A. Grafen et Ra. Johnstone, WHY WE NEED ESS SIGNALING THEORY, Philosophical transactions-Royal Society of London. Biological sciences, 340(1292), 1993, pp. 245-250
Evolutionarily stable strategy (ESS) models of biological signalling a
re important because the intimate coevolution of signalling and receiv
ing strategies is complicated. Tentative results from a numerical stud
y of error-prone signalling show the value of formal modelling. Error
in perception can create discreteness in the distribution of signals p
roduced, and so observed discreteness in nature may call for no more c
omplicated explanation. Further developments in the theory of signalli
ng may include a link with theories of aggression such as the sequenti
al assessment game. The technical device of a 'scratch space' may allo
w a natural development of 'two-way' information games in which each c
ontestant plays the roles of signaller and receiver simultaneously. Th
is device may also incidentally derive mental states from purely strat
egic considerations.