Absolute pitch in boreal chickadees and humans: Exceptions that test a phylogenetic rule

Absolute pitch in boreal chickadees and humans: Exceptions that test a phylogenetic rule
of 18
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
  Learning and Motivation 41 (2010) 156–173 Contents lists available at ScienceDirect Learning and Motivation  journal homepage: Absolute pitch in boreal chickadees and humans:Exceptions that test a phylogenetic rule Ronald G. Weisman a , ∗ , Laura-Lee Balkwill a , Marisa Hoeschele b ,Michele K. Moscicki b , Laurie L. Bloomfield b , Christopher B. Sturdy b , ∗∗ a Queen’s University, Kingston, Ontario, Canada b University of Alberta, Edmonton, Alberta, Canada a r t i c l e i n f o  Article history: Received 1 June 2009Received in revised form 25 January 2010Available online 26 May 2010 Keywords: Pitch heightPitch chromaSongbirdsHumansMusicFrequency rangeBoreal chickadeesAbsolute pitchAnimals a b s t r a c t This research examined generality of the phylogenetic rule thatbirds discriminate frequency ranges more accurately than mam-mals. Human absolute pitch chroma possessors accurately trackedtransitions between frequency ranges. Independent tests showedthat they used note naming (pitch chroma) to remap the tonesinto ranges; neither possessors nor nonpossessors were accurateat octave (pitch height) naming. Boreal chickadees discriminatedfrequency ranges less accurately than other birds; they trackedreward across several lower frequency ranges but failed at fre-quencies over 4000Hz. The results revealed the error of describingspecies differences solely in terms of their discrimination of fre-quency ranges. Exceptions to the rule disappear when the rule isrestated in terms of underlying mechanism: birds are superior tomammals in the ability to use absolute pitch height perception todiscriminate pitches and ranges of pitches.© 2010 Elsevier Inc. All rights reserved. Oscines, the true songbirds, learn to use song as an acoustic signal to defend territory and attractfemales for reproduction. It is now well known that songbirds collect information from several songfeatures (e.g., number of notes, note duration, and trill notes) to identify conspecifics (Nelson, 1988).However,onefeature,thespectralfrequencyrangesofsongnotes,iscitedrepeatedlyintheliterature ∗ Corresponding author at: Department of Psychology, Queen’s University, 62 Arch Street, Kingston, Ontario, Canada K7L 3N6. ∗∗ Corresponding author at: Department of Psychology, P217 Biological Sciences Building, University of Alberta, Edmonton,Alberta, Canada T6G 2E9. E-mail addresses: (R.G. Weisman), (C.B. Sturdy). 0023-9690/$ – see front matter © 2010 Elsevier Inc. All rights reserved.doi:10.1016/j.lmot.2010.04.002  R.G. Weisman et al. / Learning and Motivation 41 (2010) 156–173  157  Table 1 Frequencies(Hz)ofS+(responsesreinforced)andS − (responsesunreinforced)tonesintheS − firstandS+firstfrequency-rangediscrimination groups.Frequency range S − first group S+ first group Frequency (Hz)1 S −  S+ 980 1100 1220 1340 14602 S+ S −  1580 1700 1820 1940 20603 S −  S+ 2180 2300 2420 2540 26604 S+ S −  2780 2900 3020 3140 32605 S −  S+ 3380 3500 3620 3740 38606 S+ S −  3980 4100 4220 4340 44607 S −  S+ 4580 4700 4820 4940 50608 S+ S −  5180 5300 5420 5540 5660 asimportanttobothindividualandspeciesrecognition.Forexample,infieldsparrows( Spizellapusilla ),songsconsistingofnoteswithinthenormalfrequencyrangeelicitmoreterritorialdefensethansongspitch shifted two standard deviations upward (Nelson, 1989a). Moreover, Nelson (1989b) was able to show that sparrows could simultaneously use frequency to identify an individual neighbor and toidentify the range of conspecific songs. Evidence supports the hypothesis that frequency range is asuperordinatefeatureinsongrecognition.Thatis,infieldexperiments,frequencyrangepredominatesover other acoustic features (e.g., the timing of notes) in determining the strength of the territorialresponse (see Nelson, 1989b; Weary, Lemon, & Date, 1986). The finding that the frequency range of songnotesisimportanttoconspecificrecognitionhasbeenreplicatedinatleast12oscinespecies(e.g.,Dabelsteen & Pedersen, 1985; Emlen, 1972; Falls, 1963; Lohr, Weisman, & Nowicki, 1994; Nelson,1989a; Nowicki, Mitani, Nelson, & Marler, 1989; Thompson, 1969; Weary et al., 1986; Weisman &Ratcliffe, 1989; Wunderle, 1979).Thus far here, the examples of frequency-range discrimination have come from nature. However,over the past decade our research group has developed a suite of sensitive and direct laboratory testsknown as operant frequency-range discriminations. These discriminations are laboratory analogsto identifying the frequency ranges of successive notes in conspecific songs and calls. In a studythat provided a model for the present experiments, Weisman et al. (1998) trained songbirds inoperant discriminations of 40 tones, spaced 120Hz apart, in the spectral region between 980 and5660Hz, and parsed into eight ranges of five tones each (see Table 1). In the S+ first version of the task, responses (hopping or flying to the feeder) to tones in the odd ranges (1st, 3rd, 5th, and7th) were rewarded (with food) and responses to tones in the even ranges were not rewarded.In the counterbalanced, S −  first, version of the task, responses to tones in the even ranges (2nd,4th, 6th, and 8th) were rewarded. In both versions of the task, zebra finches ( Taeniopygia gut-tata ) accurately tracked every shift between reward and nonreward across the eight frequencyranges.We have studied both less difficult, three-range, and the more difficult, eight-range (Weisman etal., 1998), versions of the frequency-range discrimination in several avian and mammalian species.Amongbirds,fouroscinespecies(zebrafinches,white-throatedsparrows  Zonotrichia albicollis ,black-capped chickadees  Poecile atricapillus , and mountain chickadees  Poecile gambeli ), a parrot species(budgerigars, P.gambeli )andadovespecies(pigeons, Columbalivia )alllearnedtotrackrewardacrossthree and even eight ranges (see Weisman, Njegovan, Williams, Cohen, & Sturdy, 2004; Weisman,Williams, Cohen, Njegovan, & Sturdy, 2006 for summaries). Two mammalian species (humans,  Homosapiens , and rats,  Rattus norvegicus ) learned to track reward across three ranges only slightly lessaccurately but tracked reward across eight ranges much less accurately than birds (Weisman etal., 2004). Also, rats attend more to frequency change than frequency range to categorize complexsounds (Mercado, Ordu˜na, & Nowak, 2005). We have proposed an explanation from phylogeny forthe superiority of avian over mammalian species at tracking frequency ranges. Phylogeny, that is,the genetic heritage of species, is one of  Tinbergen’s (1965), often-cited, four causal explanations of behavior; the other three are adaptation, development, and mechanism. All four causes operate toprovide a full explanation of behavior but each can be investigated separately. Explanations fromphylogeny are not as mysterious as might first appear. For example, the evolution of color vision  158  R.G. Weisman et al. / Learning and Motivation 41 (2010) 156–173 early in their history accounts for birds’ common use of distinctively colored patches of feathers tomark species identity. In contrast, mammals, except for primates, have at best weak color visionand do not exhibit distinctive color markings. That primates have color vision and use color to markconspecifics illustrates that exceptions can signal important revisions to a phylogenetic explana-tion.Describing auditory discriminations in terms of ranges of frequencies is a kind of shorthand fora more scientifically sophisticated description in terms of pitch perception. Spectral frequenciesmay be what birds produce when they sing, but pitches are what birds perceive when they hearsong.In the early 1980s, Stewart Hulse began conducting auditory operant discriminations in thelaboratory to study absolute and relative pitch in songbirds. Absolute pitch (AP), now studiedin frequency-range discriminations in birds, is the ability to produce and perceive pitches with-out an external referent. Relative pitch is the ability to produce and perceive pitches based onordinal or ratio relationships between temporally adjacent notes. Hulse studied operant discrimi-nations between ascending and descending sequences of tones and transfer to novel sequences (e.g.,Hulse & Cynx, 1985; MacDougall-Shackleton & Hulse, 1996; Page, Hulse, & Cynx, 1989). For exam-ple, MacDougall-Shackleton and Hulse (1996) found that starlings learn to discriminate betweenascending and descending sequences and generalize to higher and lower frequency sequenceson the basis of either high versus low (absolute) pitch or ascending versus descending (relative)pitch.Mainly because of Hulse’s research, it is now commonly understood that birds’ responsesto conspecifics’ vocalizations in nature and to learned frequency-range discriminations in thelaboratory are the products of avian AP. Seen in this way, our research on frequency-range dis-criminations is a natural progression from Hulse’s work towards quantifying absolute pitch acrossspecies.AP has been studied in many species, but definitional and methodological issues have hamperedefforts to make direct comparisons, especially between humans and other animals. Many psycholo-gists that study AP require humans to name pitches using the notation of modern Western music(e.g., Takeuchi & Hulse, 1993). How can birds, or any nonhuman animal for that matter, possessaccurate AP if they cannot name pitches as required by music psychologists (see Levitin & Rogers,2005)?An important piece of the puzzle is that AP has at least two dimensions: pitch  chroma  and pitch height  . According Shepard’s (1982) elegant explanation, the chromatic circle is a geometrical spacethat shows relationships among the 12 equal-tempered pitch classes making up the familiar chro-matic scale (see Fig. 1). Pitch chroma perception refers to the similarity between tones spaced ata doubling in frequency (e.g., notes that are separated by an octave) observed in humans (Ward& Burns, 1982) and other primates (Wright, Rivera, Hulse, Shyan, & Neiworth, 2000). It is impor- tant to draw attention to the fact that attempts to demonstrate accurate pitch chroma perceptionin songbirds have not been successful (Cynx, 1993). In considering the significance of AP for pitch chroma, it is also important to know that only a tiny proportion of humans, mainly musicians thatbegan music training early in life, have highly accurate AP for pitch chroma (Levitin & Zatorre,2003).In contrast, pitch height is the rectilinear component of AP: a one-dimensional perceptual repre-sentation of pitch, in which tones spaced closer together, on a log scale, are perceived as more similarthantonesspacedfartherapart.APforpitchheightiscommoninhumansandextendsbroadlyacrossspecies.Forexample,Morton(1994)reviewedstudiesofthevocalizationsof54speciesandconcludedthat low-pitched signals are generally associated with aggressive responses, whereas high-pitchedsignals are generally associated with friendly, appeasing, or fearful responses. Also, human subjectsperceivemalevoicespitchshiftedtolowerfrequenciesasmoresociallyandphysicallydominantthanunshifted voices (Puts, Hodges, Cárdenas, & Gaulin, 2007). Clearly, AP for pitch height is adaptive. TotheseadaptivefunctionsofheightAP,weaddtheavianabilitytoaccuratelytrackthepitchesandpitchranges in conspecific vocalizations.The purpose of the present research was to examine the generality of the phylogenetic rule thatavian species have accurate AP for pitch height whereas mammalian species are much less accurate.  R.G. Weisman et al. / Learning and Motivation 41 (2010) 156–173  159 Fig. 1.  A geometric model for the representation of pitch as a function of increasing frequency. The helix places a musical notedirectlyaboveitsrepresentationinloweroctavesanddirectlybelowitsrepresentationinhigheroctavesthusprovidingabasisfortheperceptionpitchchroma.Independentofpitchchroma,increasingfrequencycanprovideadirectrepresentationofpitchheight. Adopted from Shepard (1982). Our approach was to consider possible exceptions that might test the rule that phylogeny deter-mines AP for pitch height in frequency-range discriminations. In testing the explanatory power of ageneralization, the study of exceptions has much to recommend it (Popper, 1979). In the course of  recent research, we have uncovered two rather jarring possible exceptions: in the frequency-rangediscriminationsofhumanpitchchromaAPpossessors(Experiment1)andofborealchickadees( Poecilehudsonicus , Experiment 2). Experiment 1 We undertook Experiment 1 in search of a mammalian exception to the phylogenetic determi-nant rule. In previous research, we found that human performance diverged negatively from theaccurate discriminations of avian species in the more difficult, eight-range, frequency-range discrim-ination (Weisman et al., 1998). In that research we tested people who were musicians but not AP possessors.Here,wewerethefirsttostudythefrequency-rangediscriminationsofhumanchromaAPpossessors.Our research group and colleagues were divided in their predictions about the outcome of a com-parison between chroma AP possessors and nonpossessors. Some of us predicted that AP possessorsmight not track the relationship between frequency ranges and reward any better than nonposses-sors. Their reasoning was that the pitch height AP of both groups is mediocre and about equal andtherefore not capable of tracking frequency ranges. Others predicted that AP possessors might trackranges better than nonpossessors. Their reasoning was that AP possessors might succeed in mappingthe tones used in these frequency-range discriminations into pitch chroma and then into rewarded  160  R.G. Weisman et al. / Learning and Motivation 41 (2010) 156–173 and unrewarded note classes. Here we report on the outcome of frequency-range discrimination inAP possessors and its explanation. Method Subjects We recruited subjects at the University of Alberta and Queen’s University by advertisements inlocal newspapers and posters on campus. At both universities, each with its own music program, wespecificallyaskedthatpeoplewithAPvolunteer;thesuccessoftheserequestslikelyaccountsforwhywe were able to test so many AP possessors.Subjects were 18 years of age or older with a minimum of 4 years of musical training. The Arts,Science & Law Research Ethics Board and General Research Ethics Board at the University of AlbertaandQueen’sUniversity,respectively,approvedourresearchprotocols.Of39peoplescreenedinitially,23 were available to participate in eight sessions of discrimination training.  Apparatus We presented our procedures (programmed in Visual Basic) on a Toshiba Tecra laptop (Intel Pen-tium M processor, Intel 855 series chip set, and the sound card supplied by Toshiba) and SennheiserHD 580 headphones. Procedure and stimuli In brief, the procedure had three phases: an initial screening test of AP, frequency-range discrim-ination training, and finally a second test of AP which presented the tones used in frequency-rangediscrimination training. For convenience, our AP tests are referred to as the initial and final tests and,to distinguish them from the AP tests, the earliest and latest sessions of discrimination training arereferred to as the first and last sessions, respectively.In the  first phase , we screened human chroma AP possessors from nonpossessors using a note-naming test developed by Athos et al. (2007), who studied 2213 subjects. The 40 sinewave tones presented in the note-naming test were synthesized at the frequencies of the musical notes (on the12-note semitone scale) from C2 to G8, each 1000ms in duration. These sinewave tones and otherspresentedinthisresearchwereconstructedatastandard16-bit,44.1kHzsamplingratewithoutclicksor spectral artifacts. During the test, a trial began when the subject clicked on the “play note” button,at the top of the screen, and heard a tone selected randomly and without replacement from the 40tonesavailable.Tonamethemusicalnotecorrespondingtoatone,thesubjectclickedononeofthe12black and white piano keys shown on the screen. Then to identify the octave of the note, the subjectclicked on one of eight numbered squares arranged horizontally on the screen below the keyboard.Thetestcontinuedwithoutfeedbackuntilthesubjectheardall40tones.TheAPtestbeganafterashortpractice session, eight trials, given to acquaint subjects with making mouse responses to the graphicson the computer screen and to allow subjects, individually, to adjust tone volume to the headphonesto a comfortable level, using a rotary control on the front edge of the computer. It was not possible tomonitor the subjects’ choice of amplitude as they wore the headphones throughout the experimentand could adjust the volume at anytime they chose.Inthe secondphase ,thefrequency-rangediscrimination,subjectsweretrainedtosort40sinewavetones, each heard 11 times during each training session, into 8 ranges of ascending frequencies (seeTable1).Eachrangeconsistedof5contiguoustones.IntheS+firstdiscriminationgroup,S+(go)rangesalternatedwithS − (no-go)ranges,withthe1st,3rd,5thand7thrangesdesignatedasS+ranges.IntheS − first discrimination group, the order was counterbalanced so that the 2nd, 4th, 6th and 8th rangeswere designated as S+ ranges.We used SIGNAL 4.0 for Windows (Engineering Design, Berkeley, CA, USA) to synthesize the 40sinewavetones,spacedat120Hzapartfrom980Hzto5660Hz,atastandard16-bit,44.1-kHzsamplingrate. Each tone was 440ms in duration and ramped at onset and offset for 5ms to avoid transients.
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!