I am delighted to share my new paper with coauthors @mh-christiansen.bsky.social, @erin-isbilen.bsky.social, and Dan Rubenstein. We argue that multimodality safeguards signal honesty by forming a multimodal gestalt that increases complexity and facilitates social costs on dishonesty.
A new theme issue of #PhilTransB examines the mechanisms of learning from social interaction. Read articles for free: buff.ly/K8v43YM
From birdsong to human language, acoustic communication by vocal learners involves the concatenation of sounds into sequences. Read 'The social origins of vocal sequences in songbirds and human infants': doi.org/10.1098/rstb... #PhilTransB #evolution
A new theme issue of #PhilTransB examines the mechanisms of learning from social interaction. Read articles for free: buff.ly/K8v43YM
11/11 When comparing across languages and species, learning faster vocal communication systems may well show a stronger developmental reliance on social feedback, which guides learners to compress their sequences into the range expected by their future conversation partners.
10/11 Our findings suggest that social feedback is more efficacious than previously imagined – it even guides improvements in fundamental vocal timing.
9/11 Changing our focus from individual sounds to sequences has revealed new influences of social learning on vocal development! The social facilitation effect held even when controlling for the amount of vocal practice.
8/11 In both 👶 and 🐧, the extent to which they compressed their vocal sequences over development predicted how quickly they developed mature communication (song similarity to their father for 🐧, and vocabulary development for 👶).
7/11 Does social feedback help 👶 and 🐧 learn to speed up their vocal rate to the range preferred by their communication partners? Yes! Responses to vocal sequences predicted the speed increases (compression) these learners made over early vocal development.
6/11 You know how it can be challenging to converse with someone who talks a little too slowly? Work by @davidpoeppel.bsky.social and others suggests that the brains of communicators have a preferred range of talking speed. However, early vocal learners start out with a slow rate of vocalizing.
5/11 How might 👶 and 🐧 use this social feedback loop to develop a vocal communication system with sequential structure at its foundation?
4/11 Parents’ responses, in turn, elicited infant vocal sequences above what we would expect by chance. When infants’ actions create changes in adults’ behavior, and those responses guide infants’ subsequent behaviors, this creates a social feedback loop during real-time interactions.
3/11 Starting with 👶, we discovered that babbling became more socially potent when in sequences! Sequences were significantly more likely to elicit a caregiver response than infants’ individual syllables.
2/11 w/ @mikehgoldstein.bsky.social and team, we studied both 👶 and 🐧, to find the extent to which social influences facilitate the emergence of sequential communication, in which individual vocal elements are concatenated.
1/11 Animals often communicate in phrases with sequential structure. Infant vocal learners, however, are often studied one note or syllable at a time. What are the origins of vocal sequences in animals that learn vocal communication? Find out in our new paper spanning songbirds and human infants! 🧵
Learning to speak takes place during a prolonged period of immaturity, which confers advantages for communicative development. Social partners, required for survival in early development, afford feedback for immature vocalizations like babbling and early speech. Feedback, in the form of changes to the linguistic structure of adult speech in response to infant vocalizations, may guide the earliest stages of language acquisition. In a cross-linguistic study of 1,586 transcripts, spanning 13 languages from 5 language families, we investigated whether caregiver talk was consistently influenced by children’s (aged 5–30 months) immature speech. Across languages, we found that most caregivers significantly simplified their linguistic structure in response to children’s immature speech, resulting in reduced lexical diversity, shorter utterance lengths, and higher likelihoods of single-word utterances. Children’s vocalizations elicited learnable language from caregivers, highlighting a potentially widespread feature of language use that is catalyzed by immature behavior. Thus, altriciality allows for immature speech to be a social tool, creating opportunities for learning during social interaction.
Cool new paper in @currentbiology.bsky.social — adults simplify their speech specifically in response to immature vocalizations. Seems like an important mechanism for how infant-directed vocalizations work, with a role for language learning
www.cell.com/current-biol...
SO very excited about new paper with @simonkirby.bsky.social and @ellengarland.bsky.social: We used infant-inspired tools to analyze eight years of humpback whale song, finding recurring parts with a Zipfian frequency distribution. www.science.org/doi/10.1126/...
We found a non-obvious pathway to robust language learnability across cultures & languages. Future work will assess just how widespread this pathway is across the world’s languages, and the role that contingent simplification plays in language development. 9/9
Languages don't only become learnable over evolution, they’re also learnable at precise moments during language development. Children actively shape their own learning input, eliciting learnable speech during vocal turn-taking, the central context of language use. 8/9
Languages must be learnable by the next generation, otherwise they would not exist. Here we show the simplification effect is robust across languages, contexts, and types of social interaction. Contingent simplified speech may be a cross-linguistic key to language learnability. 7/9
What this cross-cultural robustness implies is that the simplification effect of contingent speech is likely to be present in many more language than the 13 we studied in this paper. 6/9
A language we studied in this paper—Tseltal Mayan—was a particularly useful test case because Tseltal caregivers don’t use exaggerated child-directed speech typically found in industrialized societies. Despite this, we found a robust simplification in Tseltal caregiver speech. 5/9
Only parental speech that was contingent (produced immediately following their children’s vocalizations) was simplified, even though both contingent and non-contingent speech were similar in child-directed exaggerated pitch. 4/9
In response to their child’s immature speech, caregivers produce fewer unique words, shorter utterances and more single-word utterances compared to caregivers’ baseline speech complexity. We call this the simplification effect of contingent speech. 3/9
w/ @mikehgoldstein.bsky.social & Jacob Levy we show that children’s immature vocalizations & speech actively elicit language from caregivers that is linguistically simplified & more learnable. We find this in 13 languages, showing a robust social pathway for making language learning easier to do 2/9
How do languages become learnable for young children? Our new paper in Current Biology shows how “Immature vocalizations elicit simplified adult speech across multiple languages.” 🧵 of our findings below: www.cell.com/current-biol... 1/9
We found a non-obvious pathway to robust language learnability across cultures & languages. Future work will assess just how widespread this pathway is across the world’s languages, and the role that contingent simplification plays in language development. 9/9
Languages don't only become learnable over evolution, they’re also learnable at precise moments during language development. Children actively shape their own learning input, eliciting learnable speech during vocal turn-taking, the central context of language use. 8/9
Languages must be learnable by the next generation, otherwise they would not exist. Here we show the simplification effect is robust across languages, contexts, and types of social interaction. Contingent simplified speech may be a cross-linguistic key to language learnability. 7/9
What this cross-cultural robustness implies is that the simplification effect of contingent speech is likely to be present in many more language than the 13 we studied in this paper. 6/9