A5006038003- Evolutionary Game Theory and Cooperation
- T-cell and B-cell Immunology
- Culture, Economy, and Development Studies
- Experimental Behavioral Economics Studies
- Immune Cell Function and Interaction
- Language and cultural evolution
- Evolutionary Psychology and Human Behavior
- Neural dynamics and brain function
- Evolution and Genetic Dynamics
- Immunotherapy and Immune Responses
- Advanced Memory and Neural Computing
- Game Theory and Applications
- Neural Networks and Applications
- Mathematical and Theoretical Epidemiology and Ecology Models
- Animal Behavior and Reproduction
- Immune responses and vaccinations
- Psychology of Moral and Emotional Judgment
- Auction Theory and Applications
- Avian ecology and behavior
- Economic theories and models
- Reinforcement Learning in Robotics
- Optimization and Search Problems
- Primate Behavior and Ecology
- Evolutionary Algorithms and Applications
- Opinion Dynamics and Social Influence
University of Glasgow
2017-2019
Institute of Infection and Immunity
2019
Stockholm University
2014-2017
University of St Andrews
2014-2016
Queen's University
2009-2010
It Pays to Be a Copy Cat Does it pay copy what others do? Rendell et al. (p. 208 ) elected Robert Axelrod's 1979 tournament in which strategies for playing the iterated prisoner's dilemma game were pitted against each other until an overall winner emerged—the tit-for-tat strategy. In 2008 tournament, 100 social learning designed cope with changing environment competed other; winning strategy involved sampling behaviors of players periodically, rather than exploring alone.
Abstract The brain processes information through multiple layers of neurons. This deep architecture is representationally powerful, but complicates learning because it difficult to identify the responsible neurons when a mistake made. In machine learning, backpropagation algorithm assigns blame by multiplying error signals with all synaptic weights on each neuron’s axon and further downstream. However, this involves precise, symmetric backward connectivity pattern, which thought be...
Characterising the longevity of immunological memory requires establishing rules underlying renewal and death peripheral T cells. However, we lack knowledge population structure how self-renewal de novo influx contribute to maintenance compartments. Here, characterise kinetics murine CD4 cell subsets by measuring rates new cells using detailed timecourses DNA labelling that also distinguish behaviour recently divided quiescent We find both effector central comprise subpopulations with highly...
Correlations in family size across generations could have a major influence on human population the future. Empirical studies shown that associations between fertility of parents and children are substantial growing over time. Despite their potential long-term consequences, intergenerational correlations largely been ignored by researchers. We present model transition as cultural process acting new lifestyles associated with fertility. Differences parental social influences acquisition these...
The brain processes information through many layers of neurons. This deep architecture is representationally powerful, but it complicates learning by making hard to identify the responsible neurons when a mistake made. In machine learning, backpropagation algorithm assigns blame neuron computing exactly how contributed an error. To do this, multiplies error signals matrices consisting all synaptic weights on neuron's axon and farther downstream. operation requires precisely choreographed...
Social learning is common in nature, yet cumulative culture (where knowledge and technology increase complexity diversity over time) appears restricted to humans. To understand why, we organized a computer tournament which programmed entries specified when learn new refine (i.e. improve) existing knowledge. The revealed ‘refinement paradox’: refined behavior afforded higher payoffs as individuals converged on small number of successful behavioral variants, but refining did not generally pay....
Recent work on the evolution of behaviour is set in a structured population, providing systematic way to describe gene flow and behavioural interactions. To obtain analytical results one needs structure with considerable regularity. Our apply such "homogeneous" structures (e.g., lattices, cycles, island models). This regularity has been formally described by "node-transitivity" condition but mathematics, internal symmetry powerfully theory mathematical groups. Here, this provides elegant...
Laboratory mice develop populations of circulating memory CD4+ T cells in the absence overt infection. We have previously shown that these are replenished from naive precursors at high levels throughout life (Gossel et al., 2017). However, nature, relative importance and timing forces generating remain unclear. Here, we tracked generation cell subsets housed facilities differing their 'dirtiness'. found evidence for sequential to central effector development, confirmed both heterogeneous...
In certain economic experiments, some participants willingly pay a cost to punish peers who contribute too little the public good.Because such punishment can lead improved group outcomes, this costly has been conceived of as altruistic.Here we provide evidence that individual variation in propensity low contributions is unrelated altruism.First, use was uncorrelated with contribution good, contrary hypothesis punishers are proximally motivated by prosocial preferences.Second, positively...
Abstract Social learning is common in nature, yet cumulative culture (where knowledge and technology increase complexity diversity over time) appears restricted to humans. To understand why, we organized a computer tournament which programmed entries specified when learn new refine (i.e. improve) existing knowledge. The revealed ‘refinement paradox’: refined behavior afforded higher payoffs as individuals converged on small number of successful behavioral variants, but refining did not...
In a novel multiplayer extension of the famous secretary problem, multiple players seek to employ secretaries from common labour pool. Secretaries do not accept being put on hold, always job offers immediately, and leave pool once rejected by single player. All have an identical preference for secretaries, all optimize probability obtaining best n secretaries. We find that in Nash equilibrium, as number, N, searching grows, optimal strategy converges simple function N. For two-player case we...
Ineffective, aversive and harmful medical treatments are common cross-culturally, historically today. Using evolutionary game theory, we develop the following model to explain their persistence. Humans often incapacitated by illness injury, unusually dependent on care from others during convalescence. However, such caregiving is vulnerable exploitation via deception, whereby people feign or exaggerate in order gain access care. Our demonstrates that can counter-intuitively increase range of...
A requirement of culture, whether animal or human, is some degree conformity behavior within populations. Researchers gene-culture coevolution have suggested that population level may result from frequency-biased social learning: individuals sampling multiple role models and preferentially adopting the majority in sample. When learning a single model, frequency-bias not possible. We show why population-level trend, either conformist anticonformist, nonetheless be almost inevitable learn...
In certain economic experiments, some participants willingly pay a cost to punish peers who contribute too little the public good. Because such punishment can lead improved group outcomes, this costly has been conceived of as altruistic. Here we provide evidence that individual variation in propensity low contributions is unrelated altruism. First, use was uncorrelated with contribution good, contrary hypothesis punishers are proximally motivated by prosocial preferences. Second, positively...
Abstract When animals compete, hierarchies can emerge. If the outcome of competition under different conditions is dependent upon sets attributes, then we may expect to see that are domain‐specific, rather than domain general. We tested this idea by comparing prey share within shoals sticklebacks Gasterosteus aculeatus as they foraged for patchily distributed or drifting prey. found was correlated across pairs patch‐ and drift‐foraging trials, but not between two conditions, suggesting...
Abstract Laboratory mice develop populations of circulating memory CD4 + T cells in the absence overt infection. We have previously shown that these are replenished from naive precursors at high levels throughout life (Gossel et al., 2017). However, nature, relative importance and timing forces generating remain unclear. Here, we tracked generation cell subsets housed facilities differing their ‘dirtiness’. found evidence for sequential to central effector development, confirmed both...