Periodic Reporting for period 3 - COOPERATIVE PARTNER (Partner choice and the evolution of cooperation)
Período documentado: 2023-06-01 hasta 2025-03-31
Cooperation represents an evolutionary puzzle because natural selection is thought to favour selfish individuals over co-operators. However, theoretical studies and studies in humans show us that co-operators are preferred as social and sexual partners. Partner choice may therefore provide an important explanation for the evolution and stability of cooperation, alongside other known mechanisms, such as kin selection and self-serving benefits. However, we lack an understanding of whether partner choice is an important mechanism favouring cooperation in natural systems. Recent studies in captivity and some natural systems showed that animals prefer to associate with more cooperative individuals. An unequivocal test of the importance of partner choice would be to manipulate cooperation in the wild and test the associated fitness consequences. This project is pursuing this goal by combining a series of innovative experiments with a long-term study established on a highly cooperative wild bird, the sociable weaver Philetairus socius. New methodological developments are allowing us to conduct large-scale experiments in the wild and through detailed long-term tracking of individuals we can quantify life-long fitness consequences of choice. Specifically, here we are : i) using a new conceptual framework to test whether cooperative behaviour is reliable (e.g. repeatable or associated with the birds’ body condition, which is a prerequisite for being used in partner choice); ii) use state-of-the-art technology to manipulate cooperative behaviour and measure the resulting patterns of social and sexual partner choice; iii) use lifetime reproductive success and monitoring of experimental individuals over several years to examine the fitness benefits of partner choice for both co-operators and the individuals that associate with them. Finally, we will be using physiological measures (such as telomere dynamics) to investigate the mechanisms underlying the changes in fitness. Our ultimate aim is to provide a novel and robust evaluation of the roles of social and sexual selection on the evolution and maintenance of cooperation that can help us to understand the mechanisms underlying cooperation in nature, from invertebrates to humans.
The project is divided in three parts (see above). Central to all three parts, is the need to obtain detailed quantification of different cooperative behaviours at the individual level. This has required a significant upgrade from previous methods, which were essentially based on direct observations or manual video analyses, and were characterised by very low efficiency. Our main achievements so far mostly concern the development of these methods.
We are using artificial intelligence (AI) and remote sensing to develop new systems and procedures (led by L. Silva in collaboration with A. Ferreira) that have dramatically improved the speed and amount of data collected. These methods concern specifically the automation of 1) extracting behaviours and individual identity from video recordings, 2) quantifying and obtaining information about individual attributes through plumage patches or patterns (e.g. sex identification in a sexually monomorphic species; led by N. Silva in collaboration with A. Ferreira); 3) acoustic recording of individual vocalisations through the use of on-board microphones (i.e. miniature microphones attached to the individuals through a harness) and AI for processing the recordings (led by P. D’Amelio in collaboration with the Max Planck Institute for Ornithology).
These developments are now being used to address our scientific objectives. Specifically, to investigating whether some individuals have a higher propensity to cooperate than others, we are obtaining information on 1) nestling feeding by ‘helpers’, 2) cooperative nest building, 3) vigilance behaviour and 4) predator mobbing behaviour. Initial analyses on vigilance behaviour conducted by MSc student M. Marmelo revealed significant repeatability of this behaviour.
We also investigated whether a) the size of the sociable weaver black bib or b) a conspicuous type of vocalisation emitted by these birds correlate with their investment in cooperation. Bib size data were extracted automatically for >1000 individuals and analyses investigating whether this trait is linked to helping at the nest are underway. Vocalisations of 18 individuals were recorded using the on-board microphones system developed and which can record individuals continuously for >5 days.
We are using artificial intelligence (AI) and remote sensing to develop new systems and procedures (led by L. Silva in collaboration with A. Ferreira) that have dramatically improved the speed and amount of data collected. These methods concern specifically the automation of 1) extracting behaviours and individual identity from video recordings, 2) quantifying and obtaining information about individual attributes through plumage patches or patterns (e.g. sex identification in a sexually monomorphic species; led by N. Silva in collaboration with A. Ferreira); 3) acoustic recording of individual vocalisations through the use of on-board microphones (i.e. miniature microphones attached to the individuals through a harness) and AI for processing the recordings (led by P. D’Amelio in collaboration with the Max Planck Institute for Ornithology).
These developments are now being used to address our scientific objectives. Specifically, to investigating whether some individuals have a higher propensity to cooperate than others, we are obtaining information on 1) nestling feeding by ‘helpers’, 2) cooperative nest building, 3) vigilance behaviour and 4) predator mobbing behaviour. Initial analyses on vigilance behaviour conducted by MSc student M. Marmelo revealed significant repeatability of this behaviour.
We also investigated whether a) the size of the sociable weaver black bib or b) a conspicuous type of vocalisation emitted by these birds correlate with their investment in cooperation. Bib size data were extracted automatically for >1000 individuals and analyses investigating whether this trait is linked to helping at the nest are underway. Vocalisations of 18 individuals were recorded using the on-board microphones system developed and which can record individuals continuously for >5 days.
The methodological developments achieved thus far (e.g. massive increase in the efficiency of video analyses and extraction of behaviour and phenotypic data from video and photos) represents, in our view, a methodological breakthrough. The challenge now will be to develop a pipeline (or interface) that will rend these methods easy to use by the behavioural ecology community at large. These achievements were part of our initial ERC project and thus were not unplanned, although they were part of what made the ‘high-risk’ component of our project.
In addition, there are unexpected developments arising from these methods that are leading to highly innovative results. For example, we discovered that these birds' 'bib' (a plumage patch that is seemingly a sexually monomorphic trait, similar in size between males and females) encoded information about the sex of the individual birds. This is a novel result that opens the door to related discoveries concerning cryptic visual traits and individual attributes in this and other species.
Between now and the end of the project we aim to provide answers to the scientific objectives of the project. Namely, we will i) analyse whether there is individual repeatability in propensity to cooperate for the different behaviours being studied here, and we will also determine whether individuals are generally more cooperative or ‘specialise’ in some tasks; ii) determine whether individual condition (body mass and physiological stress) are associated with how much individuals cooperate; iii) manipulate social associations and individual body condition to measure the resulting patterns of social and sexual partner choice; iii) use lifetime reproductive success and monitoring of experimental individuals over several years to determine whether there fitness benefits arising from being more cooperative that are linked to partner choice, and this for both co-operators and the individuals that associate with them.
In addition, there are unexpected developments arising from these methods that are leading to highly innovative results. For example, we discovered that these birds' 'bib' (a plumage patch that is seemingly a sexually monomorphic trait, similar in size between males and females) encoded information about the sex of the individual birds. This is a novel result that opens the door to related discoveries concerning cryptic visual traits and individual attributes in this and other species.
Between now and the end of the project we aim to provide answers to the scientific objectives of the project. Namely, we will i) analyse whether there is individual repeatability in propensity to cooperate for the different behaviours being studied here, and we will also determine whether individuals are generally more cooperative or ‘specialise’ in some tasks; ii) determine whether individual condition (body mass and physiological stress) are associated with how much individuals cooperate; iii) manipulate social associations and individual body condition to measure the resulting patterns of social and sexual partner choice; iii) use lifetime reproductive success and monitoring of experimental individuals over several years to determine whether there fitness benefits arising from being more cooperative that are linked to partner choice, and this for both co-operators and the individuals that associate with them.