Avian Behavior, Ecology, and Evolution

INTRODUCTION

For decades, researchers in North America and Europe have studied the function and mechanisms of elaborate advertisement traits in male animals (Darwin 1871, Andersson and Iwasa 1996, Tobias et al. 2011). Researchers traditionally assumed that these traits evolved as a result of sexual selection acting on males. Thus, similar elaborate traits in female animals were often ignored or assumed to be non-functional. This male trait bias has significantly influenced the study of bird song, resulting in a long-held assumption that bird song is primarily a male trait (Podos et al. 2004, Catchpole and Slater 2008, Haines et al. 2020, Rose et al. 2022).

Sexual dimorphism in song was thought to have evolved from an ancestral state in which neither sex had elaborate song. Thus, dimorphism would have evolved when male birds gained song, and females did not. However, through literature surveys and phylogenetic analysis, researchers have determined that female bird song is widespread across the songbird phylogeny and is likely the ancestral trait of oscine passerines (songbirds) (Odom et al. 2014, Odom and Benedict 2018). These findings challenge the traditional assumptions of the evolutionary history of bird song. Importantly, we now know that, in many cases, sexual dimorphism in song likely evolved as a result of a loss or decrease in female song (Price et al. 2009).

A growing number of studies have found support for several potential functions of female song in both oscines and suboscines. In Stripe-headed Sparrows (Aimophila r. ruficauda), females are more vocally aggressive than males in response to intrusions, especially to same-sex playbacks (Illes and Yunes-Jimenez 2009). Female song has also been observed to function for mate attraction: female Dusky Antbirds (Cercomacra tyrannia) sing different song types when they have a mate compared with when they are unmated, and it is hypothesized that the unmated “courtship” song type serves to attract a new mate (Morton et al. 2000). In duetting species, such as Rufous Horneros (Furnarius rufus) or Venezuelan Troupials (Icterus icterus), females participate in territory defense alongside their male partners through coordinated vocalizations and displays (Odom et al. 2016, Diniz et al. 2018). In Dusky Antbirds, both sexes overlap their mates’ songs during duetting, which could warn away potential suitors (Morton 1996, Morton and Derrickson 1996). Female song may also function to maintain pair bonds in duetting species, particularly in the tropics (Langmore 1998, Hall 2004).

In many species, however, female song is reduced compared with male song or absent entirely (Austin et al. 2021, Moyer et al. 2022, Price et al. 2023). Little research has been conducted on the consistent selective pressures that may lead to a reduction or loss of female song under particular ecological conditions. Female song may be costly in certain contexts, such as increasing predation risk for females singing from their nests (Kleindorfer et al. 2016, Stracey et al. 2023). Additionally, new evidence suggests that complex female song may be lost even if duetting behavior is maintained (Price et al. 2023). Price et al. (2009) found that loss or reduction of female song in the New World blackbird family Icteridae is associated with a change from tropical to temperate breeding. This loss of female elaboration in association with migratory behavior can also be seen with elaborate plumage patterns: in the oriole genus Icterus, phylogenetic reconstructions indicate that the loss of elaborate female plumage was associated with the evolution of migration (Friedman et al. 2009).

One member of the genus Icterus is the Orchard Oriole (Icterus spurius), a migratory songbird species that breeds across much of central and eastern North America (May–July). Male orioles arrive on the breeding grounds in late April–early May and establish territories, with females arriving later (Scharf and Kren 2022). A proportion of males, particularly yearling males, remain unmated for the duration of the season (Enstrom 1993, Scharf and Kren 2022). In addition to songs, Orchard Orioles produce several types of calls, including “chatter” calls and alarm “jeets.” Alarm jeet calls are typically produced in aggressive contexts and most often by males (Sturge et al. 2016, Scharf and Kren 2022). Orchard Oriole females have completely lost elaborate plumage (Hofmann et al. 2008) and can be easily visually distinguished from mature males in the field by plumage coloration. Yearling male plumage closely resembles that of females, with the exception of black feathers around their bill and throat (Scharf and Kren 2022).

Previously, our team analyzed the song rate and structure of both male and female Orchard Orioles during the breeding season, and we found that females sang significantly less frequently than males (Moyer et al. 2022). Furthermore, we found that female song was acoustically different from male song for five of the eight variables that we measured and showed significantly more variation than male song for all eight variables (Moyer et al. 2022). However, we noted that further studies were required to determine if there are functional differences in male and female oriole song, and if the birds can distinguish between male and female vocalizations.

Playback experiments are a common method for observing animal reactions to stimuli in their natural environment, particularly to simulated territorial intrusions (Kroodsma 1989, Illes and Yunes-Jimenez 2009, Cain et al. 2015). To determine if Orchard Orioles can distinguish between male and female songs, we conducted playbacks with male and female oriole song in the field and observed the orioles’ responses. The results of these experiments may also serve to help us assess potential functional hypotheses for female song in Orchard Orioles. For example, if male and female songs both elicit strong responses under similar conditions, male and female Orchard Oriole song may serve similar functions. However, if female song is nonfunctional in the traditional male contexts of territory defense or mate attraction/guarding, we might not expect either sex to respond strongly to female song playback.

Below we provide a summary of potential playback responses that would support specific functions of female song.

1. Female song functions in territory defense: if females primarily defend against rival females, females should approach the speaker more closely and signal more aggressively to female song playback than to male song playback. If females defend against any intruders, they should respond to the playback of both sexes. It is possible that males will also respond aggressively to female song if they consider other females a territorial threat, in which case males should also approach and produce aggressive vocal signals in response to female song.
2. Female song functions for mate attraction: males, particularly yearlings who are unmated, should be attracted to female song and respond positively by approaching the speaker or singing more frequently.
3. Female song functions for mate guarding: females, but not males, should respond more aggressively to female song than to male song if they are attempting to ward off a rival female or prevent their mate from interacting with other females. Females should also sing at higher rates in response to female song.
4. Female song is not functional or has reduced function in traditional male song contexts: neither sex should respond strongly to female song.

Distinct functions for female song do not necessarily have exclusive predictions. However, interpreting our results in light of the above predictions is a good first step toward determining potential functions for female song in Orchard Orioles.

METHODS

Simulated territorial intrusions

We conducted playback studies between 0600–1130 EDT at field sites across central Maryland (Howard, Prince George’s, Carroll, and Baltimore counties; Append. 1) in May and June of 2021 and 2022. We recorded observations during each playback using a Marantz PMD 661 recorder and a Sennheiser ME67 or ME66 shotgun microphone with K6 powering module. We played all song playback stimuli using an iPod Classic and an Anchor Audio AN-MINI speaker covered by a thin brown speaker cloth for camouflage. We set the speaker volume at a constant level to produce each signal at an amplitude of approximately 85 dB at one m, measured using a Radio Shack 33-2050 sound level meter with a C-weighting curve and a fast time response. Before each playback trial, the ambient noise level, temperature, wind speed, and all other metadata were recorded by both observers to ensure that playback volume would not be obstructed.

We placed a speaker on the ground below a tree ≥5 m in height near the center of an Orchard Oriole territory, and stationed two observers 20 m away. Observers were stationed roughly at 90 degrees from each other in areas that provided the clearest line of sight to the playback area. We placed flags 5, 10, 20, and 50 m away from the speaker along the N, S, E, and W axes to improve the accuracy of distance estimates. We determined that 50 m was the radius in which vocal behaviors could be reliably monitored and attributed to the playback stimuli. If the bird left the 50 m observation radius during the trial, we did not include any of their behaviors produced outside of that radius in the analysis, as the behaviors could not be reliably attributed to specific individuals.

We created playback treatments for three stimulus categories from previously collected high-quality recordings: male Orchard Oriole songs, female Orchard Oriole songs, and male Eastern Bluebird (Sialia sialis) songs as a heterospecific control. The geographic range of Eastern Bluebirds broadly overlaps that of the Orchard Oriole during the breeding season (Gowaty and Plissner 2020), but this species is not closely related enough to be considered a potential sexual competitor, so we did not expect any response to this treatment. For each of the three playback stimulus categories, we chose four individual stimuli, each one containing a single song from a unique individual so that we would have multiple exemplars for each stimulus class as a means to test for the effects of pseudoreplication in playback treatments (Kroodsma 1989). All songs were recorded during the breeding season (late April–June), digitized at 44.1 kHz, and high-pass filtered to remove background noise (Moyer et al. 2022).

Each Orchard Oriole tested received one playback treatment of each of the three types (male song, female song, bluebird song) in a given trial. Within a trial, each individual playback treatment lasted 2.5 min, with one recorded song repeated every 10 sec, corresponding to the singing rate in natural (i.e., in the absence of any stimuli) male Orchard Oriole song bouts. Between each of the playback treatments, we enforced a 10-min interstimulus period to reduce carryover effects from one treatment type to the next. Thus, a playback session with a given individual or pair resulted in approximately 47.5 min total trial time, including a 10-min prestimulus observation period. The playback treatment used for each trial was randomly selected without replacement from our recordings of four males, four females, and four bluebirds, with the stipulation that no song recorded from a given site was used in playbacks at that location. The bluebird treatment was always played first as a control, as we expected no response to this treatment, and the male and female oriole playbacks were alternated in order. Song playback stimulus treatments were created in Audacity® audio editing and recording software (V.3.0.2, Audacity Team 2021). All playback experiments were performed at least 1 km away from one another, or ≤1 km if the focal males were of different age classes and therefore visually distinguishable.

As noted above, a proportion of males remain unpaired throughout the breeding season, but no unpaired females were ever observed, so trials were performed on pairs and solo males. Before beginning a trial, we ensured that there was only one pair or individual male visible and that there were no neighbors that were likely to enter the observation area. If there was a known nest, the speaker was placed at least 20 m from the nest tree. On one occasion where the nest was located after the experiment had already commenced, and the speaker was less than 20 m from the nest, the bird’s distance from the speaker was only recorded when the bird was not on the nest.

During the course of our playback experiments, we modified two aspects of our original experimental design to streamline our data collection and increase our overall sample size. For the first ten playback trials, we began with 10 min of prestimulus observation, where any target behaviors were noted by both observers. After none of the responses that were typically given during playback experiments (speaker approach, alarm jeets, etc.) were observed, we ceased including this prestimulus assessment period for all subsequent trials. Similarly, after no such responses were observed during the bluebird control treatment in the first 38 playback trials, we conducted all subsequent trials with only male and female Orchard Oriole playbacks (Appends. 2, 3). In total, we conducted 59 playback trials (34 to both males and females, 25 to solo males).

Playback analysis

During each playback stimulus, we recorded the number of jeets and the number of songs produced by each responding oriole, as well as the closest approach to the speaker. Only one female produced an alarm jeet during any of the playback trials, so this response variable was scored only in males. One traditional approach for interpreting count data, such as the number of jeets, would be to model the data using a Poisson regression and Generalized Linear Mixed Models (GLMMs). However, there was a high proportion of zeros in the data. Appendix 4 shows the proportion of zeros expected for a typical Poisson distribution compared with the distribution observed in this data set.

Given the high proportion of zeros we observed compared with what would be expected from typical count data, we chose to analyze playback responses using zero-inflated “glmmTMB” package in R, v.4.2.1 (Brooks et al. 2017). We built five models to compare responses to each playback type: (1) Male Jeets ~ Playback Sex, (2) Male Songs ~ Playback Sex, (3) Female Songs ~ Playback Sex, (4) Male Closest Approach ~ Playback Sex, and (5) Female Closest Approach ~ Playback Sex. For all five models, we included the order of playback type provided as a fixed effect, and the date, the identity of the responding bird, and the pair status (of males, as all females observed were paired) as random effects. To obtain F statistics and p values for both random and fixed effects, we ran a one-way ANOVA on the model output. The Male Songs model failed to converge with all effects, so we determined the combination of effects that resulted in the lowest AICc and ran a simplified model, which did not include date or order.

Zero-inflation can obscure patterns in the data and make it difficult to detect underlying trends. Instead of reporting average values for each response, which would be lower than expected as a result of the high proportion of zeros, we converted the three behaviors of the focal male and the focal female (when present) to binary variables. For a given individual, each response behavior was scored as “yes” if they performed that behavior (sang at least one song, produced at least one alarm jeet, or came within 20 m of the speaker, respectively) during a playback trial. Individuals that did not perform that behavior were scored as a “no” response for that variable. Converting these responses to binary variables allowed us to more accurately assess overall response to playbacks, despite the large proportion of zeros in the data.

As an additional analysis to confirm our model results, we compared each of these binary response variables using Fisher’s Exact Tests. We also compared male and female responses to oriole playbacks vs. the bluebird control playback for the 38 trials that received the control stimulus. Finally, we quantified the song rates of responding birds during playbacks to compare with the natural rates reported in Moyer et al. (2022). All statistics were computed using R, v4.3.2 (R Core Team 2023).

Opportunistic observations of natural female song in context

We recorded opportunistic observations of females singing throughout the breeding season and noted the contexts in which they sang. Orchard Oriole pairs (n = 28) were recorded for 20-min observation sessions between 6:00 am and 12:00 pm at least once per week for 4 wks in May and June 2022. The location and behaviors of the female, her mate, and any additional orioles nearby were noted “before” (the 10 sec preceding a female song) and “during/after” (the duration of song the song and the period immediately after, totaling 10 sec) each female song. These behavioral observations were compared across all females to determine if there were any consistent contexts in which female song was most often produced and if there were any consistent responses observed from conspecifics in the “during/after” period. Natural female song rates (calculated as songs/min during the 20-min observation) during each stage of the breeding season (pre-building, building, incubation, and provisioning) were also quantified and compared to determine if song rate was associated with breeding stage.

RESULTS

Simulated territorial intrusions

Males produced significantly more alarm jeets in response to male playback than to female playback (F_1,57 = 5.10, p = 0.028; Fig. 1). There was significant individual variation in male jeets (F_56,57 = 1.65, p = 0.032). Both paired (n = 24) and unpaired (n = 23) males produced alarm jeets more often in response to male playback, with solo males producing significantly more alarm jeets overall (F_1,57 = 6.84, p = 0.011; Table 1).

There was no significant difference in male song responses to male vs. female playback (F_1,58 = 1.26, p = 0.267; Fig. 2), although there was significant individual variation (F_57,58 = 1.66, p = 0.028). Male song rates were low overall—only 50.8% of males sang during female playback, and only 37.3% of males sang during male playback (Fig. 2). Male song rates were lower than the natural rates observed in Moyer et al. (2022) (Table 2).

Females did not sing significantly more often to either type of playback (F_1,33 = 1.51, p = 0.228), but date had a significant effect, with females singing more earlier in the breeding season (F_1,58 = 13.43, p = 0.001; Fig. 2). In response to female playback, 9.1% of females sang, and 27.3% of females sang in response to male playback (Fig. 2). However, female song rates during both playbacks were similar to the natural rates observed in Moyer et al. (2022) (Table 3).

Males were significantly more likely to approach within 20 m to male playback (F_1,57 = 4.82, p = 0.322; Fig. 3). Females did not approach differentially to male playback vs. female playback (F_1,33 = 1.69, p = 0.203; Fig. 3). All model results were confirmed by the Fisher’s Exact Tests (Append. 5).

For the 38 trials that included bluebird control playback, males and females were significantly more likely to approach and males were significantly more likely to jeet in response to male playback than to the control playback (Appends. 2, 3). Males were significantly more likely to approach to female playback than to the control playback, but for all other response variables, the response to female playback was similar to the response to control playback (Appends. 2, 3)

Opportunistic observations of natural female song in context

Opportunistic observations of natural singing behavior did not indicate any specific contexts in which female song was consistently produced (Table 4). Females sang in a variety of contexts, including during aggressive interactions with other individuals or following songs from a mate, but there were no strong associations with specific behavioral interactions, conditions, or responses. Song rate comparisons across the breeding stage indicated that female song rates were highest during the pre-building stage of the nest cycle, although there was a great deal of individual variation (Fig. 4). Female song rates are naturally significantly lower than those of males, with females singing roughly 95% less frequently than males (Moyer et al. 2022).

DISCUSSION

By all three metrics of response, male and female Orchard Orioles responded to song playback from both sexes. Although the responses to male and female playback only differed significantly for two of the five variables, all five variables showed stronger responses to male playback than to female playback. Males produced significantly more alarm jeets and approached within 20 m of the speaker significantly more often when exposed to male song than to female song. “Male songs” was the only male variable that was not significantly different between male and female playback. The production of alarm jeets likely represents a more reliable male response to a territorial intrusion than songs, as male Orchard Oriole song rates often decreased in response to territorial intrusions and were lower than the natural rates previously observed (Moyer et al. 2022). Females did not sing or approach significantly differently to male vs. female playback. Responses from both sexes to female playback was low overall, and did not differ significantly from responses to the control playback for all variables except male approach.

Below we summarize the proposed hypotheses for female song and how well they are supported by the observed responses to playbacks:

1. Female song functions in territory defense: We found low support for this hypothesis. Although males did produce alarm jeets, which suggests that female song could be a threatening signal to territory owners, they did so significantly less often in response female playback than to male playback. In addition, females did not respond more strongly to female song than to male song, nor did they sing at higher rates than natural levels in response to territorial intrusion from either sex.
2. Female song functions for mate attraction: We found low support for this hypothesis. Males did approach to playbacks of female song. However, they also produced alarm jeets, which are generally an aggressive signal. Furthermore, unpaired males also produced more alarm jeets than paired males in response to female song, which contradicts the hypothesis that female song might be an attraction signal to birds seeking a mate. Females did sing more in response to male song, but this difference was not significant. However, females did sing more earlier in the breeding season according to the Female Song model and female rate observations, which could suggest that female song may be important for coordination of early breeding activities.
3. Female song functions for mate guarding: We found low support for this hypothesis. Females did not respond more aggressively to female song than to male song. Females also did not typically sing in response to female song, as would be predicted if females were guarding their mates.
4. Female song is not functional or has reduced function in traditional male song contexts: We found moderate support for this hypothesis. If female song were non-functional as a territorial or mate attraction signal, we would not expect either sex to respond to a simulated territorial intrusion playback experiment using female song. However, female song may be similar enough to male song in acoustic characteristics to elicit some response in that context, regardless of sex.

Naturalistic observations of female song did not provide evidence for any consistent function of female song, which suggests that this behavior may be under relaxed selection pressure. However, we have observed individual females using song in a number of clearly identifiable specific contexts. For example, two females sang continuously for over 2 min while engaged in a physically aggressive altercation (personal observation, MJM). On several occasions, we observed females singing at a rapid rate while two nearby males attacked one another (personal observation, KEO). Furthermore, several females have been observed singing on the nest when their mates were nearby (personal observation, MJM), a behavior that may or may not be adaptive (Leonard 2008, Kleindorfer et al. 2016, Stracey et al. 2023). Finally, we noticed that males usually sang in long bouts, often from exposed areas such as treetops or the edge of bushes, whereas females did not typically sing from any particular prominent locations (personal observations, MJM, KEO). It is possible, therefore, that female song may be functional for some individuals in specific contexts but is not under strong selection for a consistent function across all females. Relaxed selection pressure on female song is also supported by the increased variation that we documented previously in the acoustic structure and syllable usage of female song (Moyer et al. 2022; Moreland et al., unpublished manuscript).

Few studies have examined direct selective pressures against female song or reduced selection pressure in favor of retaining female song. Orchard Orioles are New World blackbirds (Icteridae), and Price et. al. (2009) demonstrated that across the Icterid phylogeny, loss of female song was associated with temperate breeding and the gain of migratory behavior. If the environmental conditions shift so that that the benefits of a given trait no longer to outweigh the costs, then evolutionary theory predicts that the trait will be selected against and will be lost or reduced (Lahti et al. 2009). Conversely, the trait could simply no longer be selected for, which is predicted to lead to increased variation in trait expression (Lahti et al. 2009, Reinhold 2011).

In migratory species, pair bonds and territories are more ephemeral, and female song may not be maintained in favor of other energetic requirements (Price 2009, Logue and Hall 2014). For Orchard Orioles, the natural history characteristics associated with temperate breeding, such as a shorter breeding season, lack of year-round territory defense, or the energetic constraints of migration could have led to a reduction in female song structure and functionality.

Importantly, elaborate female coloration, which may have been an important signal for territory, mate, or resource competition in an ancestral oriole, has been completely lost in this species (Hofmann et al. 2008). This is the only species of oriole that has no solid black and no pure patches of color in adult females. Similarly, Orchard Oriole female song is acoustically reduced (shorter, smaller bandwidth, less frequent) compared with male song, which may indicate an analogous reduction in selective pressure for competitive vocal signaling. In Carolina Wrens (Thryothorus ludovicianus), a species with long-term pair bonds and sedentary behavior, complex female song was still lost, which suggests that there may yet be additional unknown factors impacting the maintenance of female song (Price et al. 2023).

CONCLUSION

Further research in both tropical and temperate species is needed on the specific selection pressures that may be shaping female song in comparison with male song, particularly given the documented relationships between female song and tropical natural history characteristics (Price 2009). This study demonstrated that female song in Orchard Orioles may be functional in specific contexts, but playback experiments and naturalistic observations suggest relaxed selection on female song for functions traditionally associated with male song, such as territory defense and mate attraction. Whereas both males and females approached and sang in response to both playback stimuli, the overall responses to male song were stronger than to female song. Compared with male song, female song is reduced both in its production (Moyer et al. 2022) and in the degree of response it elicits (present study). Therefore, the retention of female song in specific contexts likely is not under strong positive selection in this species.

LITERATURE CITED

Andersson, M., and Y. Iwasa. 1996. Sexual selection. Trends in Ecology and Evolution 11:53-58. https://doi.org/10.1016/0169-5347(96)81042-1

Austin, V. I., A. H. Dalziell, N. E. Langmore, and J. A. Welbergen. 2021. Avian vocalisations: the female perspective. Biological Reviews 96:1484-1503. https://doi.org/10.1111/brv.12713

Brooks, M. E., K. Kristensen, K. J. Van Benthem, A. Magnusson, C. W. Berg, A. Nielsen, H. J. Skaug, M. Mächler, and B. M. Bolker. 2017. Modeling zero-inflated count data with glmmTMB. bioRxiv. https://doi.org/10.1101/132753

Cain, K. E., A. Cockburn, and N. E. Langmore. 2015. Female song rates in response to simulated intruder are positively related to reproductive success. Frontiers in Ecology and Evolution 3:1-7. https://doi.org/10.3389/fevo.2015.00119

Catchpole, C. K., and P. J. Slater. 2008. Bird song: biological themes and variations. Cambridge University Press, Cambridge, UK. https://doi.org/10.1017/CBO9780511754791

Darwin, C. 1871. The descent of man, and selection in relation to sex. John Murray, London, UK. https://doi.org/10.5962/bhl.title.121292

Diniz, P., E. F. da Silva, Jr., M. S. Webster, and R. H. Macedo. 2018. Duetting behavior in a Neotropical ovenbird: sexual and seasonal variation and adaptive signaling functions. Journal of Avian Biology 49(4): e01637. https://doi.org/10.1111/jav.01637

Enstrom, D. A. 1993. Female choice for age-specific plumage in the orchard oriole: implications for delayed plumage maturation. Animal Behaviour 45:435-442. https://doi.org/10.1006/anbe.1993.1055

Friedman, N. R., C. M. Hofmann, B. Kondo, and K. E. Omland. 2009. Correlated evolution of migration and sexual dichromatism in the New World Orioles (Icterus). Evolution 63:3269-3274. https://doi.org/10.1111/j.1558-5646.2009.00792.x

Gowaty, P. A., and J. H. Plissner. 2020. Eastern Bluebird (Sialia sialis), version 1.0. In A. F. Poole, editor. Birds of the world. Cornell Lab of Ornithology, Ithaca, New York, USA. https://doi.org/10.2173/bow.easblu.01

Haines, C. D., E. M. Rose, K. J. Odom, and K. E. Omland. 2020. The role of diversity in science: a case study of women advancing female birdsong research. Animal Behaviour 168:19-24. https://doi.org/10.1016/j.anbehav.2020.07.021

Hall, M. L. 2004. A review of hypotheses for the functions of avian duetting. Behavioral Ecology and Sociobiology 55:415-430. https://doi.org/10.1007/s00265-003-0741-x

Hofmann, C. M., T. W. Cronin, and K. E. Omland. 2008. Evolution of sexual dichromatism. 1. Convergent losses of elaborate female coloration in New World Orioles (Icterus spp.). The Auk 125:778-789. https://doi.org/10.1525/auk.2008.07112

Illes, A. E., and L. Yunes-Jimenez. 2009. A female songbird out-sings male conspecifics during simulated territorial intrusions. Proceedings of the Royal Society B: Biological Sciences 276:981-986. https://doi.org/10.1098/rspb.2008.1445

Kleindorfer, S., C. Evans, and K. Mahr. 2016. Female in-nest chatter song increases predation. Biology Letters 12:4-7. https://doi.org/10.1098/rsbl.2015.0513

Kroodsma, D. E. 1989. Suggested experimental designs for song playbacks. Animal Behaviour 37:600-609. https://doi.org/10.1016/0003-3472(89)90039-0

Lahti, D. C., N. A. Johnson, B. C. Ajie, S. P. Otto, A. P. Hendry, D. T. Blumstein, R. G. Coss, K. Donohue, and S. A. Foster. 2009. Relaxed selection in the wild. Trends in Ecology and Evolution 24:487-496. https://doi.org/10.1016/j.tree.2009.03.010

Langmore, N. E. 1998. Functions of duet and solo songs of female birds. Trends in Ecology and Evolution 13:136-140. https://doi.org/10.1016/S0169-5347(97)01241-X

Leonard, M. 2008. An overview and comparative analysis of singing on the nest in North American birds. Canadian Journal of Zoology 86:1101-1110. https://doi.org/10.1139/Z08-092

Logue, D. M., and M. L. Hall. 2014. Migration and the evolution of duetting in songbirds. Proceedings of the Royal Society B: Biological Sciences 281: 20140103. https://doi.org/10.1098/rspb.2014.0103

Morton, E. S. 1996. A comparison of vocal behavior among tropical and temperate passerine birds. Pages 258-268 in D. E. Kroodsma and E. H. Miller, editors. Ecology and evolution of acoustic communication in birds. Cornell University Press, Ithaca, New York, USA. https://doi.org/10.7591/9781501736957-021

Morton, E. S., and K. C. Derrickson. 1996. Song ranging by the Dusky Antbird, Cercomacra tyrannina: ranging without song learning. Behavioral Ecology and Sociobiology 39:195-201. https://doi.org/10.1007/s002650050281

Morton, E. S., K. C. Derrickson, and B. J. M. Stutchbury. 2000. Territory switching behavior in a sedentary tropical passerine, the Dusky Antbird (Cercomacra tyrannina). Behavioral Ecology 11(6):648-653. https://doi.org/10.1093/beheco/11.6.648

Moyer, M. J., E. M. Rose, D. A. Moreland, A. Raza, S. M. Brown, A. L. Scarselletta, B. Lohr, K.J . Odom, and K. E. Omland. 2022. Female song is structurally different from male song in Orchard Orioles, a temperate-breeding songbird with delayed plumage maturation. Journal of Field Ornithology 93(1): 3. https://doi.org/10.5751/JFO-00073-930103

Odom, K. J., and L. Benedict. 2018. A call to document female bird songs: applications for diverse fields. The Auk 135:314-325. https://doi.org/10.1642/AUK-17-183.1

Odom, K. J., M. L. Hall, K. Riebel, K. E. Omland, and N. E. Langmore. 2014. Female song is widespread and ancestral in songbirds. Nature Communications 5: 3379. https://doi.org/10.1038/ncomms4379

Odom, K. J., K. E. Omland, D. R. McCaffrey, M. K. Monroe, J. L. Christhilf, N. S. Roberts, and D. M. Logue. 2016. Typical males and unconventional females: songs and singing behaviors of a tropical, duetting Oriole in the breeding and non-breeding season. Frontiers in Ecology and Evolution 4. https://doi.org/10.3389/fevo.2016.00014

Podos, J., S. K. Huber, and B. Taft. 2004. Bird song: the interface of evolution and mechanism. Annual Review of Ecology, Evolution, and Systematics 35:55-87. https://doi.org/10.1146/annurev.ecolsys.35.021103.105719

Price, J. J. 2009. Evolution and life-history correlates of female song in the New World blackbirds. Behavioral Ecology 20:967-977. https://doi.org/10.1093/beheco/arp085

Price, J. J., S. M. Lanyon, and K. E. Omland. 2009. Losses of female song with changes from tropical to temperate breeding in the New World blackbirds. Proceedings of the Royal Society B: Biological Sciences 276:1971-1980. https://doi.org/10.1098/rspb.2008.1626

Price, J. J., M. T. Willson, and R. W. Pare. 2023. Loss of complex female song but not duetting in the ancestors of Carolina wrens. Ethology 129(1):47-54. https://doi.org/10.1111/eth.13344

R Core Team. 2023. R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria.

Reinhold, K. 2011. Variation in acoustic signalling traits exhibits footprints of sexual selection. Evolution 65:738-745. https://doi.org/10.1111/j.1558-5646.2010.01130.x

Rose, E. M., N. H. Prior, and G. F. Ball. 2022. The singing question: re-conceptualizing birdsong. Biological Reviews 97:326-342. https://doi.org/10.1111/brv.12800

Scharf, W. C., and J. Kren. 2022. Orchard Oriole (Icterus spurius), version 2.0. In A. F. Poole, editor. Birds of the World. Cornell Lab of Ornithology, Ithaca, New York, USA. https://doi.org/10.2173/bow.orcori.02

Stracey, C. M., K. Sanchez, B. Brown, D. Hawkins, and T. Shepherd. 2023. Singing on the nest is a widespread behavior in incubating Northern Mockingbirds and increases probability of nest predation. Ornithology 140(2):1-11. https://doi.org/10.1093/ornithology/ukad010

Sturge, R. J., K. E. Omland, J. J. Price, and B. Lohr. 2016. Divergence in calls but not songs in the orchard oriole complex: Icterus spurius and I. fuertesi. Journal of Avian Biology 47:109-120. https://doi.org/10.1111/jav.00595

Tobias, J. A., V. Gamarra-Toledo, D. García-Olaechea, P. C. Pulgarín, and N. Seddon. 2011. Year-round resource defence and the evolution of male and female song in suboscine birds: social armaments are mutual ornaments. Journal of Evolutionary Biology 24:2118-2138. https://doi.org/10.1111/j.1420-9101.2011.02345.x