Language-like efficiency and structure in house finch song

Proceedings of the Royal Society B Biological Sciences, Journal Year: 2024, Volume and Issue: 291(2020)

Published: April 2, 2024

Communication needs to be complex enough functional while minimizing learning and production costs. Recent work suggests that the vocalizations gestures of some songbirds, cetaceans great apes may conform linguistic laws reflect this trade-off between efficiency complexity. In studies non-human communication, though, clustering signals into types cannot done a priori , decisions about appropriate grain analysis affect statistical in data. The aim study was assess evidence for language-like structure house finch ( Haemorhous mexicanus ) song across three levels granularity syllable clustering. results show strong Zipf's rank–frequency law, law abbreviation Menzerath's law. Additional analyses songs have small-world structure, thought systematic syntax, mutual information decay sequences is consistent with combination Markovian hierarchical processes. These patterns are robust clustering, pointing limited form scale invariance. sum, it appears has been shaped by pressure efficiency, possibly offset costs female preferences

Language: Английский

---

Negativity bias in the spread of voter fraud conspiracy theory tweets during the 2020 US election

Humanities and Social Sciences Communications, Journal Year: 2023, Volume and Issue: 10(1)

Published: Sept. 14, 2023

Abstract During the 2020 US presidential election, conspiracy theories about large-scale voter fraud were widely circulated on social media platforms. Given their scale, persistence, and impact, it is critically important to understand mechanisms that caused these spread. The aim of this preregistered study was investigate whether retweet frequencies among proponents Twitter during election are consistent with frequency bias and/or content bias. To do this, we conducted generative inference using an agent-based model cultural transmission VoterFraud2020 dataset. results show observed distribution a strong causing users preferentially tweets negative emotional valence. Frequency information appears be largely irrelevant future count. Follower count strongly predicts in simpler linear but does not appear drive overall after temporal dynamics accounted for. Future studies could apply our methodology comparative framework assess for valence theory messages differs from other forms media.

Language: Английский

---

The demographic drivers of cultural evolution in bird song

Current Biology, Journal Year: 2025, Volume and Issue: unknown

Published: March 1, 2025

Highlights•Analyzed over 100,000 great tit songs using deep learning to study vocal culture•Tested if local demographic variation and processes impact song culture in the wild•Showed that age, dispersal, turnover affect cultural diversity changeSummarySocial can give rise shared behavioral patterns persist as within animal communities,1,2 such bird whale cetacean feeding techniques.3,4,5 These traits evolve6,7,8,9 individual survival, population structure, conservation efforts.10 Although theoretical work indicates processes—like turnover, immigration, age structure—significantly influence evolution,11,12,13 empirical evidence from natural populations is limited. Using metric analyze >400 repertoires tits (Parus major), we show affects cultures small spatial temporal scales where occurs. Within-population dispersal homogenizes culture, immigrant birds adopt while increasing neighborhood through larger repertoires. Birds of similar tend have more repertoires, which provides change, with mixed-age neighborhoods showing higher diversity. We estimate a main driver change its pace also moderated by structure. findings support expectations regarding key role evolution highlighting their interaction species-specific factors timing mode acquisition.Graphical abstract

Language: Английский

---

The growing methodological toolkit for identifying and studying social learning and culture in non-human animals

Philosophical Transactions of the Royal Society B Biological Sciences, Journal Year: 2025, Volume and Issue: 380(1925)

Published: May 1, 2025

There is a growing consensus that animals' socially transmitted knowledge should be recognized when planning conservation management, but demonstrating social learning or culture can present considerable challenges, especially in the wild. Fortunately, decades of research have spawned rich methodological toolkit for exactly this purpose. Here, we review principal approaches, including: experiments; analyses natural experimentally seeded diffusions novel behaviours, sometimes using specialist statistical techniques; mapping behavioural variation across neighbouring, sympatric captive groups, at larger scales; and assessment aspects cross-generational transmission, including teaching, during ontogenetic development cumulative change. Some methods reviewed were developed studies, subsequently been adapted application wild, are useful exploring species' general propensity to learn transmit information socially. We highlight several emerging 'rapid assessment' approaches-including camera trapping, passive acoustic monitoring, animal-borne tags, AI-assisted data mining computer simulations-that prove addressing particularly urgent needs. conclude by considering how best use practice, guide further on animal cultures, maximize policy impact.This article part theme issue 'Animal culture: changing world'.

Language: Английский

---

pykanto: A python library to accelerate research on wild bird song

Methods in Ecology and Evolution, Journal Year: 2023, Volume and Issue: 14(8), P. 1994 - 2002

Published: June 8, 2023

Abstract Studying the vocalisations of wild animals can be a challenge due to limitations traditional computational methods, which often are time‐consuming and lack reproducibility. Here, I present pykanto, new software package that provides set tools build, manage, explore large sound databases. It automatically find discrete units in animal vocalisations, perform semi‐supervised labelling individual repertoires with interactive web app feed data deep learning models. pykanto used streamline research on, for example, vocal signatures acoustic similarity between individuals populations. To demonstrate its capabilities, put library test on male great tits Wytham Woods, near Oxford, UK. The results show identities birds accurately determined from their songs use improves efficiency reproducibility process.

Language: Английский

---

A densely sampled and richly annotated acoustic data set from a wild bird population

Animal Behaviour, Journal Year: 2024, Volume and Issue: 211, P. 111 - 122

Published: March 25, 2024

We present a high-resolution, densely sampled data set of wild bird songs collected over multiple years from single population great tits, Parus major, in the U.K. The includes 1 100 000 individual acoustic units 109 963 richly annotated songs, sung by more than 400 birds, and provides unprecedented detail on vocal behaviour birds. Here, we describe collection processing procedures provide summary data. also discuss potential research questions that can be addressed using this set, including behavioural repeatability stability, links between performance reproductive success, timing song production, syntactic organization production learning wild. have made associated software tools publicly available with aim other researchers benefit resource use it to further our understanding

Language: Английский

---

Statistical signals of copying are robust to time- and space-averaging

Evolutionary Human Sciences, Journal Year: 2023, Volume and Issue: 5

Published: Jan. 1, 2023

Cattle brands (ownership marks left on animals) are subject to forces influencing other graphic codes: the copying of constituent parts, pressure for distinctiveness and complexity. The historical record cattle in some US states is complete owing legal registration, providing a unique opportunity assess how sampling processes leading time- space-averaging influence our ability make inferences from limited datasets fields like archaeology. In this preregistered study, we used dataset ~81,000 Kansas (1990-2016) explore two aspects: (1) relative copying, complexity creation diffusion brand components; (2) effects statistical signals. By conducting generative inference with an agent-based model, found that patterns data consistent intermediate addition, by comparing mixed structured datasets, these signals robust to, possibly boosted by, space-averaging.

Language: Английский

---

Cultural Evolution and Music

Oxford University Press eBooks, Journal Year: 2023, Volume and Issue: unknown

Published: April 20, 2023

Abstract The universality and diversity of music in human societies make it an important research model for understanding how cultural features change over time space. In this chapter, we review on the evolution music, broken down into three major approaches: (i) corpus-based approaches that use large datasets to infer evolutionary patterns, (ii) experimental explore transmission transformation, (iii) ‘music-like’ behaviors non-human species, such as bird whale song, highlights shared mechanisms future directions. Finally, discuss applications issues like musical endangerment, copyright enforcement, algorithmic inequality. Given have yet be fully leveraged, think has potential become a powerful evolution.

Language: Английский

---

Modeling how population size drives the evolution of birdsong, a functional cultural trait

Evolution, Journal Year: 2022, Volume and Issue: 76(6), P. 1139 - 1152

Published: April 11, 2022

Oscine songbirds have been an important study system for social learning, particularly because their learned songs provide analog human languages and music. Here, we propose a different analogy: from evolutionary perspective, could birds' change over time more like arrowheads than arias? Small improvements to bird's song can lead large fitness differences its singer, which make analogous tools languages. We modify model of tool evolution accommodate cultural birdsong: each learner chooses the most skilled available tutor emulate, is likely produce inferior copy superior one. Similar evolution, our suggests that larger populations birds foster in time, even when learners restrict pool tutors subset individuals network. also demonstrate elements be simplified instead lost after population bottlenecks if lower quality traits are easier imitate higher ones. show these processes plausibly generate empirically observed patterns some traits, predictions about types shrink. More broadly, aim connect modeling approaches used nonhuman systems, moving toward cohesive theoretical framework accounts both cognitive demographic processes.

Language: Английский

---

Bridging theory and data: A computational workflow for cultural evolution

Proceedings of the National Academy of Sciences, Journal Year: 2024, Volume and Issue: 121(48)

Published: Nov. 18, 2024

Cultural evolution applies evolutionary concepts and tools to explain the change of culture over time. Despite advances in both theoretical empirical methods, connections between cultural theory evidence are often vague, limiting progress. Theoretical models influence research but rarely guide data collection analysis logical transparent ways. themselves too abstract apply specific contexts statistical inference. To help bridge this gap, we outline a quality-assurance computational workflow that starts from generative phenomena logically connects estimates real-world explanatory goals. We emphasize demonstrate validation using synthetic data. Using interplay conformity, migration, diversity as case study, present coded repeatable examples directed acyclic graphs, tailored agent-based simulations, probabilistic transmission model for longitudinal data, an approximate Bayesian computation cross-sectional discuss assumptions, opportunities, pitfalls different approaches modeling show how each can be used improve depending on structure available depth understanding. Throughout, highlight significance ethnography collecting basic demographic information about study populations call more emphasis theory-driven workflows part science reform.

Language: Английский

---