Symposium Proposal: An Integrative Approach to Sexual Communication: Brain, Behavior and Evolution in Túngara Frogs

A recent trend in the field of animal behavior is an integrative approach that not only investigates central questions at different levels of analysis but uses information collected at one level of analysis to inform interpretations derived at other levels of analysis. In many ways, this is a return to Tinbergen’s four aims of ethology, which have proven an important heuristic in organizing our field. We are involved in such an integrative analysis of sexual communication in the túngara frog, Physalameus pustulosus, and its close relatives. The fundamental goal is to understand the evolution, adaptive function, and physiological mechanisms underlying the acoustic mate recognition system that mediates reproduction in these animals. Toward this end, we have assembled a group of scientists from areas of animal behavior, neurobiology, biomechanics, and phylogenetics to attempt a deep understanding as to how and why these animals communicate as they do. The goal of the symposium will be two-fold. (1) To review findings from these various studies to present a general understanding of this communication system from which we can draw generalizations about animal communication in other taxa and other sensory modalities. (2) To highlight the advantages of taking such an interdisciplinary approach to animal behavior, specifically, and organismal biology, in general. An important caveat is that not all of the talks will be “animal behavior” in the narrow sense, but all talks will be linked to the central question of sexual communication.

Introduction to the Symposium

Dr. Michael J. Ryan, Section of Integrative Biology, University of Texas

30 mins, Sexual communication: Some major problems in its evolution.

Studies of acoustic mate recognition in frogs, insects and birds have made important contributions to species recognition, sexual selection, and the evolution of communication systems. In this talk I outline a research approach that integrates concerns of behavioral ecology, neurobiology, and phylogenetics to address questions of why animals communicate the way they do. In respect to túngara frogs, I will give several examples of how questions of adaptive function and evolution that are addressed at only one level of investigation can be misleading, and how it is necessary to conduct truly integrative studies to obtain a deep understanding of communication systems. I then address a number of issues for which we know little in this system, including: the role of learning and early experience on acquisition of and preference for advertisement calls, the role of more immediate experience, such as nest building, on the reproductive success of males and females, and the meaning, if any, of patterns of geographic variation in calls and call preferences.

Acoustic Communication: Behavior

Dr. Stan Rand, Smithsoinian Tropical Research Institute, Balboa, Panama

30 mins, A natural history of the communication behavior of túngara frog.

The túngara frog has become a model organism for the studies of a variety of aspects of sexual communication.   The results of many of these studies will be presented in this symposium. In this talk I will try to convince you that the frog itself is interesting apart from its role as a model. The túngara is a small, plump, brown frog covered with a scattering of small warts like a tiny toad.  It is neither warm nor fuzzy and is attractive only to those that know it well.  Its attraction lies not in its appearance but rather in its voice.  Even here it is not how it sounds, because it's call is not musical, but in what it says and how it says it.  Females are silent but males are persistently vocal, commonly giving several thousand calls in a night.  Males gather in small choruses, usually in small temporary puddles in low elevation dry forest habitats from Mexico to Guyana.  Túngara frogs breed in construction sites, low spots in gravel parking lots and road ruts.  It is one of the few species that benefits from human modification of the environment.  Males are not territorial but a calling male does defend its immediate vicinity from other callers.   Females are attracted to male calls and select a mate on the basis of his call.  Males typically wait for the close approach of a female before attempting amplexus.  Non calling males are commonly seen in a chorus but seldom amplex females. After the male grasps a female she carries him to an oviposition spot usually at the edge of a small pond.  The females extrudes her eggs and large amounts of jelly that the male beats into a protective white foam nest with his hind legs.  The nest floats on the water and the tadpoles hatch into the water in about 48-72 hours.  Tadpoles grow to metamorphosis in about 4-6 weeks.  Metamorphs hop away into the litter, feed and grow to maturity in some months. The adults breed repeatedly throughout the wet season.  Males may call on successive nights and females can produce eggs at 4-6 week intervals.  Few survive to breed in the next wet season.

Ms. Kathy Boul, Section of Integrative Biology, University of Texas,

15 mins, Geographic variation in the presence of complex calls and preferences for complex calls in Physalaemus petersi.

There is geographic variation for the presence of complex calls in Phylaemaus petersi. Previously it had been thought that species in Ecuador lacked complex calls while those in Peru and western Brazil produced complex calls. I showed that one population in northern and western Ecuador (Yasuni) produces a complex call, but that other Ecuadorian populations, including one, La Selva, which is only 20 km from Yasuni, produce simple calls. An analysis of larynx variation throughout the species shows that populations without complex calls have a larynx morphology similar to Physalaemus coloradorum, which also does not produce complex calls. Populations that do produce complex calls have a larynx similar to P. pustulosus, which also produces complex calls. I conducted female phonotaxis experiments at La Selva and Yasuni. Yasuni, which produces complex calls, prefers complex calls to simple calls. La Selva does not show a preference for its basic call with a Yasuni secondary component added. Each population prefers its basic call component to that of the other population. I then suggest how these data influence our interpretations of signal-receiver evolution in the Physalaemus pustulosus species group.

Ms. Ximena Bernal, Section of Integrative Biology, University of Texas,

15 mins, Sexual differences in the perception of signal variation in túngara frogs.

Advertisement calls in frogs, as other sexual dimorphic traits, have a dual function; they are used not only to attract mates but also to deter rivals. Even though females and males respond to the same signal, they might respond differently to signal variation. Different responses among the sexes can be generated if the costs and benefits for discrimination error of the signals are different for males and females. On one hand, a male that makes a recognition error would lose time and energy displaying to the wrong target. On the other hand, a female that makes such a mistake could lost her eggs and, depending on the phenology of the species, a breeding season. Sexual differences in perception of signals willl affect the evolutionary outcome in the communication system. In most studies of sexual communication, however, the interaction between male trait and female preference is often investigated without reference to the male response to that signal. In this study I examined sexual differences in perception of acoustic signals in túngara frogs in relation to task differences. Responses to signal variation in the conspecific advertisement call, as well as responses to heterospecific calls of close relatives were evaluated in both sexes. I used female and male phonotaxis experiments to quantify signal recognition. In addition, I tested males in evoked calling behavior experiments for recognition of the same signals when performing a different task. Females and males showed similar responses in phonotaxis, both being responsive only to few of the stimuli presented. Males on evoked calling experiments, however, were less selective and called back to a broader range of stimuli. Thus the recognition system in males has two components that express different responses dependent on the task to perform. This study provides support for the hypothesis of sexual differences due to the costs associated with errors and shows within sex differences in signal discrimination associated to task differences.

Dr. Steve Phelps, Department of Zoology, University of Florida,

15 mins, A psychophysical analysis of recognition and discrimination: Are there two brain modules?

Dr. Ham Farris, Center for Neuroscience, Louisiana State University

15 mins, How females assign male signals to their source: auditory grouping of complex calls by túngara frogs.

In some modalities, assigning complex sexual signals to their correct source may be difficult. Unlike visual signals, in which adjacent regions on the retina may serve to group a male’s signals, acoustic signals are less directional due to factors such as reverberation. Most research on acoustic source determination has focused on humans and the phenomenon called the 'cocktail party effect', which refers to our ability to attend to sounds produced by one source while rejecting interference from others. Many other species, however, must also identify and locate complex acoustic signals imbedded in multiple overlapping signals and ambient noise. For example, male Túngara frogs (Physalaemus pustulosus) produce complex calls consisting of two distinct components, a ~350 ms FM sweep called the "whine" followed by up to three ~40 ms harmonic bursts called "chucks". In order to choose and locate a calling male, females attending to male choruses must identify and group call components (i.e., into auditory streams) so that they can be correctly assigned to their sources. In this study we used phonotaxis as a bioassay to measure how the spectrum, temporal structure and spatial orientation of the two components affect auditory grouping by females. Furthermore, we show that once grouped the separate call components are weighted differently in recognizing and locating the call, so called ‘what’ and ‘where’ decisions, respectively

Ms. Rachel Page, Section of Integrative Biology, University of Texas

15 mins, The costs of túngara calling behavior.

The evolution of sexual communication calls is often constrained by two factors: the pressure to be conspicuous and attractive to potential mates, and the attempt to be cryptic and avoid detection. Conspicuous calls attract not only mates but predators and parasitoids, both of which incur negative costs on the signaling prey. Male túngara frogs, Physalaemus pustulosus, produce advertisement calls that attract parasitoid flies (Corethrella, Corthrellidae) and predatory bats (Trachops cirrhosus, Phyllostomidae). The bats, as the female túngara frogs, prefer the frog’s complex call to its simple call.

I tested the hypothesis that Trachops prefer complex calls because they are easier to localize. I presented the bat with a speaker concealed beneath a screen covered in leaf litter. From the speaker, I broadcast simple or complex túngara frog calls, and measured three variables: (1) latency to flight (time from the onset of the frog call to the bat’s flight from its perch), (2) time of approach (time the bat spent circling the leaf litter screen before landing on the speaker), (3) distance to speaker (distance from the bat’s first landing on the screen to the speaker). I found that bats localize complex túngara frog calls faster and with higher accuracy than simple calls, with and without masking noise. When I introduced food rewards, however, I found similar localization performance for both call types. The influence of motivation on localization performance led me to test for learning in other situations. My experiments with T. cirrhosus suggest that detection of prey is bounded by biophysical constraints, but that detection within these boundaries is flexible.

Studies of frog-fly interactions are in their early stages but show that the flies, at minimum, prove a major irritant to the frogs, and are attracted to their advertisement calls. Future studies will be able to compare how these two different predators/parasites inflict different or similar selection costs on acoustic communication in túngara frogs.

Acoustic Communication: Physiology

Dr. Walt Wilczynski, Department of Psychology, University of Texas

30 mins, Sexual communication in túngara frogs: What we need to know about their brains.

This talk will review approaches to studying the neural systems underlying animal communication, focusing on the sensory system processing the communication signal. As for integrative, evolutionary examinations of behavioral processes, an analysis at the neural level involves studies at three levels. The first centers on the individual organism: what are the basic characteristics of the auditory system, and how do these characteristics relate to the acoustic characteristics of the advertisement call; that is, how does the brain encode the call? Neurophysiological studies indicate that the auditory system is divided into two channels, one originating in the amphibian papilla and one originating in the basilar papilla of the inner ear, which process the two components of the túngara frog advertisement call, the “whine” and “chuck” respectively. The second level expands the analysis to the question of individual variation in calls and tuning within a species in order to gain insight into sexual selection processes. Individual males vary in the dominant frequency of the “chuck”, with larger males producing lower frequencies. Female basilar papilla tuning varies, but on average is tuned below the mean chuck dominant frequency. Modeling studies indicate that this should make females more sensitive to lower frequency chucks, imparting a selective advantage to low frequency calls. The third level expands the analysis further to examine species differences in calls and tuning in order to test hypotheses about evolutionary patterns. A determination of amphibian and basilar papilla tuning in several species within the Physalaemus genus has shown that the basic tuning characteristics of both papillae are very stable within the genus, and a phylogenetic reconstruction of ancestral tuning indicates little change throughout the clade despite significant changes in the advertisement call. Furthermore, comparisons of calls and tuning across the extant species shows that all Physalaemus species examined have a “whine” that stimulates the amphibian papilla. Physalaemus pustulosus is unusual in adding a second component (the “chuck”) that stimulates the basilar papilla. The fact that all extant species have a basilar papilla with tuning similar to that of Physalaemus pustulosus suggests that this call evolved to exploit an additional, preexisting, part of the auditory system. By building the understanding of neural processing up from the individual level through multispecies comparisons, with explicit consideration of the results of behavioral studies and phylogenetic analyses, we have been able to obtain a deeper understanding of the mechanisms and the evolution of a model sexual communication system.

Drs. Kim Hoke and Sabrina Burmeister, Section of Integrative Biology, University of Texas,

15 mins, Functional mapping of the auditory midbrain in the túngara frog, Physalaemus pustulosus, using immediate early gene expression.

We mapped functional differences in the anuran auditory midbrain region, the torus semicircularis, using expression of immediate early gene egr-1 mRNA in

animals exposed to auditory stimuli. We cloned a fragment of egr-1 from Physalaemus pustulosus, the túngara frog, by degenerate PCR. Male túngara frogs were caught at mating ponds, held in soundproof boxes, and exposed to conspecific advertisement calls, heterospecific calls, or no auditory stimulus. Their brains were processed for radioactive in situ hybridization using an egr-1 probe, then regions of the torus semicircularis were analyzed separately to measure egr-1 expression levels. The principal nucleus of the torus semicircularis, which receives primarily auditory inputs, showed higher egr-1 expression than the laminar nucleus. The ventral toral region, a sparse cell layer where auditory input from the hindbrain terminates, also had higher egr-1 levels than the laminar nucleus, In addition to these differences between nuclei, there are regional differences in gene expression within nuclei. Egr-1 expression levels within the principal and laminar nuclei were higher more medially, and expression in the principal nucleus increased more caudally. Males exposed to conspecific whine-chuck calls showed trends towards higher egr-1 mRNA expression in the principal nucleus compared to males listening to heterospecific Physalaemus enesefae calls (p<0.1). The laminar nucleus showed less egr-1 expression in males listening to heterospecific calls than males listening to conspecific whine-chuck calls (p<0.05). These data suggest that the laminar and principal nuclei of the torus semicircularis are important in differentiating advertisement calls.

Dr. Cathy Marler, Department of Psychology, University of Wisconsin

15 mins, AVT and its effect of sexual communication in túngara frogs.

The neuropeptide arginine vasotocin and its mammalian homologue vasopressin have been shown to influence acoustic communication in a variety of species.   For example AVT can increase call rate and call duration in some anuran species.  Very little is known, however, about the function of these changes, particularly in a natural setting.  Tungara frogs provide an ideal species in which to examine both the effect of AVT on calling behavior and the function of such changes in the framework of sexual selection.  Preliminary data suggest that AVT changes several call characteristics, but most intriguing is an increase in the probability that chucks will be expressed in male tungara frogs.  Females display a strong preference for calls that include chucks.   We will discuss how the changes in AVT-induced calls can influence female preference.

Ms. Kathleen Lynch, Department of Psychology , University of Texas

15 mins, Female permissiveness in mate choice is influenced by physiological condition.

For convenience sake, mate choice studies frequently treat female choice as if it were static. Females, however, may display plasticity in their mate choices because of the cost associated with mate sampling. In fact, recent studies provide evidence that flexibility in female mate choice occurs over a range of time scales, such as within the females’ lifetime, within a breeding season or within a single breeding cycle. Mate choices of female túngara frogs (Physalaemus pustulosus) was examined throughout their breeding cycle to determine which aspects of mate choice behavior changed as females approached the time at which they must release their eggs. We investigated female plasticity in mate choice using phonotaxis tests that assay acoustic-based mating preferences. Tests were repeated in three reproductive stages: unamplexed, amplexed and post-mated. The likelihood of responding to a mate signal (receptivity), the probability of attending to a mate signal that is less attractive than the preferred signal (permissiveness) and the likelihood of discerning the difference between alternative signals (discrimination) were measured in each reproductive stage. We found that female receptivity and permissiveness in mate choice peaked in the amplexed stage, whereas discriminatory responses were maintained in this stage but not in the subsequent post-mated stage. Furthermore, females that deposit their eggs faster (requiring less than the median time of 14.5 hours) also respond faster to conspecific mate calls. Our results show that coordinated changes in female receptivity and permissiveness occur as females approach the point at which they must release their eggs but such behavioral plasticity is not a result of an overall loss of discrimination. In contrast, females that maintain receptive behaviors while in the post-mated stage do express diminished discriminatory responses. This study provides empirical support for theoretical models predicting that females will reduce their threshold criteria for choosing mates as critical time points approach.

Comparative Approaches

Dr. David Cannatella, Section of Integrative Biology, University of Texas

30 mins, Phylogenetic relationships in the genus Physalaemus as a foundation for studies in behavioral evolution.

The current behavior and function of an organism result from immediate selective pressures as well as the accumulated history of constraints and opporunities presented to its evolutionary lineage. Given this, comparative behavioral studies should benefit from the use phylogenetic relationships as a basis for design, execution, and interpretation of behavioral experiments. The Neotropical frogs of the genus Physalaemus present a stimulating system for such studies. Phylogenetic analysis of molecular sequence data for most of the species of Physalaemus and related genera indicates evolutionary novelty at several levels: (1) a fundamental increase in larynx size characterizes Physalaemus relative to other taxa; (2) within this constraint, further modification of larynx shape is correlated with evolutionary change in certain parameters of call pitch; and (3) interspecific signal variation does not appear to track long term evolutionary history.

Dr. Marcos Gridi-Papp, Department of Zoology, State University of Sao Paulo

15 mins, The evolution of biomechanics of complex call production in Physalaemus.

Species within the Physalaemus pustulosus group presents the derived condition, within its genus, of facultatively adding suffixes with a distinctive acoustic structure to their advertisement call. They also have a derived and functionally correlated morphology, such as a pair of laryngeal tissue masses that are large and have reduced attachments. The presence of acoustic suffixes have great importance for male mating success in the P. pustulosus group, and examining its functional link to their laryngeal morphology might help to explain interspecific differences in calling structure and behavior in the group. I report on a comparative biomechanical study, involving P. pustulosus and two other Physalaemus with different acoustics and laryngeal morphology. For each species, the potential acoustic output range of the larynx of euthanized frogs was determined through manipulation of airflow across the calling apparatus. Additional measurements on larynges with manipulated morphology were conducted to simulate muscular control and to test functional hypotheses. The basic acoustic function of the larynges of all three species is very similar. While the main part of the advertisement call of all three species can be produced simply by flowing air across the larynx, the suffixes of P. pustulosus require additional muscular contraction and depend on their specific laryngeal morphology. This indicates that a functional constraint in signal generation was removed in the evolution of the P. pustulosus, and allowed for differentiation of their advertisement signals and mating behavior.

Dr. Kathrin Lampert, University of Wuerzburg, Biozentrum, Am Hubland

15 mins, Microsatellite analysis of metapopulations and mate choice for genetic compatability in túngara frogs.

Túngara frogs (Physalaemus pustulosus) are a model system for sexual selection and communication. Population dynamics and gene-flow are of major interest in this species, because they influence speciation processes and microevolution and could consequently provide a deeper understanding of the evolutionary processes involved in animal communication. A molecular analysis (7 microsatellites) of fine scaled genetic patterns in the túngara frog revealed a high level of population differentiation at 2-3 km. A river created a major gene flow barrier, while habitat type was not important for genetic structuring. Males are the dispersing sex in túngara frogs, although the genetic data also suggest that female philopatry is not exclusive. One theory about sex-biased dispersal is that it may prevent inbreeding. We therefore became interested in the question whether female túngara frogs choose their mates for their relatedness assuming that they would prefer less related males as it has been demonstrated in other species (Waldman 1992). Microsatellite analysis of couples in amplexus did not reveal a female preference for less related males. Also, overall similarities in male calls or similarities in single call parameters as dominant chuck frequency or half-fall time that seem to be important in female choice did not correlate with the males’ relatedness. Closer related males do not produce more similar calls. The microsatellite analysis of metapopulations in the túngara frog revealed a high level of population structuring mainly influenced by isolation by distance. The most interesting finding was the sex-bias in dispersal. The male bias in dispersal lowers the probability of mating with a close relative. This could explain why female túngara frogs do not have to pay attention to relatedness when choosing a mate.

Dr. Heike Prohl, San Jose, Costa Rica

15 mins, Population genetic analysis of a secondary zone of contact in túngara frogs: Did a double invasion of Middle America yield two species of túngara frogs?

Earlier studies on allozymes and DNA-sequences on a transect throughout the range of P. pustulosus suggested that this species of South American origin twice invaded Middle America and this has resulted in a zone of secondary contact in western Panama and southern Costa Rica. I conducted an analysis of microsatellite variation of more than 20 populations in this area of presumed secondary contact (North Costa Rica to Western Panama). There is clear genetic structuring of populations along the transect as well as between populations north and south from a gap in the distribution in Central Costa Rica. Introgression with northern alleles offer support to the hypothesis of secondary contact between two túngara frog groups in southern Costa Rica. Also male calls vary between populations, and female mating preferences for the calls of their own populations suggest that these two groups might be in the early stages of evolving behavioral reproductive isolating mechanisms.

Mr. Santiago Ron, Section of Integrative Biology, University of Texas

15 mins, Is character displacement influencing the evolution of communication systems in Physalaemus?

Reproductive interactions between closely related species can generate selective pressures that promote divergence in communication systems. Recent fieldwork has shown that Western Ecuador is the most speciose region within the distribution range of the Physalaemus pustulosus group. At least six species occur in an area less than 55000 km². At some localities, two species can be calling syntopically. To test whether character displacement has promoted divergence in Physalaemus calls from Western Ecuador, I will define regions of sympatry between species by modeling their distributions with the GIS Genetic Algorithm of Rule Set Prediction. Call parameters will be compared between regions of predicted allopatry and sympatry of species pairs to assess the role of divergent selection on the evolution of Physalaemus calls. I describe two new species of Physalaemus from western Ecuador. I detail their relationships to the several other species of Physalaemus know from west of the Andes. I discuss patterns of sympatry among congeners in the is region and suggest future avenues of research critical to this area.

Concluding remarks

Dr. Peter Narins, Department of Physiology, UCLA

15 mins, Tungara frogs as a model system?: Comments from an outsider looking in.

This talk will offer a synopsis of the talks given in this symposium and give a critical evaluation of what contributions have been made by studies of this model system and what data are critically lacking.