to Google:
This phrase is a direct quote from the 1955 play (and 1960 film) Inherit the Wind, spoken by the character Henry Drummond (based on Clarence Darrow) regarding the teaching of evolution. It frames scientific education and intellectual freedom as the ultimate, pure progress of human civilization, contrasting with dogmatic resistance.
Context: The line refers to the 1925 Scopes "Monkey" Trial, which debated the legality of teaching evolution in Tennessee schools.
Significance: It serves as a dramatic defense of modernism, science, and freedom of thought against traditionalist views.
Cultural Impact: While based on historical events, the play uses this line to argue that intellectual inquiry is the cornerstone of advancement.
An interesting explanation that happens to be completely hallucinated. That line doesn't appear anywhere in either the play or the movie.
ha, ha, and they say AI does not hallucinate anymore!
Okay Gemini
If you don't recognize a quote, it's obviously AI? Might want to rethink your logic, or outsource it to AI. Might help you