How social learning adds up to a culture: from birdsong to human public opinion

The demographic drivers of cultural evolution in bird song

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

Learning mimetic cuckoo call innovations from neighbors in a Chinese songbird

Some oscine passerines incorporate heterospecific sounds into their repertoires, including vocalizations of other bird species, sounds of other fauna, and even anthropogenic sounds, through vocal mimicry. However, few studies have investigated whether mimics learn heterospecific sounds from model species or from conspecific tutors. Here, we investigate mimicry acquisition using innovation in Cuculidae calls imitated by the Chinese blackbird (Turdus mandarinus). If the mimicry innovation arises and spreads among several neighbors and is not produced by model species, the mimicry must be acquired partially from conspecifics. We found that: (1) Cuculidae calls imitated by blackbirds were reasonably accurate, but with some differences between mimetic and real calls in acoustic structures. (2) We identified four unique mimetic units (mimicry innovation or copy errors), and these units only occurred at certain sites and were shared by several neighbors. In aggregate, frequency parameters (the first principal component) of unique mimetic units were higher than usual mimetic units (p < 0.001). Our findings provide further evidence that mimetic units can be partially learnt from conspecifics based on four cases of unique mimetic units. Our study and approach provide a reference and theoretical basis for the future understanding of social learning and development of vocal mimicry.

Machine learning reveals cryptic dialects that explain mate choice in a songbird

Culturally transmitted communication signals - such as human language or bird song - can change over time through cultural drift, and the resulting dialects may consequently enhance the separation of populations. However, the emergence of song dialects has been considered unlikely when songs are highly individual-specific, as in the zebra finch (Taeniopygia guttata). Here we show that machine learning can nevertheless distinguish the songs from multiple captive zebra finch populations with remarkable precision, and that 'cryptic song dialects' predict strong assortative mating in this species. We examine mating patterns across three consecutive generations using captive populations that have evolved in isolation for about 100 generations. We cross-fostered eggs within and between these populations and used an automated barcode tracking system to quantify social interactions. We find that females preferentially pair with males whose song resembles that of the females' adolescent peers. Our study shows evidence that in zebra finches, a model species for song learning, individuals are sensitive to differences in song that have hitherto remained unnoticed by researchers.

The Role of the Endogenous Opioid System in the Vocal Behavior of Songbirds and Its Possible Role in Vocal Learning

The opioid system in the brain is responsible for processing affective states such as pain, pleasure, and reward. It consists of three main receptors, mu- (μ-ORs), delta- (δ-ORs), and kappa- (κ-ORs), and their ligands – the endogenous opioid peptides. Despite their involvement in the reward pathway, and a signaling mechanism operating in synergy with the dopaminergic system, fewer reports focus on the role of these receptors in higher cognitive processes. Whereas research on opioids is predominated by studies on their addictive properties and role in pain pathways, recent studies suggest that these receptors may be involved in learning. Rodents deficient in δ-ORs were poor at recognizing the location of novel objects in their surroundings. Furthermore, in chicken, learning to avoid beads coated with a bitter chemical from those without the coating was modulated by δ-ORs. Similarly, μ-ORs facilitate long term potentiation in hippocampal CA3 neurons in mammals, thereby having a positive impact on spatial learning. Whereas these studies have explored the role of opioid receptors on learning using reward/punishment-based paradigms, the role of these receptors in natural learning processes, such as vocal learning, are yet unexplored. In this review, we explore studies that have established the expression pattern of these receptors in different brain regions of birds, with an emphasis on songbirds which are model systems for vocal learning. We also review the role of opioid receptors in modulating the cognitive processes associated with vocalizations in birds. Finally, we discuss the role of these receptors in regulating the motivation to vocalize, and a possible role in modulating vocal learning.

Unravelling the Stability of Nightingale Song Over Time and Space Using Open, Citizen Science and Shared Data

Open science approaches enable and facilitate the investigation of many scientific questions in bioacoustics, such as studies on the temporal and spatial evolution of song, as in vocal dialects. In contrast to previous dialect studies, which mostly focused on songbird species with a small repertoire, here we studied the common nightingale (Luscinia megarhynchos), a bird species with a complex and large repertoire. To study dialects on the population level in this species, we used recordings from four datasets: an open museum archive, a citizen science platform, a citizen science project, and shared recordings from academic researchers. We conducted to the date largest temporal and geographic dialect study of birdsong including recordings from 1930 to 2019 and from 13 European countries, with a geographical coverage of 2,652 km of linear distance. To examine temporal stability and spatial dialects, a catalog of 1,868 song types of common nightingales was created. Instead of dialects, we found a high degree of stability over time and space in both, the sub-categories of song and in the occurrence of song types. For example, the second most common song type in our datasets occurred over nine decades and across Europe. In our case study, open and citizen science data proved to be equivalent, and in some cases even better, than data shared by an academic research group. Based on our results, we conclude that the combination of diverse and open datasets was particularly useful to study the evolution of song in a bird species with a large repertoire.

Complex foraging behaviours in wild birds emerge from social learning and recombination of components

Recent well-documented cases of cultural evolution towards increasing efficiency in non-human animals have led some authors to propose that other animals are also capable of cumulative cultural evolution, where traits become more refined and/or complex over time. Yet few comparative examples exist of traits increasing in complexity, and experimental tests remain scarce. In a previous study, we introduced a foraging innovation into replicate subpopulations of great tits, the 'sliding-door puzzle'. Here, we track diffusion of a second 'dial puzzle', before introducing a two-step puzzle that combines both actions. We mapped social networks across two generations to ask if individuals could: (1) recombine socially-learned traits and (2) socially transmit a two-step trait. Our results show birds could recombine skills into more complex foraging behaviours, and naïve birds across both generations could learn the two-step trait. However, closer interrogation revealed that acquisition was not achieved entirely through social learning-rather, birds socially learned components before reconstructing full solutions asocially. As a consequence, singular cultural traditions failed to emerge, although subpopulations of birds shared preferences for a subset of behavioural variants. Our results show that while tits can socially learn complex foraging behaviours, these may need to be scaffolded by rewarding each component. This article is part of a discussion meeting issue 'The emergence of collective knowledge and cumulative culture in animals, humans and machines'.

The Role of the Learner in the Cultural Evolution of Vocalizations

As a uniquely human behavior, language is crucial to our understanding of ourselves and of the world around us. Despite centuries of research into how languages have historically developed and how people learn them, fully understanding the origin and evolution of language remains an ongoing challenge. In parallel, researchers have studied the divergence of birdsong in vocal-learning songbirds to uncover broader patterns of cultural evolution. One approach to studying cultural change over time, adapted from biology, focuses on the transmission of socially learned traits, including language, in a population. By studying how learning and the distribution of cultural traits interact at the population level, we can better understand the processes that underlie cultural evolution. Here, we take a two-fold approach to understanding the cultural evolution of vocalizations, with a focus on the role of the learner in cultural transmission. First, we explore previous research on the evolution of social learning, focusing on recent progress regarding the origin and ongoing cultural evolution of both language and birdsong. We then use a spatially explicit population model to investigate the coevolution of culture and learning preferences, with the assumption that selection acts directly on cultural phenotypes and indirectly on learning preferences. Our results suggest that the spatial distribution of learned behaviors can cause unexpected evolutionary patterns of learning. We find that, intuitively, selection for rare cultural phenotypes can indirectly favor a novelty-biased learning strategy. In contrast, selection for common cultural phenotypes leads to cultural homogeneity; we find that there is no selective pressure on learning strategy without cultural variation. Thus, counterintuitively, selection for common cultural traits does not consistently favor conformity bias, and novelty bias can stably persist in this cultural context. We propose that the evolutionary dynamics of learning preferences and cultural biases can depend on the existing variation of learned behaviors, and that this interaction could be important to understanding the origin and evolution of cultural systems such as language and birdsong. Selection acting on learned behaviors may indirectly impose counterintuitive selective pressures on learning strategies, and understanding the cultural landscape is crucial to understanding how patterns of learning might change over time.

Balanced imitation sustains song culture in zebra finches

Songbirds acquire songs by imitation, as humans do speech. Although imitation should drive convergence within a group and divergence through drift between groups, zebra finch songs sustain high diversity within a colony, but mild variation across colonies. We investigated this phenomenon by analyzing vocal learning statistics in 160 tutor-pupil pairs from a large breeding colony. Song imitation is persistently accurate in some families, but poor in others. This is not attributed to genetic differences, as fostered pupils copied their tutors’ songs as accurately or poorly as biological pupils. Rather, pupils of tutors with low song diversity make more improvisations compared to pupils of tutors with high song diversity. We suggest that a frequency dependent balanced imitation prevents extinction of rare song elements and overabundance of common ones, promoting repertoire diversity within groups, while constraining drift across groups, which together prevents the collapse of vocal culture into either complete uniformity or chaos.

Studying how songbirds learn songs can shed light on the development of human speech. An analysis of 160 tutor-pupil zebra finch pairs suggests that frequency dependent balanced imitation prevents the extinction of rare song elements and the overabundance of common ones, promoting song diversity within groups and species recognition across groups.

Mechanisms of Cultural Evolution in the Songs of Wild Bird Populations

Young songbirds draw the source material for their learned songs from parents, peers, and unrelated adults, as well as from innovation. These learned songs are used for intraspecific communication, and have well-documented roles for such functions as territory maintenance and mate attraction. The songs of wild populations differ, forming local "dialects" that may shift over time, suggesting that cultural evolution is at work. Recent work has focused on the mechanisms responsible for the cultural evolution of bird songs within a population, including drift, learning biases (such as conformity and rare-form copying), and selection (including sexual selection). In many songs or song repertoires, variability is partitioned, with some songs or song segments being stable and consistent, while others vary within the population and across time, and still others undergo population-wide transitions over time. This review explores the different mechanisms that shape the cultural evolution of songs in wild populations, with specific reference to a long-term investigation of a single population of philopatric Savannah sparrows. Males learn a single four-segment song during their 1st year and sing the same song thereafter. Within this song, the buzz segment is a population marker, and may be stable for decades - variant forms occur but eventually disappear. In contrast, the middle segment is highly variable both within the population and over time; changes in relative prevalence of different forms may be due to cultural drift or a rare-form learning bias. Within the introductory segment, a high note cluster was replaced by a click train between 1982 and 2010, following an S-shaped trajectory characteristic of both selective sweeps in population genetics and the replacement of one form by another in human language. In the case of the Savannah sparrows, this replacement may have been due to sexual selection. In subsequent generations, the number of clicks within trains increased, a form of cultural directional selection. In contrast to the narrowing of a trait's range during directional selection in genetic systems, variation in the number of clicks in a train increased as the mean value shifted because improvisation during song learning allowed the range of the trait to expand. Thus, in the single short song of the Savannah sparrow, at least four different mechanisms appear to contribute to three different types of cultural evolutionary outcomes. In the future, it will be import to explore the conditions that favor the application of specific (and perhaps conditional) learning rules, and studies such as the ongoing song seeding experiment in the Kent Island Savannah sparrow population will help in understanding the mechanisms that promote or repress changes in a population's song.

Towards an integrated view of vocal development

Vocal development is usually studied from the perspective of neuroscience. In this issue, Zhang and Ghazanfar propose a way in which body growth might condition the process. They study the vocalizations of marmoset infants with a wide range of techniques that include computational models and experiments that mimic growth reversal. Their results suggest that the qualitative changes that occur during development are rooted in the nonlinear interaction between the nervous system and the biomechanics involved in respiration. This work illustrates how an integrative approach enriches our understanding of behavior.