Sensory Exploitation Hypothesis
KeywordsMate Choice Female Mate Choice Advertisement Call Basilar Papilla Auditory Grouping
Mating preference for a courtship signal trait evolves as a preexisting sensory bias from a nonmating context and is then exploited by the opposite sex, increasing mating opportunities.
An important evolutionary driver of animal signal diversity is female mate choice. Throughout the animal kingdom, mating success between the sexes is asymmetric; most females in a population procure mates while some males mate multiply and many males fail to mate at all. This asymmetry results in differential selection on male traits and influences the evolution of male courtship signals (Andersson 1994). Virtually all mating systems include some component of communication whereby receivers must be able to detect and discriminate the sender’s courtship signal. This communication is critical for female mate choice, and in particular, females must be able to recognize members of their own species. Failure to perform this recognition often imposes severe fitness costs on the female, as a hybrid mating typically yields offspring with reduced vigor. Species recognition occurs through the assessment of specific characteristics of courtship signals. These signals can take on a myriad of different forms and may be communicated through virtually any sensory system (Ryan 1985; Basolo 1990; Partan and Marler 1999; ter Hofstede et al. 2015).
A large body of research has examined how the tuning of animal sensory systems generates signal recognition and influence mate choice (Capranica and Moffat 1983; Ryan 1990; Guilford and Dawkins 1991; Shaw 1995; Endler and Basolo 1998; Ryan and Cummings 2013). This work has connected signal evolution to receiver bias or more broadly, receiver psychology. Classically, sensory bias has been considered in three nonmutually exclusive contexts: sensory trap, sensory drive, and sensory exploitation (Endler and Basolo 1998). A sensory trap occurs when males produce a signal that elicits a positive mating response from a female because the signal mimics a stimulus from another context (Christy 1995). For example, in fiddler crabs (Uca spp.), females preferentially choose males that build pillars since the pillars mimic natural structures that allow females to escape predation. Sensory drive integrates evolutionary processes, environmental factors, and sensory systems that can lead to changes in signals and then behavior. For example, in two species of African cichlid, Pundamilia spp., female preference for red and blue nuptial coloring coincide with the spectral depth gradient in the natural habitat of the fish (Seehausen et al. 2008). This in turn may result in sexual selection on male coloration through female mate choice via sensory drive. Finally, Ryan (1990) proposed sensory exploitation to explain how male courtship signals evolve to exploit preexisting biases of the female’s sensory system. Mate choice via sensory exploitation predicts that male signals should evolve to increase stimulation of the female’s sensory systems and specifically should evolve to align with female sensory systems in regions that are most sensitive. Thus, females should be predisposed to prefer a particular courtship trait or character if it better stimulates her nervous system. Additionally, this preexisting bias must be present before the signal arises, otherwise the trait will fail to spread in the population. This review of the sensory exploitation hypothesis will focus on the sensory physiology, organismal behavior, and perception in three animal groups that have been extensively studied.
Animal Mating Systems via Sensory Exploitation
Often multiple species occupy a breeding pond, where males vigorously call and defend small territories against rival conspecifics. This vocal competition generates an acoustically complex environment. Females arrive at the pond and choose a mate by listening for males, distinguishing among different species, as well as individual calls of conspecific males.
Frogs possess two inner ear organs, the amphibian papilla (AP) and the basilar papilla (BP). These organs are tuned to relatively low sound frequencies (AP) or relatively higher frequencies (BP). Within a species, there is a strong correlation between the tuning of the inner ear organs and the frequency of the male courtship call; quite simply, if a male’s call falls within the region of best sensitivity of the female ear, he is most likely to be detected and hence attract a mate. This correlation between signal properties and neural sensitivity has been described as matched filter detection (Capranica and Moffat 1983).
The first note is called a “whine,” and males always produce this while calling. The second note, the “chuck,” can be voluntarily appended to the whine. Males almost never make a chuck in isolation and females do not respond to a chuck produced artificially in the absence of a whine. The whine is both necessary and sufficient for mate attraction. When males append one or more chucks to a whine, however, this makes the call five times more attractive than the whine alone. The dominant frequency of the whine matches the best sensitivity of the túngara AP, and the dominant frequency of the chuck closely matches the greatest sensitivity of the BP (Fig 2). Interestingly, a related species, Physalaemus coloradorum, also makes a “whine-type” advertisement call, but males do not make chucks. If a chuck is artificially appended to the P. coloradorum call, females prefer it, even though they have never been exposed to a chuck under natural conditions. Further, the BP in P. coloradorum exhibits similar tuning frequencies to the túngara frog, indicating that the artificial chuck is stimulating this additional inner ear organ. This illustrates that for túngara frogs, the chuck spread in the population when a mutation generated an early chuck-type note in the call, matching the spectral sensitivity (e.g., preexisting bias) of the BP.
In addition to frequency cues, female frogs have also been shown to use temporal aspects of call pulses for species recognition. This has been especially well studied in gray treefrogs. These consist of two species, Cope’s gray treefrog Hyla chrysoscelis and the gray treefrog, Hyla versicolor. H. versicolor is a tetraploid of the diploid H. chrysoscelis. The species are morphologically identical but produce distinct pulsed courtship calls. Males of H. chrysoscelis produce a courtship call typically consisting of around 50 pulses/s. Males of H. versicolor typically produce calls of about 20 pulses/s, although this can vary with temperature in both species (Gerhardt and Huber 2002). Both species use pulse rate as a mechanism for identifying conspecifics, with females showing strong mating preferences for conspecific pulse rates.
In more detailed neurobiology experiments, researchers have been able to examine the neural underpinnings of hearing perception in frogs (Wilczynski and Capranica 1984). For example, counting neurons have been identified in the auditory midbrain that responds to species-specific pulse rates in two species of frogs (Edwards et al. 2002). Recently, Schrode et al. (2014) used the minimally invasive auditory brainstem response (ABR) to develop a frequency tuning curve for Cope’s gray treefrog. They found two peak sensitivities, corresponding to the expected tuning of the two inner ear organs and also closely matching the frequencies of male’s advertisement calls (Schrode et al. 2014). In addition, this matched previously documented frequency preferences by females.
Orthopteran insects : Like frogs, Orthopteran insects (crickets and katydids) have provided a unique system to study behavior and signal evolution (Gerhardt and Huber 2002). In most species, males produce acoustic signals to attract mates by rubbing their fore wings together. The sound is produced by scraping one wing, which possess a series of teeth, against the anal margin of the other wing, which acts like a scraper. Generally, crickets communicate at frequencies around 5 kHz, and katydid calls are broadband signals with a frequency spectrum that may extend from the sonic to ultrasonic range (10–60 kHz). Both crickets and katydids possess tympanal ears for hearing on their legs. This sensory system, particularly in crickets, is tuned for the frequency of the male call. For example, in the cricket Gryllus bimaculatus, it has been shown that the tympanal sensitivity was best tuned to the carrier frequency of the male song, about 5.3 kHz, thus songs produced at higher or lower frequencies were not detected as easily. In addition to frequency, crickets and katydids produce chirps in specific temporal patterns, and detecting these species-specific pulse patterns is also an important part of recognition. In an elegant review, Hedwig (2016) showed that the species recognition system in crickets works as a set of serial filters, with each filter responding to specific features of the song such as frequency, chirp duration, and repetition rate.
Katydids are categorized by a high diversity of acoustic communication signals. In a comparison between partially sympatric sibling species, Neoconocephalus robustus and N. bivocatus, Deily and Schul (2006) found that they exhibit different call recognition mechanisms. N. robustus males produce calls with a single pulse rate and females of the same species are attracted to calls without amplitude modulation. In N. bivocatus females require amplitude modulation for call recognition. More broadly, in a phylogenetic comparative analysis of katydid call diversity, three important call properties were identified: call structure, pulse pattern, and pulse rate (Frederick and Schul 2016). This analysis compared 17 katydid species in the genus Neoconocephalus and revealed convergent evolution in call traits. Call diversity and the ability of a species to recognize conspecifics appeared to limit species diversity in particular locations. Two species with the same call trait do not co-occur. This coupled with changes in call structure and pulse rate produce clear acoustical niches, generating distinct species recognition.
Thus far, sensory exploitation has been described via neural tuning and female mate choice, the context in which it is normally considered. In the Hawaiian cricket, Teleogryllus oceanicus, a preexisting bias in males, however, has rendered a mutation adaptive. Like most crickets, males of this species produce songs to attract a mate. This song also attracts parasitoid flies (Zuk et al. 1993). A wing mutation, rendering some males silent, occurred in the population. Normally this mutation would have been eliminated from the population as silent males would fail to procure mates and pass on the trait. A preexisting behavior, the propensity for males act as satellites, rendered this trait adaptive, however, and allowed it to spread rapidly in the population. With satellite behavior, males sometimes approach another calling male, remain silent, and intercept females attracted to the caller. By exploiting the calling behavior of other males, the mutated silent males maintain their ability to procure a mate and also reduce their chance of attracting parasitoid flies (Tinghitella and Zuk 2009).
Another example of sensory exploitation in crickets is the conversion of a vibration startle response to a mate-pairing signal (ter Hofstede et al. 2015). Generally, female field crickets approach males with a low-frequency call and avoid high-frequency calls. In lebinthine crickets, the males produce an extremely high-frequency call and in response, the females produce a vibration instead of moving towards the male. The male then follows the vibrational signal to the female ter Hofstede et al. (2015) documented that field crickets closely related to the lebinthine crickets show an acoustic startle response to high-frequency sound and generate vibrations similar to those produced by female lebinthine crickets. The lebinthine vibrational signal thus likely arose from the preexisting startle response.
Fish : In contrast to crickets and frogs, courtship communication in fish primarily occurs via visual signals. Male Trinidadian guppies (Poecilia reticulata) display complex color patterns to females, and females prefer males with intense orange colors. Both male and female adult guppies are more responsive to orange-colored objects, even outside mating contexts (Rodd et al. 2002). The orange color is generated from carotenoid pigments in the diet and guppies are attracted to orange-colored food items. The cabrehash tree and several other trees produce bright orange fruit, but it is a rare, high quality food source for guppies. Thus, this preference for orange objects may have arisen in the context of food detection.
Another fish system, the genus Xipophorus, contains species where males bear a decorative extension of the caudal fin (swordtails) as well as unsworded species (platyfish). Female swordtails typically prefer to mate with males possessing longer swords. Interestingly, female platyfish, whose males lack swords, also prefer to mate with males who have had a sword artificially affixed to their tail. The sword is a derived character in this group, demonstrating that the female bias for the ornament predated the evolution of the sword (Basolo 1990). Recent data also support the hypothesis of sensory exploitation for the sword ornament in a group distantly related to Xipophorus. Mollies are composed of two clades, the long-fin clade and the short-fin clade, which differ in the presence of the sword phenotype. The tamesi molly, Poecilia latipunctata, was historically classified as a short-fin molly based on morphology and ecology but is grouped with long-fin mollies based on molecular and behavioral data. Tamesi molly females prefer the sword ornament, even though the males of this species to not express the trait (Makowicz et al. 2015).
African rift lake cichlids have undergone rapid evolutionary radiation, evolving hundreds of species within a relatively short time frame. This group exhibits a wide array of ecological and morphological variation, as well as variation in sexually selected courtship signals. These courtship signals are especially prominent as variable color patterns and have evolved through mate choice. For example, Egger et al. (2011) tested female preference for “egg spots,” the prominent light-colored spots that males of many species exhibit on their anal fins. They presented female haplochromine cichlids, Pseudocrenilabrus multicolor, with images of males where egg spots were artificially added; males of this basal species lack egg spots. Even though males of their own species lack them, females expressed a preexisting preference for these egg spots. The authors suggest that this preference may have arisen because the yellow-orange color of these spots was previously adaptive for locating carotenoid rich foods. Other work examining spectral tuning of retinal photoreceptors have provided clear evidence that different species have diverged in their sensitivities to particular wavelengths of light and these sensitivities have driven subsequent mate preference behaviors and color evolution (Seehausen et al. 2008).
Perception and Behavior
Information on sensory tuning, mate preference, and phylogenetic relationships have been important in elucidating the process of mate choice via sensory exploitation (reviewed in Ryan and Cummings 2013). More recently, some of this work has incorporated techniques from the human psychophysics literature, providing glimpses into the role of sensory perception in mate choice and signal evolution. Historically, understanding sensations and perception were considered to be outside the realm of science because these phenomenon are “private” to each individual. Researchers working in human psychophysics developed experimental techniques to test perception; specifically they measured stimuli and asked individuals if they could detect or discriminate, and hence perceive them (e.g., Fay 1988). The biology of some nonhuman organisms also lends itself well to these psychophysical methods and has allowed researchers to make important cross-taxonomic comparisons in auditory processing, perception, and mate choice. Here again, Orthopteran insects and Anuran amphibians have proved successful models (Wyttenbach and Farris 2004; Bee 2015).
Phelps et al. (2006), for example, presented female túngara frogs with a series of digitally manipulated calls. This series consisted of calls that were digitally “morphed,” from a normal túngara frog call into the call of another species that the female túngara frogs do not recognize. By presenting female túngara frogs with this series of calls, they were able to test the limits of acoustic recognition and discrimination. This allowed the researchers to then build a cognitive framework defining the acoustic recognition space in this species.
In another series of studies with the túngara frog, Farris and colleagues demonstrated a surprising auditory grouping behavior (Farris et al. 2002; Farris and Ryan 2011). Many animals use spatial cues to group segments of sound together and generate a coherent “auditory object,” that is, assigning a particular complex sound to its source location (Bregman 1990). In a series of clever experiments, Farris et al. (2002) explored the basis of auditory grouping in the túngara frog. As discussed previously in the chapter, females respond to the male’s whine call, but not an isolated chuck. In their experiments, Farris et al. (2002) presented females with a whine and chuck that were physically separated and females responded to the spatially separated chuck as if it were produced in the same location as the whine. Based on these data from a single whine and chuck, it appeared that female túngara frogs do not use spatial cues in auditory grouping. In a subsequent study, they presented females with a whine and two spatially separated chucks. In this case, females appeared to take advantage of relative comparisons, favoring chucks that were placed closer to whines in either space or time (Farris and Ryan 2011). This indicates that auditory grouping was not absent in túngara frogs. Instead, like humans communicating in noise, strategies for assigning sounds to their source by túngara frogs seem to be flexible.
In humans, a well-known phenomenon, the continuity illusion or auditory induction, occurs when brief silent gaps in sound are filled with white noise. The auditory system then fills in the masked gaps and generates the perception of an unfragmented “auditory object.” This process in humans likely improves speech comprehension in noisy environments where segments of speech become masked by noise (Bregman 1990). Baugh et al. (2016) tested auditory induction in the túngara frog and surprisingly found that they seem to lack this ability. Because túngara frogs communicate in noisy environments, analogous to noisy human speech conditions, it might be predicted that frogs would incorporate a similar mechanism for improving acoustic discrimination. Interestingly, Taylor and Ryan (2013) performed a similar type of test, but combined both acoustic and visual signals. In this study, they “sandwiched” a visual stimulus of a calling male frog between two temporally separate male call components, the whine and chuck. Normally, the temporally separated call components fail to elicit full recognition of the call. When the visual stimulus was placed in between the separated whine and chuck, however, this perceptually rescued the acoustic components and restored recognition, indicating the presence of a multisensory continuity illusion.
Since the early 1990s, tests of sensory exploitation as a mechanism for signal evolution via mate choice have required a multidisciplinary approach, employing methods from neurobiology, behavior, and phylogenetics (Ryan 1990; Shaw 1995). Further investigation of sensory exploitation still requires this integrative approach, since the evolution of the signal and receiver preference may not be closely coupled, the disconnect occurring as the preference predates the signal origin. Ryan and Cummings (2013) support a conceptual shift from the broad term sensory bias to perceptual bias to further expand our understanding of signal evolution via mate choice. The inclusion of theory from human psychophysics is an example of this shift, but there are also new approaches that will further the field. One is to gain fine-scale information about the focal communication system by utilizing genomic or transcriptomic tools. This is already being done by some behavioral researchers (Hofmann et al. 2014) and more are coming.
Another area under investigation is a focus on the female brain at the neuromolecular level, linking neural processes responsible for sensory processing to motor output leading to female mate preference behavior (Cummings 2015). Finally, an exciting area of research is beginning to focus on individual differences in sensory perception and mate choice. This area of research could focus on repeated measures of female preference, individual differences in central and peripheral signal processing, and/or individual genomic links to behavioral preference (Ronald et al. 2012; Patricelli et al. 2016). The previous several decades have provided profound insights into the role of sensory exploitation in mate choice and signal evolution. Current and developing technologies and concepts that allow researchers to link individual differences in genes, physiology, and behavior promise to offer an unprecedented understanding of this evolutionary process.
- Andersson, M. B. (1994). Sexual selection. Princeton: Princeton University Press.Google Scholar
- Bregman, A. S. (1990). Auditory scene analysis: The perceptual organization of sound. Cambridge, MA: MIT Press.Google Scholar
- Fay, R. R. (1988). Hearing in vertebrates: A psychophysics databook (p. 621). Winnetka: Hill-Fay Associates.Google Scholar
- Frederick, K., & Schul, J. (2016). Character state reconstruction of call diversity in the Neoconocephalus katydids reveals high levels of convergence. PLoS Currents 8.Google Scholar
- Gerhardt, H. C., & Huber, F. (2002). Acoustic communication in insects and anurans: Common problems and diverse solutions. Chicago: University of Chicago Press.Google Scholar
- Hedwig, B. G. (2016). Sequential filtering processes shape feature detection in crickets: A framework for song pattern recognition. Frontiers in Physiology, 46, 1–15.Google Scholar
- Hofmann, H. A., Beery, A. K., Blumstein, D. T., Couzin, I. D., Earley, R. L., Hayes, L. D., Hurd, P. L., Lacey, E. A., Phelps, S. M., Solomon, N. G., & Taborsky, M. (2014). An evolutionary framework for studying mechanisms of social behavior. Trends in Ecology & Evolution, 29, 581–589.CrossRefGoogle Scholar
- Ryan, M. J. (1985). The túngara frog: A study in sexual selection and communication. Chicago: University of Chicago Press.Google Scholar
- Ryan, M. J. (1990). Sexual selection, sensory systems and sensory exploitation. Oxford Surveys in Evolutionary Biology, 7, 157–195.Google Scholar
- Zuk, M., Simmons, L. W., & Cupp, L. (1993). Calling characteristics of parasitized and unparasitized populations of the field cricket Teleogryllus oceanicus. Behavioral Ecology and Sociobiology, 33, 339–343.Google Scholar