Relative Pitch and the Song of Black-Capped Chickadees

Chickadees, like people, have a strong sense of relative pitch. These birds use skillful, precise pitch changes to advertise their quality and attract mates

Ronald Weisman and Laurene Ratcliffe

ore than 2,000 years ago, the acer- descriptions of birdsongs. Donald Bor- Music and Relative Pitch Mbic philosopher Marcus Tullius ror, a master bioacoustician and field Relative pitch is the ability to produce Cicero observed that Roman songbirds biologist, found many of the song de- or recognize a note based on a refer- compose more excellent melodies than scriptions inadequate by modern stan- ence note. The ability to hum a simple any musician. He certainly doesn’t dards, as he wrote in his foreword to tune when given a set note with which stand alone in history on that count; it the 1967 reprinting of Mathews’s book. to start, as most people can, shows is a nearly universal human experience Borror acknowledged that Mathews relative pitch in action. A sense of to find joy and wonder in birdsong— lacked modern electronic equipment relative pitch is common among non- and to compare the songs to human and that his primary interest was in musicians even if they never use the music. People have been transcribing the musical content of birdsongs. To- term, but musical training develops melodies of birds into the notation of day, however, Mathews’s approach this knack. Trained singers have an ad- music since at least the 18th century; seems dated and quaint. Musical no- vanced appreciation of pitch change: Vivaldi’s Goldfinch concerto and Han- tation is simply unable to provide the They can produce precise relative fre- del’s Cuckoo and the Nightingale organ detailed, accurate and reproducible quency changes from one note to the concerto include musical notation for descriptions required in modern bio- next in a melody. birdsongs. In the early 1900s, the New acoustical analyses of vocal commu- To compare the appreciation for England naturalist and composer F. nication among songbirds. Mathews’s relative pitch among humans and Schuyler Mathews presented the songs approach helps musically trained peo- songbirds, it is necessary to under- of many North American birds in mu- ple recognize birdsongs, but it fails stand how musicians and bioacousti- sical notation in his Field Book of Wild to objectively describe birdsongs and cians measure pitch change. Musicians Birds and Their Music in order to help calls. For bioacousticians, accurate ob- describe the change between any two readers identify species common in the servations are a crucial first step in notes on the musical scale as an in- eastern United States. analyzing the role of songs in the life terval. The interval between any two Later biologists did not share of a species. adjacent notes is defined as a semi- Mathews’s enthusiasm for musical Modern ornithologists believe that tone. The ratio between the frequen- songbirds first appeared some 40 mil- cies (measured in cycles per second, lion to 50 million years before human or hertz [Hz]) of the higher and lower Ronald Weisman is a professor emeritus of psychology and biology at Queen’s University in beings. That people derive pleasure of two notes separated by a semitone Kingston, Ontario, where he was on the faculty for from birdsongs and recognize their is always 1.059, regardless of where 36 years. He received his Ph.D. with a concentra- musical features suggests that despite the two notes are located on the scale. tion in comparative psychology from Michigan the vast evolutionary gulf between Musicians speak of semitones and in- State University in 1964. He is the author of 90 birds and mammals, songbirds and tervals; bioacousticians use the terms articles and book chapters about animal cognition humans share some common auditory frequencies and frequency ratios. and evolution and continues to conduct research perceptual abilities. In this article, we Pianos and keyboards easily illustrate on songbird communication and the evolution focus on relative pitch, the ability to rec- the discrete intervals between notes. For of pitch perception. Laurene Ratcliffe is a profes- ognize relationships between acoustic example, hitting the E note in the fourth sor of biology, psychology and women’s studies frequencies. We review studies sug- octave (E4, 330 Hz), then counting at Queen’s University. She received her Ph.D. in biology, for research in animal behavior, from gesting that humans and songbirds down four keys (where both black and McGill University in Montréal in 1981. Address share the ability to perceive relative white keys are counted) and hitting the for Weisman: Department of Psychology, Queen’s pitch changes and exploring the evolu- C note in the same octave (C4, 262 Hz) University, Kingston, ON, K7K 3N6, Canada. tionary functions of birds’ perception produces a relative pitch change of four Internet: [email protected] and use of relative pitch. semitones, as measured by musicians, © 2004 Sigma Xi, The Scientific Research Society. Reproduction 532 American Scientist, Volume 92 with permission only. Contact [email protected]. Figure 1. Black-capped chickadees are noisy and highly social birds with two common vocalizations—their universal “chick-a-dee” call and the sweetly whistled “fee-bee” song. Males, like the one shown singing here from a post in his territory, sing the fee-bee song, primarily during the breeding season, when they use it to protect territory and to attract females. In this article, the authors show that the seemingly simple fee-bee song holds precise pitch cues that signal a male’s quality. (Photograph courtesy of Dan Mennill.) and a frequency ratio of 1.26, as mea- voice can vary in pitch continuously ent renditions of the same melody. In sured by bioacousticians. One can easily from one frequency to another, that is comparing relative pitch perceptions transpose a four-semitone pitch change not the way humans use their voices across species, we use low variabil- on a piano. For example, moving up or their musical instruments. People ity about mean frequency ratios in the the keyboard and hitting the E note in produce pitch changes with low vari- same song across singers to provide the fifth octave (E5, 659 Hz) and the C ability about their frequency ratios. evidence that songbirds, like people, note in the same octave (C5, 523 Hz) Humans recognize pitch changes have a strong sense of relative pitch. produces the same relative pitch change in discrete intervals and are good at as in the fourth octave: four semitones identifying notes relative to an exter- Black-Capped Chickadees or a frequency ratio of 1.26. Although nal reference pitch—in other words, Songbirds belong to the class Aves, they use different notations, musicians relative pitch. Most people easily rec- which includes all birds, the order and bioacousticians agree on the con- ognize a melody transposed to a differ- Passeriformes, consisting of perching stancy of pitch change across the keys. If ent key, or starting frequency, because birds, and the suborder Oscines, the one measured the frequencies produced the song maintains the same intervals, true songbirds. The true songbirds are by hitting E4 and then C4 on each of or frequency ratios, among the notes. distinguished from other perching birds 20 pianos, the average ratio should be Trained musicians, of course, are better by the anatomy of their vocal organ, the 1.26 with a very small standard devia- than others at identifying intervals and syrinx, and by the fact that they need tion among the pianos. The small stan- transposed melodies, but practically to learn their songs while juveniles dard deviation helps illustrate that pitch everyone has some sense of relative from adult members of the same spe- changes in music are discrete. pitch in simple melodies. For example, cies. Oscines are an evolutionary suc- Keyed instruments, such as pianos, people instantly recognize the song cess, comprising about half of the more are constrained to produce discrete “Happy Birthday” regardless of the than 9,000 species of birds and filling pitches, whereas keyless instruments, key in which it’s sung. niches in almost every environment on such as violins and human voices, can Human musicians have not artifi- the planet. Because songbirds are so produce pitches that vary continuous- cially adopted common scales to co- diverse in appearance and behavior, no ly. In practice, however, trained violin- ordinate their musical performances; single species can represent the whole ists and singers produce discrete pitch people, whether musicians or not, group. Our studies ask not whether all changes analogous to a piano’s keys. produce and perceive discrete, con- species of songbirds use relative pitch Put another way, although the human stant frequency ratios across differ- to produce and perceive songs but

© 2004 Sigma Xi, The Scientific Research Society. Reproduction www.americanscientist.org with permission only. Contact [email protected]. 2004 November–December 533 semitonessemitones intervals andintervals their frequency and their frequencyratios across ratios an acrossoctave an octave semitones interval name semitones interval name ratio ratio 0 1.000 0 unison unison 1.000 1 1.059 1 minor secondminor second 1.059 2 1.122 2 major secondmajor second 1.122 4 1.260 4 major thirdmajor third 1.260 C C2 8 1.498 C C2 8 fifth fifth 1.498 12 2.000 12 octave octave 2.000 2 2 1 2 2 2 2 2 1 2 2 1 2 1 Figure 2. Piano keyboard illustrates the concept of a semitone—the musical interval between any two adjacent notes. When a black piano key lies between two white keys, those white keys are separated by two semitone intervals. Twelve semitones cover an octave, a doubling in frequency. The table at right relates pitch intervals described by semitones to frequency ratios. Two strikes of the same key have a frequency ratio of 1.00, whereas hitting a key and then the seventh key over from it (eight semitones away) equals a frequency ratio of 1.498. Musicians describe pitch changes in terms of intervals and semitones; acousticians describe the same changes in terms of frequency ratios.

whether at least some Oscines do so field book of songbird music, Mathews Mathews observed that the male in order to identify the songs of their included black-capped chickadees. True chickadee’s fee-bee song consists of two own species. We have concentrated our to his description, these chickadees are or three clearly whistled notes. The first studies in particular on forest-dwell- acrobatic, noisy and highly social birds. note, “fee,” is sung at a higher pitch than ing songbirds such as the black-capped Their name derives from their call, the second one, “bee.” Sometimes, he chickadee, Poecile atricapillus (formerly which sounds like “chick-a-dee.” This noted, the bee note appears to be sung Parus atricapillus), because the songs of call is given by male and female, young as two separate notes at the same pitch. forest birds contain pure tones—that is, and old, to keep in touch with mem- Mathews suggested that chickadees tones of narrow frequencies—which are bers of the flock. The chick-a-dee call is separate the fee and bee notes in their relatively easy to measure and repro- complex, and some of the notes include songs by a clearly perceptible pitch in- duce in perception experiments, both in “buzzy” harmonics. During the breed- terval. Recall that musical intervals are the laboratory and in the field. ing season and less frequently through- simply precise discrete frequency ratios. Black-capped chickadees are familiar out the year, males sing the "fee-bee" About 70 years later, Steward Hulse of and abundant year-round in the north- song, used to hold territory and to at- The Johns Hopkins University restated ern two-thirds of the United States and tract and arouse females. We focus on Mathews’s hypothesis in modern bio- in much of Canada. In his 100-year-old the fee-bee song in this article. acoustic terms when he suggested that chickadees might hold the ratio between black-capped chickadee (Poecile atricapillus) breeding range the frequencies of their fee and bee notes constant across birds. To test Mathews’s and Hulse’s hypothesis, we, along with Ingrid Johns- rude, now at Queen’s University, and Andy Hurley, now at the University of Lethbridge, conducted a large bioacoustic analysis of several songs from each of 154 male black-capped chickadees singing on their territories in eastern Ontario. We found that, indeed, the internote ratio between the frequencies of an individual chickadee’s fee and bee notes varied by less than 2 percent about a mean value of 1.134 (just slightly more than two semitones). Also, the ratio between the fee and bee notes varied less than 2 percent across our sample of chickadees. Our spectrograms showed that the fee note is not constant in pitch but instead sweeps downward in frequency in what musicians call a glissando. The frequency ratio of the decline in the pitch from the start of the fee note until its end Figure 3. Black-capped chickadees live year-round in much of the United States and Canada. was 1.056 (just less than a semitone)

© 2004 Sigma Xi, The Scientific Research Society. Reproduction 534 American Scientist, Volume 92 with permission only. Contact [email protected]. with approximately 1 percent variabil- start to the end of the fee note and from For more than four decades, to dis- ity among the renditions either by an the end of the fee note to the start of the cover how songbirds communicate, individual bird or among birds. bee note, their pitch changes are very biologists have conducted experi- The spectrograms show that male precise, but their starting pitch varies. ments playing back songs. Recordings black-capped chickadees’ songs con- In musical terms, black-capped chick- can simulate the intrusion of a rival tain not one but two relative pitch cues: adees transpose the sequence of rela- male on or near another’s established the decline in frequency from the start tive pitch changes typical of the song territory. Skillfully recorded and re- to the end of the fee note—the glis- of their species across a wide range of produced songs of the species elicit sando ratio—and the decline from the pitches. This combination of musical forceful territorial defense, whereas end of the fee note to the beginning of abilities is rare in birds. Recordings of playing back songs with altered acous- the bee note—the internote ratio. Quite other songbirds—in particular, white- tic features may provoke a weaker unexpectedly, we found that individ- throated sparrows (Zonotrichia albicollis) territorial response, depending on the ual males that were brought into the and veeries (Catharus fuscescens)—show importance of the altered features to laboratory, isolated and recorded brief- that individual birds do not change the the potency of the song. ly occasionally sang their fee-bees on starting frequencies of their songs from We wondered whether accuracy in different starting pitches. Additional rendition to rendition. Constant note relative pitch is important to the po- field observations by Brad Hill, then a frequencies give other members of the tency of chickadee songs. In several graduate student, and Ross Lein, at the species a way to identify individual studies, we played back normal fee-bee University of Calgary, confirmed these birds. At the same time, a constant fre- songs and a variety of fee-bee songs findings. With our collaborators, Andy quency ratio in the song of the species with altered frequency ratios to male Horn and Marty Leonard, now both at helps identify the species. chickadees on their territories. One of Dalhousie University, and Scott Mac- us (Ratcliffe), along with Scott Mac- Dougall-Shackleton, now at the Uni- Song Perceptions Depend on Sex Dougall-Shackleton and Dan Weary, versity of Western Ontario, we mea- Humans may listen to birdsongs for now at the University of British Co- sured many more pitch-shifted songs solace and entertainment, but among lumbia, looked at the roles of both the from individually tagged males during songbirds, including black-capped glissando and the internote frequency their extended and persistent dawn chickadees, males listen to one another ratio in eliciting territorial defense. Ter- chorus. The chickadees surprised us. to identify potential rivals for territory, In the first dawn recording session, and females listen to find potential 6 fee-bee spectrogram each bird sang his fee-bee song at a mates who possess territories. During A wide range of frequencies, changing the breeding season, male chickadees B C frequency on average every 40 songs. defend territories of, on average, about 4 Over subsequent sessions, individual 5 hectares from rival males and attract males filled in the gaps between fre- females with whom they reproduce quencies within the range we heard. and rear young. If a rival male sings 2

These bioacoustical measurements in another’s territory, the male already frequency (kilohertz) taught us some astonishing facts about established there drives out the intrud- song production in black-capped chick- er by singing vigorously, approaching 0 adees. Individual chickadees do not use the trespasser and, if the rival persists, 0 0.5 1 1.5 discrete keys to place the starting fre- physically attacking him. time (seconds) quencies of their song but instead vary their songs continuously across the range of frequencies used by members 4.2 of the species. In other words, when internote ratio chickadees change frequencies from the glissando ratio Figure 4. Sound spectrogram (above right) represents the amplitude and frequency of 3.9 the vibrations making up the fee-bee song. The spectrogram shows that the tune consists of two loud, simple notes. The fee note has a distinct start frequency (A) and end frequency (B), whereas the bee note has one consis- 3.6 tent frequency (C). The downward frequency slide in the fee note is called a glissando, and its ratio is calculated as A/B; the frequency ratio of the downward internote pitch change higher frequency (kilohertz) is calculated as B/C. The larger graph shows 3.3 the glissando and internote ratios the authors observed in analyzing songs from 154 male black-capped chickadees. This bioacoustic analysis shows that chickadees produce pre- 3.0 cise, consistent relative pitch changes in the fee-bee song. The graph at right is redrawn 2.6 2.9 3.2 3.5 3.8 4.1 from Figure 2 in Weisman et al. 1990. lower frequency (kilohertz)

© 2004 Sigma Xi, The Scientific Research Society. Reproduction www.americanscientist.org with permission only. Contact [email protected]. 2004 November–December 535 Figure 5. Bird behavior at a feeding table displays the rank of males. A more dominant male, in flight and with a gaping beak, supplants a less dominant one, which turns away submissively. The authors observed this kind of behavior at winter feeders to rank males they studied. (Adapted from a photograph by Dan Mennill.) ritorial male chickadees in their native the male chickadees responded no less altered internote frequency ratios. In environment heard a normal song and to it than to the second altered song, one series of experiments, we digitally altered songs lacking the glissando ele- which had only an abnormal glissando copied and spliced together fee and bee ment. One altered song included a fee frequency ratio. notes from several chickadees to form note flattened to the start frequency of Our results puzzled us. We could not songs with normal internote ratios and a normal fee note, which increased the understand why male chickadees pro- songs with ratios as disparate from internote ratio relative to the bee note. duce accurate discrete pitch changes normal as 1.00 (equivalent notes) and A second altered song presented a fee in their songs only to ignore the vastly 1.29 (between four and five semitones note flattened to the end frequency of inaccurate internote pitch changes they apart). Judging by several behavioral a normal fee note, which therefore left heard in playback experiments. We measures, male chickadees persisted the fee-bee internote ratio unchanged were skeptical of this conclusion in part in defending their territories equally compared with a normal song. Male because other songbirds—in particu- strongly against all of these songs, chickadees flew closer to the speaker lar male white-throated sparrows and whether the internote ratio was normal on more occasions and called more of- veeries—are known to produce pre- or altered. Taken together, these field ten to a normal song than to the play- cise, predictable pitch changes in their experiments have led us, finally, to con- back of either altered song. The two songs, and playback experiments that clude that the glissando frequency ratio altered songs provoked about equal altered these ratios reduced territorial is far more important to territorial de- territorial defense. The first altered defense in both species. We continued fense by male black-capped chickadees song had abnormal glissando and ab- to seek reduced territorial responses in than is the internote ratio. normal internote frequency ratios, yet black-capped chickadees to songs with But what about female responses? Among most songbirds, chickadees included, females respond weakly or not at all to simulated territorial intru- sions. But female choice is a powerful evolutionary force, and we decided to address whether female chickadees care about the pitch changes sung by potential mates. One of us (Ratcliffe) and Ken Otter, now at the University of Northern British Columbia, brought female black-capped chickadees into breeding condition in the laboratory by artificially extending their day length and administering the hormone estradiol. When female songbirds are in breeding condition, the songs of males from their species elicit a dis- tinctive copulation-solicitation display. We played the receptive females two normal songs, a song with an in- Figure 6. Mated pair of chickadees is shown outside their nest cavity. Males use the fee-bee creased internote ratio and a song with song to attract female mates with whom they reproduce and rear young. (Adapted from a an altered glissando. Female chicka- photograph by Dan Mennill.) dees displayed longer to both normal

© 2004 Sigma Xi, The Scientific Research Society. Reproduction 536 American Scientist, Volume 92 with permission only. Contact [email protected]. male test songs 8 8 8 normal chickadee song altered glissando altered glissando and internote ratio

4 4 4

0 0 0

female test songs frequency (kilohertz) 8 8 8 normal chickadee songs altered glissando altered internote ratio

4 4 4

0 0 0 0 1 2 3 0 1 2 3 0 1 2 3 time (seconds) Figure 7. Sound spectrograms show test songs played in two sets of experiments. The male test songs were played in the wild to test the territorial response of males. The female test songs were played to sexually receptive females in the laboratory. The behavioral effect of songs with altered glissando frequency ratios (reduced to 1.00) and altered internote ratios (increased to 1.26) were compared with the effects of songs with normal glissando and internote ratios (1.056 and 1.134 respectively). Males showed a weaker territorial response to songs with an altered glissando ratio, but did not respond less to songs whose internote ratio was also altered. Females had weaker copulation displays to songs with either an altered glissando or internote ratio. songs than to the song with an altered cords from these birds as well as from who are themselves skilled singers. internote ratio, and they displayed isolated birds reared by Steve Nowicki This conclusion, in turn, led us to ask more often to the normal songs than at Duke University, along with his stu- whether experience with males and to the one with an altered glissando. dents Melissa Hughes, now at the Col- their songs is necessary for chickadees In other words, the females chickadees lege of Charleston, and Bernard Lohr, to develop accurate relative-pitch per- appeared to discriminate sexually in now at the University of Maryland. ception. Isolated males sing only aber- favor of normal songs over songs with Unlike chickadees reared normally, iso- rant notes, but, we wondered, is this either glissandos or internote ratios al- lated males sing quavering, low-volume simply an impairment in relative-pitch tered. The surprising implication of our notes and seldom produce consistent production, or do the birds lack accu- playback experiments is that although pitch changes between notes. rate relative-pitch perception? males need to produce a precise inter- Some songbirds can learn the song To address this question, one of us note pitch change to attract and stim- of their species from tape recordings, (Weisman) and Milan Njevovan, now ulate females, the males themselves but many species, including chicka- at the University of Alberta, housed seem not to care at all about whether dees, learn best when they interact both field-reared and isolation-reared their territorial rivals produce wildly with live adult males. Of more than a chickadees individually in sound-re- inaccurate internote ratios. dozen male black-capped chickadees sistant chambers and taught them to reared with taped tutors by one of fly to a feeder to obtain food. Once The Role of Experience us (Ratcliffe) with Scott MacDougall- a chickadee was standing on a perch Like human musicians, songbirds learn Shackleton, and by Don Kroodsma directly opposite the feeder, we played their songs; oscines do not produce pre- and his colleagues at the University a sequence of two pure-tone notes cise frequency ratios in their songs by of Massachusetts, only three birds from a speaker beside the feeder. In some feat of gene transcription. Most learned good approximations of taped this discrimination task, flying to the humans, of course, have some innate fee-bee songs, and only one bird shift- feeder after some two-note sequences musical ability and can sing and recog- ed the pitch in his songs. The remain- was rewarded with food, whereas fly- nize at least some melodies, and trained ing birds sang several whistled notes ing to the feeder after other two-note musicians clearly learn to perfect their and sometimes changed pitch dramat- sequences was not rewarded. relative-pitch skills. Such observations ically mid-note. The logic of the experiment was led us to consider whether black-capped From these studies of isolated and that chickadees should be able to dis- chickadees learn to produce the precise, tape-tutored males, we concluded that criminate rewarded note sequences discrete pitch changes in their fee-bee young male chickadees acquire their faster when they can categorize the songs. We have reared male chickadees precise control over the production of rewarded note sequences by a com- isolated from adults on several occasions relative pitches in their songs by listen- mon frequency ratio than when they and have examined spectrographic re- ing to and interacting with adult males need to memorize a random collection

© 2004 Sigma Xi, The Scientific Research Society. Reproduction www.americanscientist.org with permission only. Contact [email protected]. 2004 November–December 537 nore the constant internote ratio in rivals’ songs during territorial interac- tions and, second, that males reared in isolation have difficulty in perceiving, as well as in producing, the internote relative pitch change in fee-bee songs.

Evolutionary Benefits of Relative Pitch We wondered about the possible evolutionary benefits of male chickadees’ ability to shift pitch and female chickadees’ knack for identifying pitch changes. In what follows, it may seem that we have granted chickadees re- markable foresight and cognitive powers to choose ways of behaving that will maximize their reproductive success. In fact, we are only using a shorthand common among biologists to describe the way selection has fa- vored behaviors that result in greater reproductive success. Chickadees are only one of several Figure 8. Young males normally learn the fee-bee song from older males; black-capped chicka- songbirds that transpose their songs dees reared in isolation seldom produce consistent pitch changes between the low-volume, over a wide range of frequencies. For quavering notes they sing. The authors tested whether birds reared in isolation were impaired in their ability to perceive relative pitch. In their discrimination experiment, both field-reared species with a single type of song, and isolation-reared males were placed in cages where they were rewarded for flying to a transposing a song to match a rival’s feeder after certain two-note songs. (The speaker is covered in cloth here.) In one group, all pitch might be an aggressive act, simi- the rewarded songs had a common frequency ratio. In the other, the rewarded songs shared no lar to singing the same song type as a common ratio. (Adapted from a photograph by Chris Sturdy and Isabelle Charrier.) rival in a species with multiple song types. Over most of North America, of rewarded note sequences one by sions to acquire the discrimination as male black-capped chickadees sing a one. Not surprisingly, when we played field-reared birds that were in the con- single song type, the fee-bee song. One field-reared birds the note sequences stant-ratio group. Most important, iso- of us (Ratcliffe) in collaboration with with randomly rewarded frequency lation-reared birds needed four times Dan Mennill, now at the University ratios, they needed twice as many ses- as many sessions as field-reared birds of Windsor, found that in duels over to learn the constant-ratio discrimina- territory, male chickadees indeed use 5.4 tion. It appears that field-reared male pitch matching to escalate conflicts reward 5.1 no reward chickadees can perceive the constant with their rivals. Maintaining fixed 4.8 ratio in the rewarded note sequences, frequency ratios in song might be whereas isolation-reared males are one way chickadees provide consis- 4.5 1.26 4.2 impaired in relative-pitch perception tent cues to their identity across song and instead need to memorize the re- transpositions. But given that males do 3.9 1.12 warded note sequences one by one. not seem to discriminate particularly 3.6 1.00 Extrapolating from our results to the among songs with varying internote 3.3 perception of fee-bee songs, we can pitch intervals, we concluded that the 3.0 conclude, first, that free-living male adaptive benefits of pitch matching 2.7 constant ratio chickadees perceive but seem to ig- must lie elsewhere. 2.4 Figure 9. Frequencies are shown for two- We turned to the dawn chorus, 5.4 note sequences (similar to chickadee songs) that critical time of day when, among 5.1 played in the discrimination experiment. many songbirds, males advertise for In one group (constant ratio), the reward- mates, and females, even those with 4.8 ed notes (red) all have a frequency ratio of partners, prospect for mating oppor- 4.5 1.12. In the other group (pseudo ratio), the tunities. (Mating outside the pair bond

frequency of note one (kilohertz) 4.2 rewarded sequences varied in frequency is called extra-pair copulation.) Male 3.9 ratio. Without the consistent relative pitch chickadees regularly shift their songs 3.6 cue, field-reared birds took twice as long to during dawn singing, when overt ter- distinguish the rewarded sequences. Isola- 3.3 ritorial conflicts are rare. This suggests 3.0 tion-reared birds needed four times longer than field-reared birds to learn constant-ratio the motivation for frequent pitch trans- 2.7 pseudo ratio discrimination; clearly, the isolation-reared positions during dawn singing might 2.4 birds had far less ability to perceive relative be communication with females. 2.4 2.7 3.0 3.3 3.6 3.9 4.2 pitch. Graphs are adapted from Njegovan This led Ratcliffe and former gradu- frequency of note two (kilohertz) and Weisman 1997. ate students Peter Christie and Dan

© 2004 Sigma Xi, The Scientific Research Society. Reproduction 538 American Scientist, Volume 92 with permission only. Contact [email protected]. Mennill to study singing during the Evidently, the female social partners of Mennill, D. J., and L. M. Ratcliffe. 2004. Over- dawn chorus in more detail. We sus- the recorded males engaged in more lapping and matching in the song contests of black-capped chickadees. Animal Behav- pected that males might be shifting extra-pair copulations when they per- iour 67:441–450. pitch to convey information about their ceived a change for the worse in the Hughes, M., S. Nowicki and B. Lohr. 1998. Call quality to females, information females quality of their social mates. These learning in black-capped chickadees (Parus might value. Female chickadees pre- results suggest male chickadees have atricapillus): The role of experience in the fer high-ranking males as social mates, good reason to advertise their qual- development of “chick-a-dee” calls. Ethol- with rank referring to dominance po- ity by holding constant the internote ogy 104:232–249. sition in the previous winter’s flocks. frequency ratio in their songs—female Hurly, T. A., L. Ratcliffe and R. Weisman. 1990. Relative pitch recognition in white-throated If females are relegated to a lower chickadees are listening carefully. sparrows, Zonotrichia albicollis. Animal Be- ranking social mate, they choose high- haviour 40:176–181. ranking males as partners in extra-pair The Virtuosity of Chickadees Kroodsma, D. E., D. J. Albano, P. W. Houlihan copulations. Such matings account for When we began studying songbirds and J. A. Wells. 1995. Song development by the paternity of young in as many as almost twenty years ago, we chose black-capped chickadees (Parus atricapillus). a third of their clutches, as revealed by black-capped chickadees because they Auk 112:129–143. genotype analysis of the mother, the were abundant and sang simple songs. Mathews, F. S. 1967. Field Book of Wild Birds and young and the social mate. Subsequently, our work has revealed Their Music. New York: Dover Press. First published 1904 by GP Putnam’s Sons. Male song can divulge the devel- that chickadees’ seemingly simple Njegovan, M., and R. Weisman. 1997. Pitch opmental conditions, genetic quality songs present precise, discrete rela- discrimination in field- and isolation-reared and current vigor of the singer and is tive pitch cues. Males shift the starting black-capped chickadees (Parus atricapillus). known to affect females’ choices in so- pitch of their songs to match a rival’s Journal of Comparative Psychology 111:294–301. cial mates and copulation partners. We song as a threat and to display their Ratcliffe, L., and K. Otter. 1996. Sex differences measured the social rank of chickadees vocal virtuosity to females during the in song recognition. In Ecology and Evolution during aggressive encounters at winter dawn chorus. By preserving the rela- of Acoustic Communication in Birds, ed. D. E. Kroodsma and E. H. Miller. Ithaca, N. Y.: feeders where high-quality males pre- tive-pitch relations between notes dur- Cornell University Press, pp. 339–355. vail, then compared these dominance ing these shifts, males provide reliable Ratcliffe, L., and R. G. Weisman. 1985. Fre- rankings with several fee-bee song fea- clues about their identity and quality. quency shift in the fee bee song of the black- tures recorded during dawn singing. Music has been part of the human capped chickadee. Condor 87:555–556. The most important finding was that experience since our inception as a spe- Ratcliffe, L., and R. Weisman. 1992. Pitch across multiple pitch transpositions, cies. Like chickadees, young humans processing strategies in birds: a comparison high-ranking males maintain the inter- learn about relative pitch from adult of laboratory and field studies. In Playback note frequency ratio with much low- humans. Humans use precise, discrete and Studies of Animal Communication, ed. P. K. McGregor. New York: Plenum Press, pp. er variability than do lower-ranking pitch changes in music and transpose 211–223. males. In musical terms, when higher melodies while keeping the intervals Shackleton, S. A., and L. Ratcliffe. 1993. Develop- ranked males transpose their songs between the notes constant. As among ment of song in hand-reared black-capped upward in pitch during dawn singing, songbirds, humans tend to find musi- chickadees. Wilson Bulletin 105:637–644. they hold the interval—that is, keep cians sexually attractive. Under the cir- Shackleton, S. A., and L. Ratcliffe. 1994. Matched the internote ratio constant—whereas cumstances, it seems remarkably easy counter-singing signals escalation of aggres- lower ranking males sing sharp—that to relate to black-capped chickadees. sion in black-capped chickadees (Parus atricapillus). Ethology 94:310–316. is, they reduce the frequency ratio. It appears likely that at least several, Previously, we found that female if not most or all, oscine species can Shackleton, S. A., L. Ratcliffe and D. M. Weary. 1992. Relative frequency parameters and chickadees appreciate the quality of perceive relative pitch changes. As a song recognition in black-capped chicka- male songs in laboratory tests, but do resource, relative pitch may be ignored dees. Condor 94:782–785. female chickadees really listen to song by some songbirds and used by others, Weary, D. M., R. G. Weisman, R. E. Lemon, duels in the wild? Ratcliffe together but it is certainly exploited relentlessly T. Chin and J. Mongrain. 1991. Use of the with Peter Boag, also of Queen’s Uni- by black-capped chickadees. relative frequency of notes by veeries in versity, and Dan Mennill conducted a song recognition and production. Auk 108:977–981. striking test that showed females do Bibliography Weisman, R., L. Ratcliffe, I. Johnsrude and indeed eavesdrop on their mates’ en- T. A. Hurly. 1990. Absolute and relative counters with rivals. In 6-minute play- Christie, P. J., D. J. Mennill and L. M. Ratcliffe. pitch production in the song of the black- back tests, we played recordings of 2004 Pitch shifts and song structure indicate capped chickadee. Condor 92:118–124. male quality in the dawn chorus of black- high-ranking males having simulated capped chickadees. Behavioral Ecology and encounters with an intruder. The mock- Sociobiology 55:341–348. intruder either sang submissively, Hill, B. G., and M. R. Lein. 1987. Function of avoiding matching the resident’s pitch, frequency-shifted songs of black-capped or aggressively, pitch matching the chickadees. Condor 89:914–915. For relevant Web links, consult this resident. The results showed that high- Horn, A. G., M. L. Leonard, L. Ratcliffe, S. issue of American Scientist Online: ranking males that lost in the simulated A. Shackleton and R. G. Weisman. 1992. encounter with an aggressive intruder Frequency variation in the songs of black- http://www.americanscientist.org/ capped chickadees (Parus atricapillus). Auk IssueTOC/issue/661 also lost paternity. Extra-pair paternity 109:847–852. in their mates’ clutches rose from 10 Mennill, D. J., L. M. Ratcliffe and P. T. Boag. percent to 50 percent after submissive 2002. Female eavesdropping on male song and aggressive playback respectively. contests in songbirds. Science 5569:873.