When honesty and cheating pay off: The evolution of honest and dishonest equilibria in a conventional signalling game

Research output: Contribution to journalArticle


Background: The reliability of signals is a key issue in the study of animal communication. Both empirical work and theoretical models show that communication need not be entirely honest, and thus signals can be deceitful. Aggressive communication appears to be a prime candidate for such deceitful communication, because bluffing has been described in several species. Bluffing in these situations are supposed to be maintained by frequency dependent selection where the fitness of a given type depends on the frequencies of the other types in the population. Previous efforts to model such a scenario through individual based simulations have yielded conflicting results. Studies have either found a rich set of polymorphic strategies including the traditional cheating scenario or found none. Thus, the modelling assumptions responsible for these diverging conclusions remain unclear. Results: In this study, I investigate the effects of four modelling assumptions: the role of an extended strategy set, the initial population composition (seeding), the differences in pay-offs and finally different parameter spaces. I investigate the effects of these factors on the evolvability of both honest and mixed cheating strategies. I show that both honest and cheating equilibria readily evolve and that the investigated parameter range and the seeding of the starting populations have the greatest influence on the outcome. Conclusions: Both honest signalling and polymorphic cheating equilibria are more likely to evolve from a narrow strategy set than from a random mixture of strategies. A large potential strategy set is not a setback for the evolution of communication -honest or cheating- as long as the initial population is seeded with only a few strategies. In addition, different sections of the parameter space show consistently different behaviour. Thus, frequency dependent selection has the potential to explain various empirical observations that show consistent differences in aggressive behaviour.

Original languageEnglish
Article number270
JournalBMC Evolutionary Biology
Issue number1
Publication statusPublished - Dec 28 2017


  • Aggressive communication
  • Animal personality
  • Deception
  • Frequency dependent selection
  • Honest signalling
  • Mixed ESS

ASJC Scopus subject areas

  • Ecology, Evolution, Behavior and Systematics

Fingerprint Dive into the research topics of 'When honesty and cheating pay off: The evolution of honest and dishonest equilibria in a conventional signalling game'. Together they form a unique fingerprint.

  • Cite this