Signalling theory
Within evolutionary biology, signalling theory is a body of theoretical work examining communication between individuals, both within species and across species. The central question is how organisms with conflicting interests, such as in sexual selection, are expected to provide honest signals rather than deceive or cheat, given that the passing on of pleiotropic traits is subject to natural selection, which aims to minimize associated costs without assuming any conscious intent. Mathematical models describe how signalling can contribute to an evolutionarily stable strategy.
Signals are given in contexts such as mate selection by females, which subjects the advertising males' signals to selective pressure. Signals thus evolve because they modify the behaviour of the receiver to benefit the signaller. Signals may be honest, conveying information which usefully increases the fitness of the receiver, or dishonest. An individual can cheat by giving a dishonest signal, which might briefly benefit that signaller, at the risk of undermining the signalling system for the whole population.
The question of whether the selection of signals works at the level of the individual organism or gene, or at the level of the group, has been debated by biologists such as Richard Dawkins, arguing that individuals evolve to signal and to receive signals better, including resisting manipulation. Amotz Zahavi suggested that cheating could be controlled by the handicap principle, where the best horse in a handicap race is the one carrying the largest handicap weight. According to Zahavi's theory, signallers such as male peacocks have "tails" that are genuinely handicaps, being costly to produce. The system is evolutionarily stable as the large showy tails are honest signals. Biologists have attempted to verify the handicap principle, but with inconsistent results. The mathematical biologist Ronald Fisher analysed the contribution that having two copies of each gene would make to honest signalling, demonstrating that a runaway effect could occur in sexual selection. The evolutionary equilibrium depends sensitively on the balance of costs and benefits.
The same mechanisms can be expected in humans, where researchers have studied behaviours including risk-taking by young men, hunting of large game animals, and costly religious rituals, finding that these appear to qualify as costly honest signals.
Sexual selection
When animals choose mating partners, traits such as signalling are subject to evolutionary pressure. For example, the male gray tree frog, Dryophytes versicolor, produces a call to attract females. Once a female chooses a mate, this selects for a specific style of male calling, thus propagating a specific signalling ability. The signal can be the call itself, the intensity of a call, its variation style, its repetition rate, and so on. Various hypotheses seek to explain why females would select for one call over the other. The sensory exploitation hypothesis proposes that pre-existing preferences in female receivers can drive the evolution of signal innovation in male senders, in a similar way to the hidden preference hypothesis which proposes that successful calls are better able to match some 'hidden preference' in the female. Signallers have sometimes evolved multiple sexual ornaments, and receivers have sometimes evolved multiple trait preferences.Honest signals
In biology, signals are traits, including structures and behaviours, that have evolved specifically because they change the behaviour of receivers in ways that benefit the signaller. Traits or actions that benefit the receiver exclusively are called "cues". For example, when an alert bird deliberately gives a warning call to a stalking predator and the predator gives up the hunt, the sound is a "signal". But when a foraging bird inadvertently makes a rustling sound in the leaves that attracts predators and increases the risk of predation, the sound is not a signal, but a cue.Signalling systems are shaped by mutual interests between signallers and receivers. An alert bird such as a Eurasian jay warning off a stalking predator is communicating something useful to the predator: that it has been detected by the prey; it might as well quit wasting its time stalking this alerted prey, which it is unlikely to catch. When the predator gives up, the signaller can get back to other tasks such as feeding. Once the stalking predator is detected, the signalling prey and receiving predator thus have a mutual interest in terminating the hunt.
Within species, mutual interests increase with kinship. Kinship is central to models of signalling between relatives, for instance when broods of nestling birds beg and compete for food from their parents.
File:yellow-banded.poison.dart.frog.arp.jpg|thumb|upright|The yellow-banded poison dart frog, Dendrobates leucomelas, gives an honest signal of its toxicity to warn off predators and reduce the frog's risk of injury.
The term honesty in animal communication is controversial because in non-technical usage it implies intent, to discriminate deception from honesty in human interactions. However, biologists use the phrase "honest signals" in a direct, statistical sense. Biological signals, like warning calls or resplendent tail feathers, are honest if they reliably convey useful information to the receiver. That is, the signal trait tells the receiver about an otherwise unobservable factor. Honest biological signals do not need to be perfectly informative, reducing uncertainty to zero; all they need to be useful is to be correct "on average", so that some behavioural response to the signal is advantageous, statistically, compared to the behaviour that would occur in absence of the signal. Ultimately the value of the signalled information depends on the extent to which it allows the receiver to increase its fitness.
One type of honest signal is the signalling of quality in sexually reproducing animals. In sexually reproducing animals one sex is generally the 'choosing sex' and the other the 'advertising sex'. The choosing sex achieves the highest fitness by choosing the partner of the highest quality. This quality cannot be observed directly, so the advertising sex can evolve a signal, which advertises its quality. Examples of these signals include the tail of a peacock and the colouration of male sticklebacks. Such signals only work, i.e. are reliable, if the signal is honest. The link between the quality of the advertising sex and the signal may depend on environmental stressors, with honesty increasing in more challenging environments.
Another type of honest signal is the aposematic warning signal, generally visual, given by poisonous or dangerous animals such as wasps, poison dart frogs, and pufferfish. Warning signals are honest indications of noxious prey, because conspicuousness evolves in tandem with noxiousness. Thus, the brighter and more conspicuous the organism, the more toxic it usually is. The most common and effective colours are red, yellow, black and white.
The mathematical biologist John Maynard Smith discusses whether honest signalling must always be costly. He notes that it had been shown that "in some circumstances" a signal is reliable only if it is costly. He states that it had been assumed that parameters such as pay-offs and signalling costs were constant, but that this might be unrealistic. He states that with some restrictions, signals can be cost-free, reliable, and evolutionarily stable. However, if costs and benefits "vary uniformly over the whole range" then indeed honest signals have to be costly.
Dishonest signals
Because there are both mutual and conflicting interests in most animal signalling systems, a central problem in signalling theory is dishonesty or cheating. For example, if foraging birds are safer when they give a warning call, cheats could give false alarms at random, just in case a predator is nearby. But too much cheating could cause the signalling system to collapse. Every dishonest signal weakens the integrity of the signalling system, and so reduces the fitness of the group. An example of dishonest signalling comes from Fiddler crabs such as Austruca mjoebergi, which have been shown to bluff about their fighting ability. When a claw is lost, a crab occasionally regrows a weaker claw that nevertheless intimidates crabs with smaller but stronger claws. The proportion of dishonest signals is low enough for it not to be worthwhile for crabs to test the honesty of every signal through combat.Richard Dawkins and John Krebs in 1978 considered whether individuals of the same species would act as if attempting to deceive each other. They applied a "selfish gene" view of evolution to animals' threat displays to see if it would be in their genes' interests to give dishonest signals. They criticised previous ethologists, such as Nikolaas Tinbergen and Desmond Morris, for suggesting that such displays were "for the good of the species". They argued that such communication ought to be viewed as an evolutionary arms race in which signallers evolve to become better at manipulating receivers, while receivers evolve to become more resistant to manipulation. The game theoretical model of the war of attrition similarly suggests that threat displays ought not to convey any reliable information about intentions.
Deceptive signals can be used both within and between species. Perhaps the best-known example of inter-species deception is mimicry: when individuals of one species mimic the appearance or behaviour of individuals of another species. A variety of mimicry types exist, including Batesian, Müllerian, host mimicry and "aggressive" mimicry. A very frequent type is ant mimicry. Deception within species can be bluffing or sexual mimicry where males or females mimic the patterns and behaviour of the opposite sex. A well-known example is the bluegill sunfish where mimic males look like and behave like females in order to sneak into the guarded nests of territorial males and fertilize some of the eggs.