Are aposematic signals honest? A review

We explore the relevance of honest signalling theory to the evolution of aposematism. We begin with a general consideration of models of signal stability, with a focus on the Zahavian costly signalling framework. Next, we review early models of signalling in the context of aposematism (some that are consistent and some inconsistent with costly honest signalling). We focus on controversies surrounding the idea that aposematic signals are handicaps in a Zahavian framework. Then, we discuss how the alignment of interests between signaller and predator influences the evolution of aposematism, highlight the distinction between qualitative and quantitative honesty and review theory and research relevant to these categories. We also review recent theoretical treatments of the evolution of aposematism that have focused on honest signalling as well as empirical research on a variety of organisms, including invertebrates and frogs. Finally, we discuss future directions for empirical and theoretical research in this area.     

Separating equilibria in continuous signalling games

Much of the literature on costly signalling theory concentrates on separating equilibria of continuous signalling games. At such equilibria, every signaller sends a distinct signal, and signal receivers are able to exactly infer the signaller's condition from the signal sent. In this paper, we introduce a vector-field solution method that simplifies the process of solving for separating equilibria. Using this approach, we show that continuous signalling games can have low-cost separating equilibria despite conflicting interests between signaller and receiver. We find that contrary to prior arguments, honesty does not require wasteful signals. Finally, we examine signalling games in which different signallers have different minimal-cost signals, and provide a mathematical justification for the argument that even non-signalling traits will be exaggerated beyond their phenotypic optimum when they are used by other individuals to judge condition or quality.     

Partner choice creates competitive altruism in humans

Reciprocal altruism has been the backbone of research on the evolution of altruistic behaviour towards non-kin, but recent research has begun to apply costly signalling theory to this problem. In addition to signalling resources or abilities, public generosity could function as a costly signal of cooperative intent, benefiting altruists in terms of (i) better access to cooperative relationships and (ii) greater cooperation within those relationships. When future interaction partners can choose with whom they wish to interact, this could lead to competition to be more generous than others. Little empirical work has tested for the possible existence of this 'competitive altruism'. Using a cooperative monetary game with and without opportunities for partner choice and signalling cooperative intent, we show here that people actively compete to be more generous than others when they can benefit from being chosen for cooperative partnerships, and the most generous people are correspondingly chosen more often as cooperative partners. We also found evidence for increased scepticism of altruistic signals when the potential reputational benefits for dishonest signalling were high. Thus, this work supports the hypothesis that public generosity can be a signal of cooperative intent, which people sometimes 'fake' when conditions permit it.     

Error-proneness as a handicap signal

This paper describes two discrete signalling models in which the error-proneness of signals can serve as a handicap signal. In the first model, the direct handicap of sending a high-quality signal is not large enough to assure that a low-quality signaller will not send it. However, if the receiver sometimes mistakes a high-quality signal for a low-quality one, then there is an indirect handicap to sending a high-quality signal. The total handicap of sending such a signal may then still be such that a low-quality signaller would not want to send it. In the second model, there is no direct handicap of sending signals, so that nothing would seem to stop a signaller from always sending a high-quality signal. However, the receiver sometimes fails to detect signals, and this causes an indirect handicap of sending a high-quality signal that still stops the low-quality signaller of sending such a signal. The conditions for honesty are that the probability of an error of detection is higher for a high-quality than for a low-quality signal, and that the signaller who does not detect a signal adopts a response that is bad to the signaller. In both our models, we thus obtain the result that signal accuracy should not lie above a certain level in order for honest signalling to be possible. Moreover, we show that the maximal accuracy that can be achieved is higher the lower the degree of conflict between signaller and receiver. As well, we show that it is the conditions for honest signalling that may be constraining signal accuracy, rather than the signaller trying to make honest signals as effective as possible given receiver psychology, or the signaller adapting the accuracy of honest signals depending on his interests.     

Do aggressive signals evolve towards higher reliability or lower costs of assessment?

It has been suggested that the evolution of signals must be a wasteful process for the signaller, aimed at the maximization of signal honesty. However, the reliability of communication depends not only on the costs paid by signallers but also on the costs paid by receivers during assessment, and less attention has been given to the interaction between these two types of costs during the evolution of signalling systems. A signaller and receiver may accept some level of signal dishonesty by choosing signals that are cheaper in terms of assessment but that are stabilized with less reliable mechanisms. I studied the potential trade‐off between signal reliability and the costs of signal assessment in the corncrake (Crex crex). I found that the birds prefer signals that are less costly regarding assessment rather than more reliable. Despite the fact that the fundamental frequency of calls was a strong predictor of male size, it was ignored by receivers unless they could directly compare signal variants. My data revealed a response advantage of costly signals when comparison between calls differing with fundamental frequencies is fast and straightforward, whereas cheap signalling is preferred in natural conditions. These data might improve our understanding of the influence of receivers on signal design because they support the hypothesis that fully honest signalling systems may be prone to dishonesty based on the effects of receiver costs and be replaced by signals that are cheaper in production and reception but more susceptible to cheating.     

Attention-Seeking Displays

Animal communication abounds with extravagant displays. These signals are usually interpreted as costly signals of quality. However, there is another important function for these signals: to call the attention of the receiver to the signaller. While there is abundant empirical evidence to show the importance of this stage, it is not yet incorporated into standard signalling theory. Here I investigate a general model of signalling - based on a basic action-response game - that incorporates this searching stage. I show that giving attention-seeking displays and searching for them can be an ESS. This is a very general result and holds regardless whether only the high quality signallers or both high and low types give them. These signals need not be costly at the equilibrium and they need not be honest signals of any quality, as their function is not to signal quality but simply to call the attention of the potential receivers. These kind of displays are probably more common than their current weight in the literature would suggest.     

Linking signal fidelity and the efficiency costs of communication

The handicap principle has been the overarching framework to explain the evolution and maintenance of communication. Yet, it is becoming apparent that strategic costs of signalling are not the only mechanism maintaining signal honesty. Rather, the fidelity of detecting signals can itself be strongly selected. Specifically, we argue that the fidelity of many signals will be constrained by the investment in signal generation and reception by the signaller and perceiver, respectively. Here, we model how investments in signal fidelity influence the emergence and stability of communication using a simple theoretical framework. The predictions of the model indicate that high‐cost communication can be stable whereas low‐cost intermediates are generally selected against. This dichotomy suggests that the most parsimonious route to the evolution of communication is for initial investment in communicative traits to be driven by noncommunicative functions. Such cues can appeal to pre‐existing perceptual biases and thereby stimulate signal evolution. We predict that signal evolution will vary between systems in ways that can be linked to the economics of communication to the two parties involved.     

The evolution of strategic male mating effort in an information transfer framework

Sperm competition theory predicts that males should use cues indicating the risk and intensity of sperm competition to tailor their sperm investment accordingly. Rival males are an important source of social information regarding sperm competition risk. However, revealing such information may not be in the rival males' interest. Here, we use a theoretical approach based on informed and uninformed games to investigate when information transfer about sperm competition risk to competitors is beneficial for a male, and when it is not. The results show that signalling to potential future mates that a female has already mated is beneficial when the signalling male has a sperm competition disadvantage, whereas it is unfavourable when the signaller has an advantage. The reason for this counterintuitive result is that the rival males' optimal response is to reduce sperm investment when the signaller has a disadvantage and, conversely, to increase investment when the signaller has an advantage. Furthermore, we analysed scenarios where males use alternative reproductive tactics. In this situation, signalling the awareness of sperm competition risk rarely pays; instead, it is beneficial to maintain an information advantage. Thus, it may be beneficial for bourgeois males to accept cuckoldry instead of revealing their sperm competition awareness to reproductive parasites. These results provide new insight into the evolution of communication between rivals in the context of sperm competition.     

The Handicap Principle: how an erroneous hypothesis became a scientific principle

The most widely cited explanation for the evolution of reliable signals is Zahavi's so‐called Handicap Principle, which proposes that signals are honest because they are costly to produce. Here we provide a critical review of the Handicap Principle and its theoretical development. We explain why this idea is erroneous, and how it nevertheless became widely accepted as the leading explanation for honest signalling. In 1975, Zahavi proposed that elaborate secondary sexual characters impose ‘handicaps’ on male survival, not due to inadvertent signalling trade‐offs, but as a mechanism that functions to demonstrate males' genetic quality to potential mates. His handicap hypothesis received many criticisms, and in response, Zahavi clarified his hypothesis and explained that it assumes that signals are wasteful as well as costly, and that they evolve because wastefulness enforces honesty. He proposed that signals evolve under ‘signal selection’, a non‐Darwinian type of selection that favours waste rather than efficiency. He maintained that the handicap hypothesis provides a general principle to explain the evolution of all types of signalling systems, i.e. the Handicap Principle. In 1977, Zahavi proposed a second hypothesis for honest signalling, which received many different labels and interpretations, although it was assumed to be another example of handicap signalling. In 1990, Grafen published models that he claimed vindicated Zahavi's Handicap Principle. His conclusions were widely accepted and the Handicap Principle subsequently became the dominant paradigm for explaining the evolution of honest signalling in the biological and social sciences. Researchers have subsequently focused on testing predications of the Handicap Principle, such as measuring the absolute costs of honest signals (and using energetic and other proximate costs as proxies for fitness), but very few have attempted to test Grafen's models. We show that Grafen's models do not support the handicap hypothesis, although they do support Zahavi's second hypothesis, which proposes that males adjust their investment into the expression of their sexual signals according to their condition and ability to bear the costs (and risks to their survival). Rather than being wasteful over‐investments, honest signals evolve in this scenario because selection favours efficient and optimal investment into signal expression and minimizes signalling costs. This idea is very different from the handicap hypothesis, but it has been widely misinterpreted and equated to the Handicap Principle. Theoretical studies have since shown that signalling costs paid at the equilibrium are neither sufficient nor necessary to maintain signal honesty, and that honesty can evolve through differential benefits, as well as differential costs. There have been increasing criticisms of the Handicap Principle, but they have focused on the limitations of Grafen's model and overlooked the fact that it is not a handicap model. This model is better understood within a Darwinian framework of adaptive signalling trade‐offs, without the added burden and confusing logic of the Handicap Principle. There is no theoretical or empirical support for the Handicap Principle and the time is long overdue to usher this idea into an ‘honorable retirement’.     

Can behavioural constraints alter the stability of signalling equilibria?

In traditional models of signalling one class of individual, the signaller, presents a signal which another class of individual, the receiver, examines. Receivers are typically assumed to have fitness returns that depend on their ability to determine the utility of the signaller to them. Each signaller must decide what level to signal at, which is a function of the quality of the signaller. In addition, a signaller's quality is assumed to be synonymous with the signaller's utility to a receiver. However, there is no reason to believe that signalling costs are incurred in the same currency as the receivers are paid and, thus, no reason to believe that the relationship between signaller quality and utility is linear or even increasing. For instance, in signalling between prey and predators, the utility of a prey item may be its fat reserves, whereas an individual prey pays for signalling (and thus measures quality) in terms of increased risk of capture; quality and utility are synonymous only if a high risk of capture is associated with high fat reserves. In addition, several recent studies have documented increased signalling as utility decreases. If utility and quality are decoupled, so that increasing quality does not always mean increasing utility, then traditional signalling models predict that no signalling equilibrium will exist. I show that if receiver fitness is modelled by a set of behavioural responses, which have both costs and benefits, then a signalling equilibrium can sometimes be recovered. An example of signalling between mates is presented in order to demonstrate this equilibrium.     

Signal modulation as a mechanism for handicap disposal

Signal honesty may be compromised when heightened competition provides incentive for signal exaggeration. Some degree of honesty might be maintained by intrinsic handicap costs on signalling or through imposition of extrinsic costs, such as social punishment of low quality cheaters. Thus, theory predicts a delicate balance between signal enhancement and signal reliability that varies with degree of social competition, handicap cost, and social cost. We investigated whether male sexual signals of the electric fish Brachyhypopomus gauderio would become less reliable predictors of body length when competition provides incentives for males to boost electric signal amplitude. As expected, social competition under natural field conditions and in controlled lab experiments drove males to enhance their signals. However, signal enhancement improved the reliability of the information conveyed by the signal, as revealed in the tightening of the relationship between signal amplitude and body length. Signal augmentation in male B. gauderio was independent of body length, and thus appeared not to be curtailed through punishment of low quality (small) individuals. Rather, all individuals boosted their signals under high competition, but those whose signals were farthest from the predicted value under low competition boosted signal amplitude the most. By elimination, intrinsic handicap cost of signal production, rather than extrinsic social cost, appears to be the basis for the unexpected reinforcement of electric signal honesty under social competition. Signal modulation may provide its greatest advantage to the signaller as a mechanism for handicap disposal under low competition rather than as a mechanism for exaggeration of quality under high competition.     

Proportional processing of a visual mate choice signal in the green swordtail,Xiphophorus hellerii

During mate choice, receivers often assess the magnitude (duration, size, etc.) of signals that vary along a continuum and reflect variation in signaller quality. It is generally assumed that receivers assess this variation linearly, meaning each difference in signalling trait between signallers results in a commensurate change in receiver response. However, increasing evidence shows receivers can respond to signals non-linearly, for example through Weber's Law of proportional processing, where discrimination between stimuli is based on proportional, rather than absolute, differences in magnitude. We quantified mate preferences of female green swordtail fish, Xiphophorus hellerii, for pairs of males differing in body size. Preferences for larger males were better predicted by the proportional difference between males (proportional processing) than the absolute difference (linear processing). This demonstration of proportional processing of a visual signal implies that receiver perception may be an important mechanism selecting against the evolution of ever-larger signalling traits.     

The cost of dishonesty

The handicap principle states that stable biological signals must be honest and costly to produce. The cost of the signal should reflect the true quality of the signaller. Here, it is argued that honest signalling may be maintained although the used signals are not handicaps. A game theoretic model in the form of a game of signalling is presented: all the existing evolutionarily stable strategies (ESSs) are found. Honest and cheap signalling of male quality is shown to be evolutionarily stable if females divorce the mate if it turns out that he has cheated about his quality. However, for this ESS to apply, the cost of lost time must not be too great. The stability of the honest signalling is based on deceivers being prevented from spreading in the population because they suffer from a cost of divorce. Under some fairly strict conditions, a mixed polymorphism of dishonesty and honesty represents another possible ESS.     

Cost,competition and information in communication between relatives

Recent signalling models have shown that honest, cost-free communication between relatives can be stable. Moreover, cost-free signalling equilibria are in some cases more efficient than costly equilibria. However, we show that they are also relatively uninformative, particularly when relatedness between signaller and receiver is low. We explore the trade-off between signal cost and information, and further demonstrate that incorporating competition among signallers into a model of communication between relatives can reduce the propensity of any one signaller to display. As a result, there is a general increase in the amount of broadcast information in a non-costly signal with increasing competitor number.     

Investment in attending to cues and the evolution of amplifiers

Signals and cues are extensively used in social interactions across diverse communication systems. Here, we extend an existing theoretical framework to explore investment by emitters and perceivers in the fidelity with which cues and signals associated with the former are detected by the latter. Traits of the emitter that improve cue or signal fidelity without adding information are termed ‘amplifiers’. We assume that each party can invest in improving fidelity but that it is increasingly costly the more fidelity is improved. Our model predicts that evolution of amplifier traits of a pre‐existing cue occurs over a broader range of circumstances than evolution of signalling in situations where the emitter offered no pre‐existing cue to the perceiver. It further predicts that the greater the intrinsic informational value of a cue, the more likely it is that the perceiver (and not the emitter) will invest in the fidelity of detecting that cue. A consequence of this predicted asymmetry is that true communication with reciprocal adaptations in emitters and perceivers to improve signal fidelity is likely to occur predominantly for traits of intermediate reliability. The corollary is that uncertainty of the perceiver will then be a key feature of communication. Uncertainty can arise because perceivers misinterpret signals or do not perceive them correctly, but here we argue that uncertainty is more fundamentally at the root of communication because traits that are intrinsically highly informative will induce only the perceiver and not the emitter to invest in improved fidelity of perception of that trait.     

Cry-wolf signals emerging from coevolutionary feedbacks in a tritrophic system

For a communication system to be stable, senders should convey honest information. Providing dishonest information, however, can be advantageous to senders, which imposes a constraint on the evolution of communication systems. Beyond single populations and bitrophic systems, one may ask whether stable communication systems can evolve in multitrophic systems. Consider cross-species signalling where herbivore-induced plant volatiles (HIPVs) attract predators to reduce the damage from arthropod herbivores. Such plant signals may be honest and help predators to identify profitable prey/plant types via HIPV composition and to assess prey density via the amount of HIPVs. There could be selection for dishonest signals that attract predators for protection from possible future herbivory. Recently, we described a case in which plants release a fixed, high amount of HIPVs independent of herbivore load, adopting what we labelled a ‘cry-wolf’ strategy. To understand when such signals evolve, we model coevolutionary interactions between plants, herbivores and predators, and show that both ‘honest’ and ‘cry-wolf’ types can emerge, depending on the assumed plant–herbivore encounter rates and herbivore population density. It is suggested that the ‘cry-wolf’ strategy may have evolved to reduce the risk of heavy damage in the future. Our model suggests that eco-evolutionary feedback loops involving a third species may have important consequences for the stability of this outcome.     

Polymorphic signals of harassed female odonates and the males that learn them support a novel frequency-dependent model

For mate-searching species, the learned mate recognition (LMR) hypothesis assumes that sexual harassment favours signal variation among females, which exploits the receiver ability of males. The model predicts that coevolving males have responded to the female sexual foil by learning to recognize female variants as potential mates. I translate the LMR hypothesis into the language of signal detection theory to explain its novelty as a dynamic, coevolutionary, negative frequency-dependent selection model. Due to gene-environment interactions, males cueing to the morph detected most often should generate positive but often asymmetrical, detection-dependent harassment towards females. Females are expected to sort to an ideal free distribution where harassment costs are equal. At equilibrium, morph fitness, but not necessarily morph frequency, is predicted to be equal. The LMR hypothesis is consistent with recent experimental data and the distribution of colour polymorphisms in the Odonata, predicts general conditions favouring variation in sexual signals, and provides a novel mechanism for speciation via sexual signalling.     

Visual signalling by asymmetry: a review of perceptual processes.

Individual levels of asymmetry in traits that display fluctuating asymmetry could be used as visual signals of phenotypic (and perhaps genotypic) quality, as asymmetry can often be negatively related to fitness parameters. There are some data to support this hypothesis but the experimental protocols employed have commonly resulted in asymmetries far larger than those observed in nature. To date, there has been little consideration of the ability of animals to accurately discriminate small asymmetries (of the magnitude observed in the wild) from perfect symmetry. This is key to assessing the plausibility of the asymmetry-signalling hypothesis. Here, I review the perceptual processes that may lead to the discrimination of asymmetry and discuss a number of ecologically relevant factors that may influence asymmetry signalling. These include: signal orientation, distance of trait elements from the axis of symmetry, trait complexity, trait contrast and colour, and the behaviour of both signaller and receiver. I also discuss the evolution of symmetry preferences and make suggestions as to where researchers should focus attention to examine the generality of asymmetry-signalling theory. In highly developmentally stable signalling systems the magnitude of asymmetry may be too small to be detected accurately and reliably, hence asymmetry signalling is unlikely to have evolved in these situations.     

Signalling among relatives. I. Is costly signalling too costly?

Zahavi''s handicap principle,originally proposed as an explanation for sexual selection ofelaborate male traits, suggests that a sufficient cost to dishonest signals can outweigh the rewards of deception and allow individuals to communicate honestly. Maynard Smith (1991) and Johnstone and Grafen (1992) introduce the Sir Philip Sidney game in order to extend the handicap principle to interactions among related individuals, and to demonstrate that stable costly signalling systems can exist among relatives.In this paper we demonstrate that despite the benefits associated with honest information transfer, the costs incurred in a stable costly signalling system may leave all participants worse off than they would be in a system with no signalling at all. In both the discrete and continuous forms of the Sir Philip Sidney game, there exist conditions under which costly signalling among relatives, while stable, is so costly that it is disadvantageous compared with no signalling at all. We determine the factors which dictate signal cost and signal benefit in a generalized version of this game, and explain how signal cost can exceed signal value. Such results raise concerns about theevolutionary pathways which could have led to the existence of signalling equilibria in nature. The paper stresses the importance of comparing signalling equilibria with other possible strategies, beforedrawing conclusions regarding the optimality of signalling.     

Testing Thayer's hypothesis: can camouflage work by distraction?

One of the oldest theories of animal camouflage predicts that apparently conspicuous markings enhance concealment. Such 'distraction' marks are hypothesized to work by drawing the viewer's attention away from salient features, such as the body outline, that would otherwise reveal the animal. If distraction marks enhance concealment, then they offer a route for animals to combine camouflage markings with conspicuous signalling strategies, such as warning signals. However, the theory has never been tested and remains controversial. By using camouflaged artificial prey presented to wild avian predators, we test whether distractive markings enhance concealment. In contrast to predictions, we find that markings, both circular and irregular shapes, increase predation compared with unmarked targets. Markings became increasingly costly as their contrast against the prey increased. Our experiments failed to find any empirical support for the hypothesis that distraction markings are an important aspect of camouflage in animals.