Chapter 10 Summary

Evolution is a process of gradual change, as new adaptations become layered on top of preceding ones. Evolutionary biologists use morphological traits and DNA-based genetic traits to construct phylogenetic trees depicting clusters of closely related species. Robust trees can be employed to map the sequence of adaptations arising during speciation and taxonomic radiation. We can also assess the evolutionary progression of behavioral traits, including signal traits and receiver response traits, with these phylogenetic trees. But to understand the adaptive function of signals, we need to analyze the coevolutionary interactions and the fitness costs and benefits to senders and receivers. A signal cannot evolve and be maintained unless senders and receivers both benefit from the information provided.
Some new signals evolve from sender precursors, preexisting behavioral, physiological, or morphological traits that already provide informative cues to receivers. If the sender benefits from the decisions and responses made by receivers as a result of these cues, then the cue can be modified into a signal via the process of ritualization. Ritualization helps get the message across by making the cue easier to detect in a noisy environment, easier to discriminate from other behaviors, and, if learning is relevant, more memorable. If the ritualized signal becomes emancipated from the original cue association, it can become highly elaborated to provide new information to receivers, or it might be used to deceive receivers. Well-described sources of sender-derived signals include: locomotory and feeding appendages; intention movements; ambivalent, displacement, and redirected behaviors emerging from conditions of motivational conflict; visual and olfactory manifestations of autonomic nervous system functions; acoustic manifestations of the respiratory system; endocrine system products; and anti-predation defense tactics. New displays can also be co-opted from other existing displays. Signals derived from sender precursors provide information about the sender’s intentions, mood, and physiological state, and may also be ritualized into mate attraction signals.
Receiver preadaptations also play an important role in the evolution of signals. The ritualization process leads to highly efficacious signal designs given the receiver’s sensory capabilities and the signaling environment. The fine-tuning of signal features with respect to receiver sensitivity and environmental background is called sensory drive. A more specific source of new signals, however, is the exploitation by senders of preexisting sensory biases in the receiver that evolved to detect important environmental stimuli such as food or predators. A signal that mimics the features of these environmental stimuli, and may also exploit the associated receiver response, such as approach to food or flight from a predator, is called a sensory trap. If the receiver benefits from the sender’s matching trait, then the trait may become elaborated further. More likely, the receiver pays a cost, and a round of antagonistic sender/receiver coevolution may ensue. To escape the trap, the receiver must improve its ability to distinguish between the sender’s mimicking trait and the important natural stimulus, and evolve different response rules for each one. Signals derived from receiver precursors are usually mate attraction signals.
Honesty is the provision of accurate information by the sender, and deception is the provision of unreliable information. Honest communication occurs when both parties benefit on average from the signaling exchange. If sender and receiver have a conflict of interest because they rank the payoffs of alternative receiver responses differently, senders will be selected to exaggerate, bluff, lie, or withhold information. Receivers will counter by either improving their ability to detect such cheaters, or they will cease paying attention to the signal.
Evolutionary game theory is a powerful and essential tool for modeling the coevolutionary interactions, where the relative fitness for an individual playing a given strategy depends on the frequencies of strategies being played by others in the population. Strategies with higher fitness can invade, and those with lower fitness may be lost, leading to a balance in payoffs for one or more strategies called an evolutionary stable strategy. At this equilibrium point, no party can do better by shifting to another strategy. If a population is at a stable equilibrium, small deviations from this equilibrium lead to trajectories favoring a return to it, a property called convergence stability. Trajectories surrounding an unstable equilibrium move the population away from the point. Drift in small finite populations can greatly affect evolutionary trajectories. Game models can take many forms: alternative strategies may be discrete or continuous; interactions can be modeled in matrix-based normal form analysis as contest games with fixed payoffs, or scrambles with frequency-dependent payoffs; dynamic games with a sequence of successive actions require extensive form analysis; and games may be symmetric, where all players have access to the same strategy set, or asymmetric, with different strategy sets or payoffs for players with different roles.
Communication games are usually asymmetric, because senders and receivers have different roles and payoffs. Models that find conditions in which signaling and paying attention are an ESS generally include some payoff cost to senders that is greater when they cheat. Classifying signals on the basis of the type of cost that guarantees signal reliability provides important insights into the design of signals and the kinds of information that can be encoded in the signal. Handicap signals are those with an associated production cost, such as energy expenditure, predation risk, developmental investment, or vulnerability to attack. For the classic quality handicap signal, receiver selection pressure favors signal forms that impose the type of cost that reveals those aspects of the sender’s qualifications of importance to the receiver. Higher-quality individuals can bear the cost of the display better, resulting in high-quality senders delivering more intense or vigorous signals. A variant of the model shows that if senders vary in the benefit they obtain from some level of display, display vigor can reliably indicate sender need or motivation.
Index signals do not entail differential costs, but are reliable because their performance is functionally and incorruptibly linked to specific sender attributes. Quality indices are typically constrained by body size, strength, age, or fighting experience; senders of insufficient size, maturity, or strength simply cannot produce the high-quality signal. Informational constraint creates another type of index signal, in which signal production is linked to the sender’s knowledge.
Proximity signals are threat signals given in conflict contexts, where the cost that guarantees honesty is a close approach to the rival. Senders must approach close enough to risk attack and injury for the display to be effective. There are three subcategories: tactical threat signal, in which the sender assumes a posture that enables it to immediately launch an attack; vulnerability handicap signal, in which the sender exposes vulnerable body parts to the rival; and defensive threat signal, in which the sender attempts to protect its most vulnerable body regions and places weapons in a position to defend itself.
Conventional signals are neither constrained nor costly to produce, and may be arbitrary in form. Their meaning is established by an agreed-upon convention. If there is no conflict of interest between sender and receiver, because they rank the receiver response alternatives similarly, then cost-free signals can evolve. If sender and receiver have conflicting interests, then a socially imposed cost is required to maintain reliability. Receivers test the conviction or call the bluff of a sender giving a strong signal by approaching closely, and may retaliate or punish if a mismatch between the signal and the sender’s true state is detected.
Modifiers are traits that provide no information directly, but make it easier (amplifiers) or harder (attentuators) for receivers to assess certain aspects of a sender’s character. Modifiers can work in conjunction with any of the signal classes mentioned above.
Most signals are believed to be honest most of the time because of the cost controls, but dishonest signaling is expected to occur some of the time, and receivers tolerate a certain amount of deception as long as there is an average net gain by attending to the signal. Causes of unreliable signaling include: out-of-equilibrium systems; perceptual error by receivers; multiple classes of senders with different benefits and costs that receivers cannot distinguish; multiple classes of receivers; and tactical deception.