96:1441-1455, 2006. First published May 24, 2006;  doi:10.1152/jn.01138.2005 J NeurophysiolMimi H. Kao and Michael S. Brainard Context-Dependent Changes to Song Variability Lesions of an Avian Basal Ganglia Circuit Prevent

You might find this additional information useful...

42 articles, 15 of which you can access free at: This article cites http://jn.physiology.org/cgi/content/full/96/3/1441#BIBL

including high-resolution figures, can be found at: Updated information and services http://jn.physiology.org/cgi/content/full/96/3/1441

can be found at: Journal of Neurophysiologyabout Additional material and information http://www.the-aps.org/publications/jn

This information is current as of August 14, 2006 .  

http://www.the-aps.org/.American Physiological Society. ISSN: 0022-3077, ESSN: 1522-1598. Visit our website at (monthly) by the American Physiological Society, 9650 Rockville Pike, Bethesda MD 20814-3991. Copyright © 2005 by the

publishes original articles on the function of the nervous system. It is published 12 times a yearJournal of Neurophysiology

on August 14, 2006

jn.physiology.orgD

ownloaded from

Lesions of an Avian Basal Ganglia Circuit Prevent Context-DependentChanges to Song Variability

Mimi H. Kao and Michael S. BrainardKeck Center for Integrative Neuroscience, Departments of Physiology and Psychiatry, University of California, San Francisco, California

Submitted 27 October 2005; accepted in final form 17 May 2006

Kao, Mimi H. and Michael S. Brainard. Lesions of an avian basalganglia circuit prevent context-dependent changes to song variability.J Neurophysiol 96: 1441–1455, 2006. First published May 24, 2006;doi:10.1152/jn.01138.2005. Trial-by-trial variability is important infeedback-based motor learning. Variation in motor output enablesevaluation mechanisms to differentially reinforce patterns of motoractivity that produce desired behaviors. Here, we studied neuralsubstrates of variability in the performance of adult birdsong, acomplex, learned motor skill used for courtship. Song performance ismore variable when male birds sing alone (undirected) than when theysing to females (directed). We test the role of the anterior forebrainpathway (AFP), an avian basal ganglia–forebrain circuit, in thissocially driven modulation of song variability. We show that lesionsof the lateral magnocellular nucleus of the anterior nidopallium(LMAN), the output nucleus of the AFP, cause a reduction in themoment-by-moment variability in syllable structure during undirectedsong to the level present during directed song. This elimination ofsong modulation is immediate and long-lasting. We further show thatthe degree of syllable variability and its modulation are both attenu-ated in older birds, in concert with decreased variability of LMANactivity in these birds. In contrast to the requirement of LMAN forsocial modulation of syllable structure, we find that LMAN is notrequired for modulation of other features of song, such as the numberof introductory elements and motif repetitions and the ordering ofsyllables or for other motor and motivational aspects of courtship. Ourfindings suggest that a key function of avian basal ganglia circuitry isto regulate vocal performance and plasticity by specifically modulat-ing moment-by-moment variability in the structure of individual songelements.

I N T R O D U C T I O N

A hallmark of well-learned skills is that they exhibit a highdegree of trial-by-trial reproducibility. Such stereotyped motoroutput is often considered an endpoint of motor learning, andresidual variability in motor performance may be construed asbiological noise that either the nervous system is unable toeliminate or that is below threshold for behavioral importance.In the context of learning and ongoing calibration of motorperformance, however, trial-by-trial motor variability plays apotentially critical role; variation in motor output enablesmotor exploration such that evaluation mechanisms can differ-entially reinforce patterns of motor activity that produce moredesired outcomes and weaken or punish patterns of activity thatproduce worse outcomes. While such an adaptive function ofvariability is well established in models of reinforcement-basedmotor learning (Doya and Sejnowski 2000; Sutton and Barto1998), whether and how the nervous system actively regulates

variability in the context of motor practice and ongoing motorperformance is little understood. Here, we studied neural sub-strates of variability in the motor control of adult birdsong.

Song learning shares many features with other forms ofvertebrate motor skill learning (Doupe and Kuhl 1999). Ini-tially, during a sensory learning period, birds memorize thesound of an adult tutor song. Subsequently, during a sensori-motor learning period, young birds begin to vocalize and useauditory feedback to gradually refine their vocalizations untilthey closely approximate the tutor song. Initial vocalizationsare highly variable from one rendition to the next. However, asthe bird’s song approaches the end state (a good match to thetutor song), it becomes increasingly stereotyped. In the zebrafinch, this process is completed by �90 days posthatch (dph),at which point the song is said to be crystallized (Immelmann1969). Despite the apparent stability of adult zebra finch song,disruptions of feedback, such as deafening, lead to a gradualdeterioration of performance, indicating that adult experiencecontinues to contribute to maintenance of song (Leonardo andKonishi 1999; Nordeen and Nordeen 1992). Several studieshave suggested that learning and maintenance of song may relyin part on the regulation of song variability for purposes ofmotor exploration (Doya and Sejnowski 2000; Jarvis et al.1998; Kao et al. 2005; Olveczky et al. 2005; Scharff andNottebohm 1991).

Here, we studied the role of the anterior forebrain pathway(AFP), an avian basal ganglia–forebrain circuit (Perkel 2004),in the regulation of song variability. The AFP is an accessoryloop that is interconnected with the motor structures for songproduction (Fig. 1A) and, hence, is well positioned to influencesong. Previous lesion studies have shown that the AFP iscritical for song learning and feedback-based vocal plasticityand have suggested a possible role for this pathway in modu-lating song variability (Bottjer et al. 1984; Brainard and Doupe2000, 2001; Kao et al. 2005; Olveczky et al. 2005; Scharff andNottebohm 1991; Williams and Mehta 1999). In juvenile zebrafinches, lesions or inactivation of the lateral magnocellularnucleus of the anterior nidopallium (LMAN), the output nu-cleus of the AFP, disrupt song development and induce pre-mature stereotypy, resulting in highly repetitive, simplifiedsongs (Bottjer et al. 1984; Olveczky et al. 2005; Scharff andNottebohm 1991). In adult zebra finches, lesions of LMANhave little overt effect on previously learned song, but theyprevent song plasticity induced by experimental manipulationof auditory and/or proprioceptive feedback (Bottjer et al. 1984;Brainard and Doupe 2000, 2001; Nordeen and Nordeen 1993;

Address for reprint requests and other correspondence: M. S. Brainard,Dept. of Physiology, Box 0444, 513 Parnassus Ave., Univ. of California, SanFrancisco, CA 94143-0444 (E-mail: msb@phy.ucsf.edu).

The costs of publication of this article were defrayed in part by the paymentof page charges. The article must therefore be hereby marked “advertisement”in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

J Neurophysiol 96: 1441–1455, 2006.First published May 24, 2006; doi:10.1152/jn.01138.2005.

14410022-3077/06 $8.00 Copyright © 2006 The American Physiological Societywww.jn.org

on August 14, 2006

jn.physiology.orgD

ownloaded from

Scharff and Nottebohm 1991; Williams and Mehta 1999).Consistent with a role in regulating song plasticity, LMANneurons exhibit robust singing-related activity that is correlatedwith song structure (Hessler and Doupe 1999a,b). Moreover,manipulation of neural activity in LMAN during singing canalter the spectral structure of individual song elements, orsyllables, on a moment-by-moment basis (Kao et al. 2005).These findings suggest that a key function of the AFP inregulating vocal plasticity may be to modulate ongoing pat-terns of activity and synaptic connections in the motor pathway(Jarvis et al. 1998; Kao et al. 2005; Kittelberger and Mooney1999; Scharff and Nottebohm 1991). Such a modulatory influ-

ence could serve to introduce variability in the motor pathwayand subsequent song, which is critical in reinforcement learn-ing (Brainard 2004; Doya and Sejnowski 2000; Kao et al.2005; Olveczky et al. 2005; Sutton and Barto 1998).

To examine the hypothesis that LMAN actively introducesvariability into patterns of activity in the motor pathway, wefocused on naturally occurring, context-dependent regulationof variability in adult song structure. Song serves as a courtshipsignal in zebra finches, and previous studies have shown that avariety of aspects of song structure, including its variability,are sensitive to social context (Kao et al. 2005; Sossinka andBohner 1980). When a male zebra finch sings alone (undi-rected song), variability in syllable structure is consistentlygreater than when he sings to a female (directed song) (Kao etal. 2005; Olveczky et al. 2005). Moreover, these context-dependent changes in song structure are associated with con-text-dependent changes in neural activity in LMAN. Singing-related activity in LMAN is greater in magnitude and morevariable in pattern across repeated renditions during undirectedsong than during directed song (Hessler and Doupe 1999b;Jarvis et al. 1998; Kao et al. 2005). These findings suggest thatgreater variability in LMAN activity during undirected singingmay actively generate variability in ongoing activity in thesong motor pathway and subsequent song output (Fig. 1B).Alternatively, or in addition, stereotyped patterns of LMANactivity during directed song may facilitate reproducibility inmotor pathway activity and song.

To examine how LMAN modulates song variability, wecompared songs produced in the two social contexts before andafter lesions. We previously reported that lesions of LMANeliminate context-dependent differences in the variability ofsyllable structure (Kao et al. 2005). Here, we further study thenature of the effects of LMAN lesions. If LMAN activelygenerates variability in undirected song, lesions should reduceor eliminate variability in undirected song (Fig. 1, B and C).Alternatively, if LMAN actively reduces variability duringdirected song, lesions should cause an increase in variability indirected song toward the level present in undirected song. Inaddition, to test the specificity and nature of the role of LMANin modulating variability, we measured the influence of LMANlesions on other features of song and behavior that are knownto differ with social context. In particular, we asked whetherlesions specifically eliminate the social modulation of individ-ual syllables or whether they more generally eliminate modu-lation of the entire suite of behaviors that differentiate directedand undirected song (Sossinka and Bohner 1980). Finally, tofurther test the relationship between LMAN activity and songvariability, we measured these variables in older adult birds,whose songs are less plastic than those of young adults (Brai-nard and Doupe 2001; Lombardino and Nottebohm 2000). Ourfindings support a role for LMAN in actively introducingvariability to specific aspects of song structure during undi-rected song.

M E T H O D S

Subjects

Thirty male zebra finches (Taeniopygia guttata) were used forexperiments. All birds were raised in individual breeding cages withtheir parents and siblings until �60 dph. Birds were selected on thebasis of their size, singing frequency, and song complexity, and were

FIG. 1. Schematic model for the contribution of an avian basal ganglia–forebrain circuit to social influences on song production. A: song systemnuclei. Motor pathway (gray) is required for normal song production through-out life and includes HVC (used as a proper name; Reiner et al. 2004), robustnucleus of the arcopallium (RA), and tracheosyringeal portion of the hypo-glossal nucleus (nXIIts). Anterior forebrain pathway (AFP; black) is necessaryfor vocal motor plasticity but is not required for song production. AFP includesArea X, which is homologous to mammalian basal ganglia, medial nucleus ofthe dorsolateral thalamus (DLM), and lateral magnocellular nucleus of theanterior nidopallium (LMAN). B: when a male sings alone (undirected song),variability in patterns of LMAN activity may introduce variability into activityin the premotor nucleus RA, resulting in variable song output. In contrast,when a male sings to a female (directed song), stereotyped patterns of LMANactivity may facilitate reproducibility in activity of RA neurons and subsequentsong output. C: according to this hypothesis, lesions of LMAN shouldeliminate socially driven differences in song variability.

1442 M. H. KAO AND M. S. BRAINARD

J Neurophysiol • VOL 96 • SEPTEMBER 2006 • www.jn.org

on August 14, 2006

jn.physiology.orgD

ownloaded from

housed individually in sound-attenuating chambers (Acoustic Sys-tems, Austin, TX). All procedures were performed in accordance withprotocols approved by the UCSF Institutional Animal Care and UseCommittee.

Surgical procedures

Before all surgical procedures, birds were deprived of food andwater for 1 h and anesthetized with an intramuscular injection ofEquithesin (0.85 g chloral hydrate (Sigma, St. Louis, MO), 0.21 gpentobarbital (Abbott Laboratories, North Chicago, IL), 0.42 gMgSO4, 2.2 ml 100% ethanol, and 8.6 ml propylene glycol to a totalvolume of 20 ml with water). After surgery, all skin incisions weresealed with a cyanoacrylate adhesive.

LESIONS. Bilateral electrolytic lesions were stereotaxically targetedat LMAN, with five penetrations per side and one or two currentinjections per penetration (50 or 100 �A for 60 s). The amount ofLMAN that was removed bilaterally ranged from 40 to 100% (medianlesion size: 80%; see Table 1; n � 5 birds), as confirmed by histologyat the end of experiments. “Sham” lesions were entirely dorsal toLMAN (Table 1; n � 2 birds). In all birds, the medial magnocellularnucleus of the anterior nidopallium (MMAN) remained intact. Lesionswere made in birds between 101 and 123 dph (n � 7 birds). We didnot find any correlation between the extent of the lesions and theireffects on song output using a variety of measures, including theaverage degree of modulation of fundamental frequency by socialcontext after lesions of LMAN (P � 0.69) and the average degree ofmodulation of song tempo by social context after lesions of LMAN(P � 0.49). Moreover, the reported effects of LMAN lesions on thesong of the bird with the smallest lesions ranked third overall (me-dian) for changes in the variability of fundamental frequency (70%lesion on the left; 10% lesion on the right as well as a visible reductionin CGRP labeling in the lateral RA on the left and dorsal RA on theright). Hence, we pooled all lesion birds in the subsequent analyses.Restriction of the data set to the four birds with the largest lesions(mean lesion size of 88%) did not change the significance of any ofthe reported results.

CHRONIC IMPLANTS. Electrodes were implanted chronically as de-scribed previously (Hessler and Doupe 1999a). Briefly, a lightweightmicrodrive (UCSF and Caltech Machine Shops) carrying two tungstenelectrodes (2–5 MOhm) insulated with epoxylite (FHC, Bowdoinham,ME) or glass (A. Ainsworth, Northhamptonshire, UK; Merrill andAinsworth 1972) was positioned stereotaxically such that the elec-trode tips were �700 �m above LMAN. A reference ground electrode(uninsulated tungsten electrode, A-M Systems, Carlsborg, WA) wasimplanted such that its tip was located within �2 mm of the targetedLMAN. The microdrive and connector socket (FHC) were secured tothe skull with epoxy (Devcon, Wood Dale, IL) and dental cement(Dentsply, Milford, DE), and a protective cap was fixed around themicrodrive. All electrodes were implanted in the right hemisphere.

Anatomy

After the final recordings, birds were deeply anesthetized withMetofane (Schering-Plough, Union, NJ) and transcardially perfusedwith 0.9% saline, followed by 3.7% formaldehyde in 0.025 M phos-phate buffer. Brains were postfixed for 4 h, cryoprotected, and cutcoronally in 40-�m sections with a freezing microtome. Every thirdsection was stained with cresyl violet acetate, and a second set ofintervening sections was reacted with an antibody to calcitonin gene-related peptide (Fig. 2B; CGRP; Sigma, St. Louis, MO) (Bottjer et al.1997; Brainard and Doupe 2001).

Sound recording

Acoustic signals were recorded by a microphone located above thebirdcage and filtered between 200 and 9,000 Hz (Krohn-Hite, Avon,MA). Custom-written acquisition software (C. Malek and A. Leo-nardo, Caltech, and C. Roddey, UCSF) recorded the acoustic signalsbefore and after the sound amplitude crossed a threshold level. Thethreshold level was determined experimentally to ensure detection ofall vocalizations (calls and songs) and to minimize detection of othersounds, such as the bird flapping its wings. Each bird’s behavior wasmonitored and recorded by a video camera.

For all experiments, undirected song was recorded when the malewas isolated in a sound-attenuating chamber. The amount of undi-rected song was often greater when the experimental bird could hearthe calls of other birds outside the recording chamber. To elicitdirected song, one or more female zebra finches was presented in aseparate cage to the male zebra finch being recorded. The recordedbird usually moved to the edge of its cage and sang while facing thefemale(s). Each female presentation lasted for �2 min, regardless of

-2 wk -1 wk +1 wk +2 wk +3 wk +4 wk-1 d < 3d

Lesion

+8 wk

SHAM lesion

Area X

MMANLMAN

RA

LMAN lesionMMAN

Area X

RA

A

B

FIG. 2. Experimental design and song system nuclei in intact and lesionedbirds. A: interleaved bouts of directed and undirected song were recordedweekly before and after lesions at each time-point indicated by tick marks. B:coronal sections through AFP (top) and arcopallium (bottom) labeled with anantibody against calcitonin gene–related protein (CGRP; Bottjer et al. 1997).Electrolytic lesions were stereotaxically targeted either at LMAN or at a sitedorsal to LMAN (sham). In a sham control (left), LMAN, its axonal projectionsto target nuclei RA and Area X, and medial MAN (MMAN) are all immuno-labeled. In a lesioned bird (right), CGRP labeling in LMAN is absent, andthere is a corresponding loss of labeled axonal projections to RA and Area X.MMAN remains intact.

TABLE 1. Extent of lesions

Bird Extent of Lesion (Left/Right)

Lesionpur24w55 60% lesion/80% lesionpur92w89 100% lesion/100% lesionb19o61 �90% lesion/70% lesionp94g48 100% lesion/100% lesionp59w36 70% lesion/10% lesion

Controlpur100b28 Sham lesion—no damage to LMANpur69b29 Sham lesion—no damage to LMAN

LMAN, lateral magnocellular nucleus of the anterior nidopallium.

1443LMAN LESIONS PREVENT CHANGES TO SONG VARIABILITY

J Neurophysiol • VOL 96 • SEPTEMBER 2006 • www.jn.org

on August 14, 2006

jn.physiology.orgD

ownloaded from

whether or not the male sang, and songs were classified as directedonly when the male bird faced the female(s). Bouts of directed songwere interleaved with undirected song during each recording. Record-ings lasted between 30 min and several hours depending on theamount of singing.

For experiments examining the effects of lesions of LMAN oncontext-dependent differences in song, interleaved bouts of directedand undirected song were recorded as shown in Fig. 2A. Recordingswere made in each bird once a week for 2–3 wk preceding surgery,again on the day before bilateral lesions, and after lesions as follows:birds were recorded every day after lesions until enough songs werecollected in both social conditions to characterize initial effects oflesions (1–3 days) and once a week for 4 wk after the lesions. A finalrecording was made during the eighth week after lesions (n � 5 birdswith LMAN lesions; n � 2 birds with sham lesions). Three additionalage-matched control birds were also recorded weekly. The range ofages of the birds at the first recording was 88–102 dph. Data from asubset of these recordings have been presented in a brief report (Kaoet al. 2005).

To study the effect of age on context-dependent differences in songstructure, we recorded songs from an additional 6 birds �4 yr old and11 birds �6 mo old.

Physiological recording

In a separate group of birds (n � 7), we recorded neural activity atmultiple sites in LMAN (1–6 sites/bird) in directed and undirectedconditions. During each recording, a flexible lead terminating in asmall operational amplifier circuit was connected to the socket on thebird’s head, and the other end was connected to a commutator(Caltech Machine Shop). Electrodes were positioned at sites inLMAN where action potentials of single and multiple neurons couldclearly be differentiated from the background neural activity (spikeamplitudes ranged from 300 �V to �1 mV, peak-to-peak; Hessler andDoupe 1999a). The neural activity signal passed through the commu-tator to a differential amplifier (A-M Systems) and was filteredbetween 300 Hz and 10 kHz. The acoustic signal was recorded asdescribed above. Custom-written software (A. Leonardo, Caltech, andC. Roddey, UCSF) recorded the acoustic and neural signals, and thebird’s behavior was monitored and recorded by a video camera.

Simultaneous song and neural recordings were made at intervals of1d to several weeks over a period of weeks to months. Neural activitywas recorded during nonsinging and singing periods in both undi-rected and directed conditions. After completion of recordings in eachbird, small electrolytic lesions (30 �A for 10 s) were made atpreviously recorded sites. Locations of recording sites were confirmedin 40-�m Nissl-stained brain sections by their positions relative to thedepth of the marker lesions.

Behavioral analysis

SONG STRUCTURE. Zebra finch song can be classified into threelevels of organization: syllables, which are individual song elementsseparated by silent intervals �5 ms in duration; motifs, which arestereotyped sequences of syllables; and bouts, which are defined asperiods of singing separated by silent intervals �2 s in duration(Sossinka and Bohner 1980). Song bouts usually consist of a series ofintroductory elements followed by a variable number of repetitions ofthe same motif. Song bouts may either be aimed at another bird(directed) or sung when the male is alone or not orienting toward anyother bird in particular (undirected) (Dunn and Zann 1996; Sossinkaand Bohner 1980).

ANALYSIS OF SYLLABLE STRUCTURE. To characterize differences inthe structure of individual syllables, we measured the fundamentalfrequency (FF) of syllables that have constant frequency components(Kao et al. 2005). For a particular syllable, we calculated the auto-

correlation of a segment of the sound waveform that had clearharmonic structure (median segment used to calculate FF: �50% ofthe total syllable duration; range � 30–85%). The FF was defined asthe distance, in frequency, between the zero-offset peak and thehighest peak in the autocorrelation function (ACF) within a range oftime lags corresponding to fundamental frequencies typical of zebrafinch syllables. To improve the resolution of the frequency estimates,we performed a parabolic interpolation of the peak of the ACF (deCheveigne and Kawahara 2002). This algorithm was applied tosyllables that had clear harmonic structure with a well-defined FF. Toexamine differences between directed and undirected songs, the FFwas calculated for a minimum of 16 renditions (range: 16–90 rendi-tions) in each behavioral context.

ANALYSIS OF OTHER SONG FEATURES. We analyzed four otherfeatures of song that previously have been reported to differ acrosssocial context (Sossinka and Bohner 1980): number of introductoryelements, number of motifs per bout, the stereotypy of syllableordering, and song tempo.

To count the number of introductory elements per song bout, westarted with the introductory element preceding the first syllable in thesong motif and counted backward until there was �500 ms of silenceor a different type of vocalization, such as a loud distance call. Usingthis method, we found consistent differences in the number of intro-ductory elements between directed and undirected song bouts (seeRESULTS). For the number of motifs per bout, we counted all motifs inwhich at least the first half of the motif was sung because zebrafinches often truncate the last motif in a song bout (e.g., if thecanonical motif was abcdef, all motifs that included at least abc werecounted).

The stereotypy of syllable ordering was quantified by using mea-sures of sequence linearity and sequence consistency, similar to thoseoriginally described by Scharff and Nottebohm (1991). We followedthe practice of recent studies that have examined stereotypy ofsyllable sequencing and excluded introductory elements from thesemeasures (Foster and Bottjer 2001; Olveczky et al. 2005; Zevin et al.2004).

Sequence linearity quantifies the different possible transitions thatcan be observed after each unique syllable of song. In this study, weused the internal linearity measure of Foster and Bottjer (2001), whichexcludes variability in the syllable that precedes song terminations.Internal linearity is defined as

Linearity ��number of different syllables� � 1

number of syllable-to-syllable transitions

For example, if a bird always sings the motif abcde, there are fiveunique syllables and four unique transitions, and the internal linearityyields a value of 1.00. Any variation in the sequencing of syllableswithin the motif will yield a value �1.00. However, songs in whichmotifs are terminated after more than one syllable still receive a scoreof 1.00 (i.e., occurrences of truncated motifs abc and abcd would notreduce the internal linearity score). Because internal linearity excludesvariability in song terminations, it always yields a value that is equalto or higher than the sequence linearity score calculated by Scharffand Nottebohm (1991).

To explicitly characterize any context-dependent differences inpremature song terminations, we calculated the percentage of syllabletypes within a motif that were followed by song terminations. Forexample, if a bird’s typical motif was abcde, but the truncatedversions abc and abcd were also observed, the percent of syllables thatterminate song would be 60 (3/5).

To measure the frequency with which the main motif sequenceoccurred, we used the sequence consistency score of Scharff andNottebohm (1991), excluding variability in introductory elements.Sequence consistency measures the proportion of syllable transitionsthat conform with the most frequent, or dominant, transition for agiven syllable and is defined as

1444 M. H. KAO AND M. S. BRAINARD

J Neurophysiol • VOL 96 • SEPTEMBER 2006 • www.jn.org

on August 14, 2006

jn.physiology.orgD

ownloaded from

Consistency �� dominant transitions/song

� all transitions/song

Complete consistency yields a score of 1.00. Because transitions tothe end of a song bout are included in this measure, it is influenced byvariability at the end of songs.

Because these measures of sequence stereotypy are sensitive to thenumber of songs that are analyzed, we calculated linearity, consis-tency, and percent syllables terminating song from randomly selectedsubsets of recorded songs that were matched to have equal numbers ofmotif initiations in the directed and undirected conditions (mean andrange of motifs per bird: 67; 36–125).

To characterize any context-dependent differences in song tempo,or the rate at which song is delivered, we measured for each bird thedurations of complete motifs during directed and undirected songs.For each recording, we computed the mean motif duration for bothbehavioral conditions.

ANALYSIS OF COURTSHIP BEHAVIOR. Directed song serves as acourtship signal and is often accompanied by a rhythmic, pivotingdance that includes orienting toward and approaching the female,hopping to and fro, head-tail twists and bows, beak wipes, and acharacteristic posture (reviewed in Zann 1996). To characterize thesemotor aspects of courtship, we scored video clips of directed andundirected singing. Video clips of directed and undirected song bouts(�30–60 s) recorded 1 day prelesions and 7 days postlesions wereinterleaved, and the clips were organized in blocks by bird because theexact form of the courtship dance varies across males. Two observershighly familiar with zebra finch behavior and blind to the experimen-tal manipulation of each bird (lesions of LMAN or sham lesions)judged the degree of arousal and the vigor of the courtship display ofeach male using a scale of 0 (relatively inactive and no dancemovements) to 3 (highly aroused courtship display). The observerswere instructed to note the following features of the courtship dance:the male’s posture, orienting toward the female, approaching thefemale, hopping to and fro, and beak wiping. During undirectedsinging, when no female was present, scores could be �0 if the malesang in the same position or oriented in the same direction as it didduring directed singing. Scores were averaged across song bouts in aparticular social context on each day to derive a mean behavioralscore. Scores were averaged across the observers.

Analysis of neural signals

Analysis of singing-related activity in LMAN was performed asdescribed previously (Hessler and Doupe 1999a,b). Briefly, rectified,smoothed neural activity waveforms were aligned using a template forthe amplitude envelope of each bird’s motif. Both the mean activitylevel and the coefficient of variation of activity across motif renditionswere calculated.

R E S U L T S

Previous studies have characterized several differences insong depending on the social context in which it is produced(Kao et al. 2005; Olveczky et al. 2005; Sossinka and Bohner1980). Sossinka and Bohner (1980) reported that when a malesings to a female (directed song), song is preceded by moreintroductory elements per bout, includes more repetitions of astereotyped sequence of syllables, or motif, per bout, is morestereotyped in the sequencing of syllables, and is delivered ata faster tempo than when the male sings alone or to noparticular audience (undirected song; Figs. 3A and 8A). Inaddition, stereotypy in the structure of individual song ele-ments is greater during directed song than during undirectedsong (Fig. 3B; Kao et al. 2005). Finally, directed song is often

accompanied by changes in the orientation, posture, and posi-tion of the male, which collectively constitute a courtshipdance (Williams 2001; Zann 1996). In the following sections,we describe the contribution of LMAN to each of thesecontext-dependent changes to song and behavior.

Context-dependent differences in variability of syllablestructure require LMAN

To assess the contribution of LMAN to ongoing motorperformance and song plasticity, we first characterized thespectral structure of individual syllables during directed andundirected song. We focused our analysis on the fundamentalfrequency of syllables that contain constant frequency compo-

US

DS

8000

Fre

quen

cy

1000 ms

0

i i i i i i i i i i b c a b c i i b c a b c a b c a b c a b c

i i i i b c a b c a b c

100 ms

a b c

US

DS

Cou

ntC

ount

Cou

ntC

ount

Fundamental Frequency (Hz)Fundamental Frequency (Hz)500 520 540 560

500 520 540 560 715 725 735 745 755

715 725 735 745 755

0

10

30

20

0

10

30

20

0

10

20

0

10

20

A

B

syllable 'a' syllable 'c'

FIG. 3. Context-dependent differences in song output. A: spectrogram (plotof frequency vs. time) of 1 bout of song recorded when a male was alone[undirected song (US); top] and when it sang to a female [directed song (DS);bottom]. Song bouts consist of a series of introductory elements followed by avariable number of repetitions of a stereotyped sequence of syllables, or motif(abc, indicated by bars). Number of introductory elements and motifs per boutare typically lower during US than during DS (Sossinka and Bohner 1980). B:variability in syllable structure is greater in the undirected condition. Spectro-gram of the same bird’s motif (abc; top). Syllables a and c contain constantfrequency components with a well-defined fundamental frequency (FF). Dot-ted lines indicate portion of syllable for which FF was measured. Histogramsof the FF for indicated portions of syllables a (left) and c (right). Variability inthe FF was greater during US than during DS (syllable a: SD: 8.8 vs. 3.4; P �0.0001; syllable c: SD: 7.2 vs. 2.5; P � 0.0001, F test for equality of variance).

1445LMAN LESIONS PREVENT CHANGES TO SONG VARIABILITY

J Neurophysiol • VOL 96 • SEPTEMBER 2006 • www.jn.org

on August 14, 2006

jn.physiology.orgD

ownloaded from

nents with clear harmonic structure. FF is a learned parameterof song that is precisely controlled (Tchernichovski et al. 2001)and robust to differences across recording conditions. In addi-tion, microstimulation in LMAN during singing can inducechanges in FF on a moment-by-moment basis, supporting theidea that neural activity in LMAN modulates ongoing perfor-mance of this vocal feature (Kao et al. 2005).

As previously reported (Kao et al. 2005), we found that innormal adult zebra finches, there was a striking modulation inthe variability of FF by social context. Figure 3B shows thevariability of FF for two syllables (a and c) produced by anadult male in interleaved bouts of directed and undirected song.As shown here, the distribution of FF for an individual syllableduring undirected song was typically broader than the distri-bution of FF for the same syllable produced during directedsong.

In control and sham-lesioned birds, context-dependent dif-ferences in the variability of syllable structure were extremelyrobust. Each symbol in Fig. 4A shows the variability in FFexpressed as the coefficient of variation (CV) for a singlesyllable produced in interleaved recordings of directed andundirected song. Data are shown for all measured syllables inall recording sessions from two control and two sham-lesionedbirds. The diagonal line indicates equal variability under thetwo social conditions. For all groups of birds, the data pointsare distributed significantly above the diagonal, indicatinggreater variability during the undirected condition (control:2–6 syllables/bird; 9 recording sessions; n � 72; presham-lesion: 2–4 syllables/bird; 3 recording sessions; n � 18; andpostsham-lesion: 2–4 syllables/bird; 6 recording sessions; n �34; P � 0.0001, paired sign test).

Lesions of LMAN completely eliminated context-dependentdifferences in the variability of FF that were present in normaladults. Figure 4B shows variability in FF before and afterbilateral lesions of LMAN for all measured syllables in allrecording sessions from five birds. Before lesions (open sym-bols), there was a robust modulation of syllable variability bysocial context; data points lie above the diagonal line (2–6syllables/bird; 3 recording sessions; n � 57; P � 0.0001,paired sign test), indicating greater variability during undi-rected song. After lesions (filled symbols, 2–6 syllables/bird; 6recording sessions; n � 111), the data points were distributedalong the diagonal, and there was no longer a significantmodulation of variability by social context.

The elimination of context-dependent modulation of syllablevariability by lesions of LMAN was rapid and long-lasting.Figure 5 shows the variability of FF at each experimentaltime-point for both undirected (open bars) and directed condi-tions (shaded bars). For control and sham-lesioned birds, therewere significant context-dependent differences in song vari-ability for all experimental time points (Fig. 5A). In contrast,context-dependent differences in the variability of syllablestructure were absent in the first recordings postlesions (1–3days after) and remained absent for as long as birds werefollowed (�8 wk; Fig. 5B).

The loss of context-dependent differences in variabilitycould be attributed primarily to a reduction in the variability ofsyllables produced during undirected song to the level presentduring directed song. This is shown in Fig. 4B by the predom-inantly downward shift in the population of syllables to thediagonal line (open and filled X indicates population means for

pre- and postlesion recordings, respectively). Indeed, lesions ofLMAN caused a significant decrease in the variability ofundirected songs (Fig. 5B, P � 0.005, paired sign test for 1 dayprelesions vs. 1st recording postlesions) but had no effect onthe variability of directed songs (Fig. 5B; P � 0.05, paired signtest). Moreover, the residual variability present in the undi-rected songs in the first recording postlesions was not signifi-cantly different from the variability present in directed songs 1day prelesions (Fig. 5B, P � 0.05, paired sign test). These dataare thus consistent with a model in which greater variability insyllable structure during undirected song is actively introducedby LMAN (Fig. 1, B and C).

Lesions of LMAN, however, did not abolish all variability insyllable structure. We did not observe a decrease in the mini-mum level of variability in FF (present during directed song)after lesions (Figs. 4B and 5B). This suggests that variability inFF during directed singing, when the patterns of activity inLMAN are stereotyped across repeated renditions of the same

0.030

0.010

0.015

0.020

0.025

0.005

00.030.01 0.020

0.030

0.010

0.015

0.020

0.025

0.005

00.030.01 0.020

A

B

Pre-sham lesionPost-sham lesion

Control

Pre-LMAN lesionPost-LMAN lesion

Directed - Variability in FF (c.v.)

Directed - Variability in FF (c.v.)

Und

irec

ted

- V

aria

bilit

yin

FF (

c.v.

)U

ndir

ecte

d -

Var

iabi

lityi

n FF

(c.

v.)

FIG. 4. Lesions of LMAN reduce the moment-by-moment variability in FFduring undirected song to the level present during directed song. Variability inFF during DS vs. variability in FF during US. Each symbol represents CV ofFF for 1 syllable in both social contexts during 1 recording session. A:context-dependent differences in FF were present in control birds (open blackcircles; n � 9 syllables in 2 birds) and in birds before (open red triangles) andafter sham lesions (filled red triangles; n � 6 syllables in 2 birds). B: data areplotted for 19 syllables for all recordings before (open symbols) and after(filled symbols) lesions of LMAN. Black Xs denote means for populations.Note that after lesions of LMAN, the absolute level of variability in FF duringundirected song was reduced to the level present during directed song (down-ward shift in mean after lesions of LMAN).

1446 M. H. KAO AND M. S. BRAINARD

J Neurophysiol • VOL 96 • SEPTEMBER 2006 • www.jn.org

on August 14, 2006

jn.physiology.orgD

ownloaded from

motif, represents a residual level of variability in syllablestructure in adult birds. This residual variability may reflectmotor constraints in the periphery, such as limits in the efficacyof the vocal musculature. Alternatively, other sources of vari-ability may exist elsewhere in the brain, such as variabilityintrinsic to the motor structures for song production.

Courtship behavior is not affected by lesions of LMAN

The effect of lesions of LMAN on song variability raises thequestion of whether or not other behaviors of males are grosslyaffected by lesions. Are the context-dependent differences insyllable variability eliminated by lesions of LMAN simplybecause the males no longer court females? To examine thisissue directly, we scored video clips of directed and undirectedbouts of song recorded 1 day prelesions and �7 days postle-sions. In contrast to undirected song, directed song is oftenaccompanied by a rhythmic dance that includes approachingthe female, pivoting the body from side to side, and changingthe head position (e.g., bowing and beak wiping), partly todisplay to the female the physical traits that are specificallymale (Williams 2001; Zann 1996).

To assess the courtship behavior of males, human observerswho were familiar with zebra finch behavior but blind to theexperimental manipulation were asked to judge the degree ofarousal and the vigor of the courtship display of each maleusing a scale of 0 (relatively inactive and no dance movements)to 3 (highly aroused courtship display). The scores wereaveraged across song bouts in each social context before andafter lesions to assess the degree to which each male’s court-ship behavior had changed. Figure 6 shows the difference inthe behavior of male zebra finches between social contexts.One day before lesions of LMAN, all birds exhibited robustcourtship displays when a female was present but were lessactive when they were alone (Fig. 6, left; n � 5 birds prelesionsof LMAN and 2 birds presham lesions). One week after lesionsof LMAN, the males continued to vigorously court females

(Fig. 6, right), and their behavior was similar to that of birdswith sham lesions. Thus the elimination of context-dependentdifferences in song variability by lesions of LMAN cannot beattributed simply to a reduction in the motivation of males tocourt females. Moreover, these results suggest that the modu-lation of the courtship display by social context is mediated bystructures other than the AFP.

Other song features modulated by social context and theirdependence on LMAN

This study (Figs. 4 and 5) and other recent work (Kao et al.2005; Olveczky et al. 2005) have shown a robust modulation ofthe variability of syllable structure by social context. In addi-

FIG. 5. Lesions of LMAN eliminate con-text-dependent differences in variability in FF.Bars indicate mean CV of the FF (�SE) forsyllables during US (open bars) and DS(shaded bars). Data are plotted for the samesyllables in the same birds across all recordingsessions. A: summary for all syllables in birdswith sham lesions and in age-matched controlbirds that were recorded weekly. Context-de-pendent differences in song variability per-sisted in control birds and in birds with shamlesions (n � 15 syllables in 4 birds; P � 0.01for each recording, paired sign test). B: sum-mary for all syllables before and after lesionsof LMAN. Bilateral lesions of LMAN abolishthe socially driven differences in song vari-ability by reducing variability in undirectedcondition to the level present in the directedcondition (19 syllables in 5 birds; P � 0.0001for each recording prelesions, paired sign test;no significant differences in song variabilitywere found for any recording postlesions).Variability in FF during US was significantlylower in the 1st recording postlesion comparedwith that 1 day before lesions (P � 0.005,paired sign test). For this and all subsequentfigures, ***P � 0.0001, **P � 0.005, and*P � 0.01.

FIG. 6. Lesions of LMAN do not abolish other courtship behaviors. Sum-mary for all birds of the average level of arousal and/or vigor of courtshipdisplay in each social context 1 day before and 1 wk after lesions of LMAN (E)or sham lesions (■ ). Males continued to vigorously court females after lesionsof LMAN and sham lesions. Lines connect data points for each bird acrosssocial context.

1447LMAN LESIONS PREVENT CHANGES TO SONG VARIABILITY

J Neurophysiol • VOL 96 • SEPTEMBER 2006 • www.jn.org

on August 14, 2006

jn.physiology.orgD

ownloaded from

tion, a previous study by Sossinka and Bohner (1980) exam-ined whether other features of song structure are modulated bysocial context. They reported context-dependent differences infour parameters of song: 1) the number of introductory ele-ments per bout was consistently greater in directed versusundirected song (7/7 birds); 2) the number of motifs per boutwas consistently greater in directed versus undirected songs(6/7 birds); 3) the regularity of the sequence of syllables withina motif appeared to be greater during directed songs thanduring undirected songs (2/3 birds); and 4) directed songs wereusually produced at a faster tempo than undirected songs (6/8birds). It has been suggested that such differences in song mayderive from the differential activation of neurons in the AFPacross social context (Jarvis et al. 1998). We therefore exam-ined each of these song features to determine the degree towhich it is modulated by social context and whether thismodulation is affected by lesions of LMAN.

NUMBER OF INTRODUCTORY ELEMENTS. Consistent withSossinka and Bohner (1980), we found a robust modulation ofintroductory elements by social context. Figure 7B shows thesocial modulation of the number of introductory elements ininterleaved bouts of directed and undirected song produced byone adult male before lesions of LMAN. In this case, theaverage number of introductory elements preceding song wasapproximately twice as large during directed versus undirectedsong. In all birds, before lesions of LMAN, the number ofintroductory elements was significantly greater during directedsong than during undirected song (Fig. 7D, left; P � 0.0005,Mann-Whitney U test for comparisons within 1 recordingsession for each bird).

Lesions of LMAN did not affect the modulation of intro-ductory elements by social context. For each of the birds whosesong was characterized before lesions, we counted introductoryelements in songs produced 1 wk after lesions of LMAN (Fig.7D, right). We found that the number of introductory elementscontinued to be significantly greater during directed songs thanduring undirected songs for all birds (P � 0.0005; Mann-Whitney U test), indicating that signals from LMAN are notrequired for social modulation of this song feature.

NUMBER OF MOTIFS. Figure 7C shows the social modulation ofthe number of motifs in interleaved bouts of directed andundirected song produced by one adult male before lesions ofLMAN, and Fig. 7E summarizes data from each of the birds.Our results were again consistent with those of Sossinka andBohner (1980). We found a consistent increase in the numberof motifs per bout in directed song versus undirected song.This difference was present in all birds and achieved signifi-cance for four of five birds (Fig. 7E, left; P � 0.005, Mann-Whitney U test for comparisons within 1 recording session foreach bird).

Lesions of LMAN did not affect the social modulation ofnumbers of motifs per bout. One week after lesions, directedsongs continued to have more motifs per bout than undirectedsongs in four of five birds (Fig. 7E, right), indicating thatsignals from LMAN also are not required for social modulationof this higher-level song feature.

STEREOTYPY OF SYLLABLE SEQUENCING. Sossinka and Bohner(1980) also reported a tendency for the sequence of syllableswithin a motif to be more stereotyped during directed song than

during undirected song. This phenomenon was less robust thanthe other socially driven changes to song (observed in 2/3birds), and the study did not provide any quantitative analysisof the variability of syllable sequence. From the exampleshown in that study (Fig. 5 of Sossinka and Bohner 1980) andthe corresponding description, it is apparent that the principledifference observed in the stereotypy of syllable sequencingwas greater variability in the transitions that could occur aftera given syllable during undirected song than for the samesyllable during directed song.

To quantify the stereotypy of syllable sequencing, we usedmeasures of sequence linearity and sequence consistency,

FIG. 7. Context-dependent differences in higher-level features of song arenot affected by lesions of LMAN. A: schematic of US and DS bouts. B and C:frequency distributions of number of introductory elements per motif andnumber of motifs per bout during US (open bars) and DS (shaded bars) for 1recording session pre-LMAN lesion (bird pur24w55). Both the number ofintroductory elements and the number of motifs per bout were significantlylower during US than during DS (P � 0.0001, Mann–Whitney U test). D andE: number of introductory elements and number of motifs per bout remainhigher during DS after lesions of LMAN. Each point represents the averagenumber of introductory elements per song bout or average number of motifsper bout for 1 recording session, and lines connect data points for the same birdacross behavioral contexts (n � 5 birds). Bars indicate group means; error barsindicate SE. Data plotted in B and C are indicated by closed squares. F and G:context-dependent differences in higher-level features of song are also presentin older adult birds (same convention as above; n � 6 birds).

1448 M. H. KAO AND M. S. BRAINARD

J Neurophysiol • VOL 96 • SEPTEMBER 2006 • www.jn.org

on August 14, 2006

jn.physiology.orgD

ownloaded from

which have been used successfully for this purpose in previousstudies of zebra finch song (Foster and Bottjer 2001; Olveczkyet al. 2005; Scharff and Nottebohm 1991; Zevin et al. 2004).Sequence linearity quantifies the number of types of transitionsthat can follow a given syllable but is insensitive to howfrequently these different possible transitions occur. For thismeasure, we followed recent studies and included only transi-tions between two syllables of the bird’s motif and did notinclude transitions from syllables to song terminations (Fosterand Bottjer 2001; Olveczky et al. 2005; Zevin et al. 2004).Sequence consistency quantifies how often the observed tran-sitions from a given syllable conform to the most commontransition for that syllable. Both of these measures should besensitive to the types of changes in syllable sequencing re-ported by Sossinka and Bohner (1980). Indeed, for the examplebird shown in that study, this measure of sequence linearityincreases from 0.64 during undirected song to 0.82 duringdirected song.

We calculated these two measures for interleaved directedand undirected songs from control birds and from experimentalbirds before lesions (Table 2, n � 10). In contrast to the reportof Sossinka and Bohner (1980), we found that, for bothmeasures, there was no difference in the stereotypy of syllablesequencing between directed and undirected conditions. Meanlinearity scores were 1.00 � 0.00 (SD) for directed songs and1.00 � 0.00 for undirected songs from the same birds (Table2), and mean consistency scores were 0.91 and 0.92, respec-tively (Table 2; P � 0.73; paired sign test). The differencebetween our findings and those of Sossinka and Bohner islikely to reflect a difference in the degree of stereotypy in theundirected songs of birds in the two studies. In our study, theundirected songs of adult zebra finches were very stereotyped,with high linearity and consistency scores similar to those ofother recent studies. For example, Zevin et al. (2004) reporteda mean linearity score of 0.98 and a mean consistency score of0.96 for the undirected songs of 27 adult zebra finches. Thishigh level of sequence stereotypy presents little opportunity forsongs to become even more stereotyped in the directed condi-tion. In contrast, for the one bird for which data are availablein the study by Sossinka and Bohner (1980), the linearity score

of undirected song can be calculated as 0.64 (insufficient dataare available to calculate sequence consistency). This value islower than that for any of the adult zebra finches to which thismeasure was applied in our study (n � 10) or in the study byZevin et al. (2004; n � 27). Because the birds from the studyof Sossinka and Bohner were a combination of wild caught anddomesticated birds of unknown ages, we speculate that thegreater variability in the sequence of undirected songs in theirstudy reflects either age or strain differences. With respect tothe former possibility, it will likely be informative to test thesocial modulation of song stereotypy in a population of juve-nile birds before song crystallization.

Because we found no evidence for the social modulation ofsequence stereotypy, we did not have a strong expectation thatlesions of LMAN would affect this parameter of song. How-ever, in juvenile birds, lesions and inactivation of LMAN havebeen reported to increase the stereotypy of syllable sequencing(Bottjer et al. 1984; Olveczky et al. 2005; Scharff and Notte-bohm 1991). Hence, we measured sequence linearity andconsistency for directed and undirected songs from birds 1 wkpostlesion and compared these values to those from the samebirds prelesions (Table 2). We found no influence of LMANlesions on these measures for either directed or undirectedsongs (linearity: P � 0.99, 1 sample sign test; consistency: P �0.99, paired sign test). Our finding that LMAN lesions do notincrease sequence stereotypy in adult birds may again resultfrom a ceiling effect: the prelesion songs in our study werealready sufficiently stereotyped to make it difficult to detectany further increases.

Although song linearity and consistency did not differ sig-nificantly across social context, we did notice a differencebetween these conditions in the stereotypy with which birdsterminated their songs. In particular, there were often moretypes of song terminations in directed songs than in undirectedsongs. We quantified this by calculating the percentage ofsyllables within a motif that were observed to precede songterminations. Across the population of normal adult birds, therewere significantly more types of song terminations in directedversus undirected songs (Table 2; 9/10 birds, P � 0.02, pairedsign test). This difference, which reflects greater variability in

TABLE 2. Analysis of song sequence

Bird Condition

Sequence Linearity Sequence ConsistencyPercent syllables

that terminate song

US DS US DS US DS

pur70b30 Age-matched control 1.00 1.00 0.93 0.91 33 37pur90w7 Age-matched control 1.00 1.00 0.98 0.98 57 44pur58g67 Age-matched control 1.00 1.00 0.92 0.89 25 60pur69b29 1 day pre-SHAM lesion 1.00 1.00 0.88 0.85 17 29pur100b28 1 day pre-SHAM lesion 1.00 1.00 0.94 0.94 14 43pur24w55 1 day pre-LMAN lesion 1.00 1.00 0.99 0.97 50 71pur92w89 1 day pre-LMAN lesion 1.00 1.00 0.90 0.92 33 80b19o61 1 day pre-LMAN lesion 1.00 1.00 0.88 0.90 43 50p94g48 1 day pre-LMAN lesion 1.00 1.00 0.85 0.87 33 67p59w36 1 day pre-LMAN lesion 1.00 1.00 0.90 0.86 33 50pur24w55 1 wk post-LMAN lesion 1.00 1.00 0.97 0.97 43 67

pur92w89 1 wk post-LMAN lesion 1.00 1.00 0.90 0.93 60 80b19o61 1 wk post-LMAN lesion 1.00 1.00 0.89 0.91 50 67p94g48 1 wk post-LMAN lesion 1.00 1.00 0.84 0.84 33 67p59w36 1 wk post-LMAN lesion 1.00 1.00 0.88 0.84 17 33

US, undirected song; DS, directed song; LMAN, lateral magnocellular nucleus of the anterior nidopallium.

1449LMAN LESIONS PREVENT CHANGES TO SONG VARIABILITY

J Neurophysiol • VOL 96 • SEPTEMBER 2006 • www.jn.org

on August 14, 2006

jn.physiology.orgD

ownloaded from

directed versus undirected songs, may result from the lesspredictable, and potentially disruptive, auditory and visualstimulation from the female bird that is present in the directedcontext. Indeed, it has been reported previously that unex-pected auditory and visual stimuli can elicit premature songterminations Cynx 1990; Cynx and Von Rad 2001). Regardlessof the source, we found that this difference between directedand undirected song persisted in all birds after lesions ofLMAN (Table 2).

SONG TEMPO. To characterize context-dependent changes tosong tempo, we measured, for each bird, the durations ofcomplete motifs produced in directed and undirected songs.Figure 8A shows data from one bird for which there was arobust modulation of song tempo by social context. In thiscase, song was 2.6% slower in the undirected condition than in

the directed condition (P � 0.0001; Mann-Whitney U test).The degree of tempo modulation by social context variedacross birds and across recording sessions for individuals.Across all control recordings of interleaved directed and undi-rected songs (3 recording sessions each for 5 birds prelesions,2 birds presham lesions, and 3 control birds), song was fasterin the directed context for 24/30 recording sessions. Thisdifference was significant for 20 of these sessions (P � 0.05;Mann-Whitney U test for comparisons within 1 recordingsession for each bird). For 1 of 30 sessions, song was signifi-cantly slower in the directed context (P � 0.02; Mann-WhitneyU test). Overall, there was a highly significant effect of socialcontext on song tempo, with an average slowing of song by 1%in the undirected context (P � 0.0001; paired sign test, n � 30control recordings; P � 0.02, 1 sample sign test for averageeffect per bird, n � 10 birds). This finding is in accordancewith that of Sossinka and Bohner (1980), who reported anaverage slowing of undirected song by �2.7% relative todirected song and significant context-dependent tempo differ-ences in six of eight birds.

To assess whether the modulation of song tempo by socialcontext is dependent on inputs from LMAN, we compared themagnitude of tempo modulation for recordings 1 day beforelesions and 1 wk after lesions (Fig. 8B). In all cases, beforelesions, the tempo was faster in the directed context than theundirected context. This difference was significant for four offive birds considered individually (P � 0.01, Mann-Whitney Utest). In contrast, 1 wk after lesions of LMAN, context-dependent differences in song tempo were no longer consis-tent: four of five birds did not exhibit a significant difference intempo between directed and undirected contexts (P � 0.05,Mann-Whitney U test for recordings in each bird), and in theremaining bird, undirected song was significantly faster thandirected song (P � 0.01, Mann-Whitney U test). Thereforelesions of LMAN abolished the social modulation of songtempo.

Stabilization of syllable structure with age

Previous studies in adult zebra finches have shown that songplasticity in response to manipulation of auditory feedbackdeclines with age (Brainard and Doupe 2001; Lombardino andNottebohm 2000). Because variability in motor output is animportant component of feedback-based motor learning, weasked whether the decline in plasticity is accompanied by anage-dependent decline in song variability. Such a decline issuggested by the gradual reduction in the variability of undi-rected song for control and sham-lesioned birds over the 8- to10-wk period of recordings (Fig. 5A). To explicitly examinethis issue, we recorded directed and undirected songs for 17additional adult zebra finches. For these birds, we indeed foundthat socially driven modulation of song variability is attenuatedwith age. In young adults (�6 mo old), variability in FF wassignificantly greater during undirected song than during di-rected song (Fig. 9A, left; 28/29 syllables in 11 birds; P �0.0001, paired sign test). These differences were significant for19/29 syllables (F-test for equality of variance across condi-tions for each syllable). In contrast, variability in FF did notdiffer across social context in birds �4 yr old (Fig. 9A, right;n � 13 syllables in 6 birds; P � 0.5811, paired sign test). Thelack of context-dependent differences in syllable variability in

US

DS

100 msec

Motif duration

Cou

nt

Motif duration (msec)

4

8

12

0630 650640 660 670

Mean US durationMean DS duration

= 1.026

***

A

B1.02

1.01

1.00

0.99

0.981d pre-lesion

~7d post-lesion

Alone = Directed

Alone > Directed

Alone < Directed

Mea

n un

dire

cted

mot

if d

urat

ion

Mea

n di

rect

ed m

otif

dur

atio

n

FIG. 8. Context-dependent differences in song tempo are dependent oninput from LMAN. A: top: DS is typically produced at a faster tempo than US.Spectrograms of a representative motif during US (top) and DS (bottom) of 1bird are aligned by onset of the 1st syllable. Dotted lines indicate onset timeand offset times of motifs. Bottom: histogram of motif durations during US(open bars, n � 37) and DS (green bars, n � 35). Motif duration wassignificantly shorter during DS (means: 640.6 vs. 657.2 ms; P � 0.0001,Mann-Whitney U test). B: lesions of LMAN eliminate context-dependentdifferences in song tempo. Undirected motifs were significantly slower thandirected motifs in 4 of 5 birds before lesions of LMAN (P � 0.01, Mann-Whitney U test for comparisons within 1 recording session for each bird). Afterlesions of LMAN, song tempo was not significantly different in 4 of 5 birds(P � 0.14, Mann-Whitney U test). In the remaining bird, undirected song wassignificantly faster than directed song (P � 0.04, Mann-Whitney U test). Linesconnect data points from the same bird. Bars indicate group means; error barsindicate SE.

1450 M. H. KAO AND M. S. BRAINARD

J Neurophysiol • VOL 96 • SEPTEMBER 2006 • www.jn.org

on August 14, 2006

jn.physiology.orgD

ownloaded from

older birds seemed to be the result of an age-dependentreduction in the absolute level of variability in FF duringundirected song to the level present during directed song (Fig.9A). That is, variability in FF during undirected song wassignificantly lower in older birds than in young adults (Fig. 9A;P � 0.0015, Mann-Whitney U test), reminiscent of the lesion-induced stabilization of syllable structure (cf. Figs. 5B and 9A).

As with LMAN lesioned birds, the age-dependent loss ofsocially driven modulation of FF did not result from a generaldisruption of courtship behavior. Older adult males exhibitedrobust context-dependent differences in higher-level featuresof song. In six of six birds, the number of introductoryelements was significantly greater during directed song thanduring undirected song (Fig. 7F; P � 0.05, Mann-Whitney Utest for comparisons within 1 recording session for each bird),and in four of six birds, the number of motifs per bout wassignificantly greater during directed song (Fig. 7G; P � 0.05,Mann-Whitney U test). Thus the social modulation of higher-level features of song persisted despite the absence of context-dependent differences in syllable structure, consistent with thehypothesis that different parameters of song are independentlycontrolled.

The stabilization of syllable structure induced by age and bylesions of LMAN raises the question of whether or not themechanism for song stabilization is the same in the two groupsof birds. The observed age-dependent stabilization of syllablestructure may reflect a normal developmental decline in factorsfrom the AFP that promote variability. For example, variabilityin the pattern of AFP activity may decline with age or may notdiffer across social context in older adult birds. Alternatively,signals from the AFP may not change with age. Rather, assynaptic connections in RA are strengthened, the ability ofextrinsic signals to modulate ongoing motor patterns and sub-sequent song output may decline with age (Lombardino andNottebohm 2000).

To examine these possibilities, we compared the singing-related activity in LMAN in birds over a wide range of ages.Figure 9B shows the singing-related activity during one record-ing session for a male bird �5 yr old. Across repeated rendi-tions of the bird’s motif, the pattern of activity was morereproducible during directed song than during undirected song.Consistent with previous reports, we found that variability inthe pattern of multiunit activity in LMAN was significantly

< 6 months old > 4 years old

0.020

0.012

0.016

0.008

0.004

0.024

***

**

** ***

0.020

0.15

0.10

0.00

0.05

0.025

< 6 months old > 4 years old

US DS US DS

A

C

V

aria

bilit

y in

FF

(c.v

.)

Var

iabi

lity

in L

MA

N a

ctiv

ity (

c.v.

)

US DS US DS

100 ms

8000

0Freq

uenc

yM

otif

-rel

ated

activ

ityA

ctiv

ity le

vel/

back

grou

nd

BDSUS

1

32

0

0.4

0.2

Var

iabi

lity

acro

ss tr

ials

(c.

v.)

**

FIG. 9. Age-dependent stabilization of syllable structure despite persistentcontext-dependent differences in LMAN activity. A: context-dependent differ-ences in song variability decline with age. CV of FF is plotted for syllablesrecorded in both behavioral contexts (1–7 syllables/bird); lines connect datapoints for the same syllable. Bars indicate group means. Variability in FF wassignificantly greater during US than during DS in birds �6 mo old (left: 28/29syllables in 11 birds; P � 0.0001, paired sign test) but not in older birds (right:�4 yr old; 13 syllables in 6 birds; P � 0.5811, paired sign test). Variability inFF during US was significantly lower in older adults compared with youngadults (P � 0.0015, Mann-Whitney U test). B: context-dependent differencesin LMAN multiunit activity in an adult male (�5 yr old). Top: singing-relatedactivity in LMAN during 3 renditions of the motif during US (left) and DS(right). Middle: rectified, smooth waveforms of LMAN multiunit activityduring 10 renditions of motif are superimposed for each behavioral condition.Waveforms are normalized by mean spontaneous activity (dotted line). LMANactivity was greater in magnitude and more variable in pattern across repeatedrenditions during US. Bottom: CV of activity level was calculated in 1-ms binsacross repeated undirected (left) and directed (right) motifs (thick black lines).CV across undirected motifs (gray) is replotted for comparison with CV acrossdirected motifs (right). Arrows denote the mean time-averaged CV in eachcondition. C: context-dependent differences in LMAN activity are present inbirds across a range of ages. Time-averaged CV during US and DS are plottedfor each site. Data in B are indicated by closed squares. Variability in LMANactivity was significantly greater during US than during DS in birds of differentages (left: �6 mo old: 10/10 sites in 5 birds; P � 0.002, paired sign test; right:�4 yr old: 9/9 sites in 2 birds; P � 0.004, paired sign test). Althoughcontext-dependent differences were present in older birds, the absolute level ofvariability in LMAN activity declined significantly in both behavioral contextswith age (undirected CV: 0.155 vs. 0.123; P � 0.01, Mann-Whitney U test;directed CV: 0.087 vs. 0.068; P � 0.004, Mann-Whitney U test).

1451LMAN LESIONS PREVENT CHANGES TO SONG VARIABILITY

J Neurophysiol • VOL 96 • SEPTEMBER 2006 • www.jn.org

on August 14, 2006

jn.physiology.orgD

ownloaded from

greater during undirected song than during directed song inyoung adult birds (�6 mo old; Fig. 9C, left; 10/10 sites in 5birds; P � 0.002, paired sign test). Similarly, in older adults(�4 yr old), variability in LMAN activity was also greaterduring undirected song (Fig. 9C, right; 9/9 sites in 2 birds; P �0.004, paired sign test). Moreover, the degree to which vari-ability in LMAN activity differed across social context wassimilar between birds of different ages (mean ratio of theundirected neural CV to the directed neural CV in young andolder adults: 1.78 and 1.81, respectively).

However, despite the presence of context-dependent differ-ences in LMAN activity in birds of different ages, we foundthat the overall level of variability in LMAN activity wassignificantly lower in older adults than in young adults (Fig.9C). This age-dependent decline in the absolute level ofLMAN variability was apparent in both behavioral contexts(Fig. 9C; undirected CV 0.155 vs. 0.123; P � 0.01, Mann-Whitney U test; directed CV: 0.087 vs. 0.068; P � 0.004,Mann-Whitney U test).

Together, these findings suggest that multiple factors con-tribute to the age-dependent stabilization of syllable structure:1) the susceptibility of the motor pathway to sources ofperturbation may decline with age and 2) the factors thatpromote variability in motor activity, such as variability inLMAN activity, may decline with age.

D I S C U S S I O N

Contributions of LMAN to context-dependent regulation ofsong and behavior

In zebra finches, adult song and behavior are stronglymodulated by social context. During directed songs producedin a courtship context, the number of introductory notes, thenumber of song motifs, and song tempo are increased, and thevariability of syllable structure is decreased relative to that ofundirected songs produced by birds in isolation (Dunn andZann 1996; Kao et al. 2005; Sossinka and Bohner 1980).Moreover, directed songs are accompanied by a courtshipdance that is not produced when the birds are alone (Williams2001; Zann 1996). This switching of song and behavioral stateoccurs within the first moments after introduction of a femaleto the presence of a male and reverses equally rapidly after herremoval (Hessler and Doupe 1999b).

Several previous observations have supported the possibilitythat signals from the AFP contribute to this rapid, context-dependent modulation of song and behavior: 1) neural activityin the AFP is greater and more variable under conditions ofundirected song than directed song, indicating that modulationof neural activity in the AFP is correlated with modulation ofbehavior (Hessler and Doupe 1999b; Jarvis et al. 1998; Kao etal. 2005) and 2) active manipulation of signals from the AFP,by microstimulation of LMAN during singing, can alter theacoustic structure of song elements, indicating that signalsfrom the AFP are sufficient to drive changes in song in realtime (Kao et al. 2005). Here, we showed that the nucleusLMAN is necessary for a specific subset of the context-dependent changes in song and behavior. We show that, in

normal adult birds, there is a robust decrease in the variabilityof syllable structure in directed (courtship) song versus undi-rected song and that this difference in song variability iseliminated by lesions of LMAN. In addition, lesions of LMANeliminate the social modulation of song tempo. In contrast,lesions do not prevent context-dependent changes in otherfeatures of song, including the number of introductory ele-ments that precede song and the number of motifs that areproduced during a bout of singing. Similarly, lesions of LMANdo not prevent the postural changes and courtship dance thatnormally accompany directed song.

The selectivity of the effects of LMAN lesions has twomajor implications. First, the effects of lesions on songmodulation are not the consequence of globally disruptingthe male bird’s ability to detect and respond to social cues.When females are presented, the courtship vigor of lesionedmales is as robust as that of control birds. Moreover, lesionsof LMAN did not affect the social modulation of syntacticfeatures of song, such as the number of introductory ele-ments and the number of motifs per bout. These resultsindicate that LMAN is not an obligatory structure forintegrating and representing information about social con-text. Rather, LMAN must be downstream from brain regionsthat register such information.

Second, the selectivity of the effects of lesions of LMANindicates that separate brain regions participate in the modula-tion of different components of song and courtship behavior.This finding is consistent with the anatomy of the song system:neurons in LMAN project directly to the premotor nucleus RA(Fig. 1A), which is thought to provide motor commands thatcontrol the precise structure of individual song elements (Vu etal. 1994; Yu and Margoliash 1996). Moreover, microstimula-tion in LMAN at low intensities modulates syllable structure ofadult birds without affecting syllable sequencing (Kao et al.2005; Vu et al. 1994). In contrast, other aspects of songstructure, such as patterning of introductory elements andmotifs, are thought to be controlled by nuclei earlier in thepremotor pathway for song, such as HVC and its inputs (Fosterand Bottjer 2001; Vu et al. 1994).

Recent measurements of the social modulation of songtempo have shown that expiratory pulses, which contribute tosyllable production, are longer during undirected song thanduring directed song, whereas inspiratory pulses are largelyunchanged across social context (Cooper and Goller 2006).These observations are consistent with the social modulation ofsong tempo reported by Sossinka and Bohner (1980) and withthe observation by Glaze and Troyer (2006) that durations ofsyllables can be regulated independently from durations ofintervening intervals. These data suggest that increases in songduration may arise from an accumulation of increases in thedurations of individual song elements. In this regard, it ispossible that the observed effect of lesions of LMAN oncontext-dependent differences in song tempo reflects the directmodulation by LMAN of pattern generation circuitry in RAthat contributes to the structure of individual syllables. Alter-natively, or in addition, lesions of LMAN might affect songtempo by indirectly influencing activity in other song premotorareas, such as HVC, which are ultimately modulated by activ-ity in LMAN and RA through recurrent connections (Ashmoreet al. 2005; Vu et al. 1998).

1452 M. H. KAO AND M. S. BRAINARD

J Neurophysiol • VOL 96 • SEPTEMBER 2006 • www.jn.org

on August 14, 2006

jn.physiology.orgD

ownloaded from

Regardless of the mechanism, the notion that LMAN canmodulate song tempo is consistent with previous reports thatlesions of LMAN result in a gradual acceleration of the tempoof undirected songs (Brainard and Doupe 2001; Williams andMehta 1999). Although such increases in tempo occur over aperiod of many days to weeks, in contrast to the rapid modu-lation of song tempo by social context, they are nonethelessconsistent with our observation that LMAN can influence songtempo.

Possible functions of different song types (practiceversus performance)

For song learning, variability in motor output may berequired for evaluation mechanisms to differentially rein-force those patterns of motor activity that produce bettersongs (i.e., closer to desired targets) and/or to punish themotor commands that result in worse songs. After song islearned, the ability to modify it likely remains important tocompensate for age-related perturbations in the motor pro-gram and song structure. Such perturbations may arise fromchanges in hormone levels, changes in the peripheral motorstructures for song production (e.g., muscle tone and inner-vation), and/or the birth, death, and incorporation of newneurons in the motor pathway (Alvarez-Buylla 1990; Nor-deen and Nordeen 1988; Scharff 2000).

Given the importance of variability in motor learning, it isquite striking that behavioral variability between directed andundirected conditions is robustly modulated. While the reasonsfor this modulation of song variability remain unclear, oneintriguing possibility is that undirected song is a state of motorpractice in which male birds try out alternative motor com-mands, produce a range of vocalizations, and use auditoryfeedback to optimize and/or maintain the song. In contrast,directed song may reflect a state of motor performance inwhich individuals exploit what they have already learned toselect the patterns of motor activity that produce their “best”current version of song (Jarvis et al. 1998; Kao et al. 2005;Sutton and Barto 1998).

These considerations render it plausible that the AFP is partof nervous system circuitry that specifically contributes to thegeneration of motor variability for the purposes of learning(Bottjer 2004; Brainard 2004; Doya and Sejnowski 2000; Kaoet al. 2005; Olveczky et al. 2005). In principle, the reducedvariability of directed song could reflect an active process inwhich the nervous system tightly controls activity in the motorpathway to produce optimal performance, whereas the greatervariability of undirected song could reflect a noisier defaultmotor performance in the absence of an audience. However,our data show that removal of inputs from the AFP to the motorpathway by lesions of LMAN causes a reduction in the vari-ability in syllable structure present during undirected song tothe level present during directed song (Figs. 4B and 5B). Theseresults support the view that the increased variability present inundirected song is actively introduced by the AFP. They aretherefore consistent with the hypothesis that one of the criticalcontributions of the AFP to vocal plasticity may rely on itsability to introduce variability to the song motor pathway togenerate motor exploration during undirected song. A furtherimplication is that the AFP, and perhaps basal ganglia circuits

in other species, may contribute to an active switching of motoractivity and behavior between states of practice and perfor-mance.

Previous work has shown that lesions of LMAN canprevent experience-dependent changes to adult song (Brai-nard and Doupe 2000; Morrison and Nottebohm 1993;Williams and Mehta 1999), suggesting two possible rolesfor LMAN: 1) LMAN may provide experience-dependentinstructive signals that actively drive changes in the songmotor pathway to produce a good match to the tutor song or2) LMAN may provide permissive factors for song plastic-ity, without which song cannot change, but which by them-selves need not provide the direction for change (for dis-cussion, see Brainard 2004). In principle, variability intro-duced into adult song by the AFP could be construed as apermissive factor for plasticity: the songs of adult birds withLMAN lesions may remain stable because motor explora-tion is effectively prevented. However, the finding thatLMAN can direct moment-by-moment changes in song alsoindicates that signals from the AFP have the potential toactively instruct song plasticity, perhaps by persistentlybiasing motor output toward desired targets (Kao et al.2005; Troyer and Doupe 2000).

Stabilization of song by age and by lesions of the AFP

If variability of syllable structure contributes to motorexploration and song plasticity, one might expect to seechanges in the degree of variability under natural conditionsassociated with song stabilization. One such condition isaging, which in adult zebra finches (beyond the normal ageof crystallization at �90 days) leads to a progressive stabi-lization of song. This stabilization has been documented asa decline in the effects of removing auditory feedback asbirds age (Brainard and Doupe 2001; Lombardino andNottebohm 2000). Here, we show that the age-dependentstabilization of song is indeed accompanied by changes inthe variability of syllable structure. Although there was noapparent change in the variability of syllables produced byolder males during directed singing, the variability of syl-lables produced during undirected singing dropped to thelevel present in directed song (Fig. 9A). Hence aging andLMAN lesions had remarkably similar effects on both thelevel of syllable variability and its modulation by socialcontext.

This similarity between the effects of aging and LMANlesions on syllable variability raises the question of whetherLMAN is functionally inactive in older birds. We found thatthere was indeed a decline in variability of LMAN neuralactivity between younger and older birds (Fig. 9C). However,even in the oldest birds (�4 yr), where social modulation ofsong variability was no longer present, there continued to besocial modulation of variability of neural activity withinLMAN (Fig. 9C). This suggests that enhanced stability of adultsong derives from multiple processes, including 1) reducedvariability in the patterns of neural activity present withinLMAN and 2) reduced sensitivity of the motor pathway to suchactivity. Such reduced sensitivity could derive from a progres-sive weakening of inputs from LMAN to RA and/or a progres-sive strengthening of connections within the motor pathway

1453LMAN LESIONS PREVENT CHANGES TO SONG VARIABILITY

J Neurophysiol • VOL 96 • SEPTEMBER 2006 • www.jn.org

on August 14, 2006

jn.physiology.orgD

ownloaded from

itself, such that it becomes less influenced by inputs fromLMAN.

The finding that the modulation of song variability de-clines with age despite the persistent modulation of LMANactivity suggests that factors in addition to patterned neuralactivity from LMAN are likely to contribute to regulation ofsong variability. These factors may include slower timescale processes, such as neural or trophic influences onsynapse formation and the survival and incorporation of newneurons in the motor pathway (Johnson and Bottjer 1994;Kittelberger and Mooney 1999, 2005; Scharff and Notte-bohm 1991). Indeed, previous studies have shown thatLMAN itself provides trophic support to RA during devel-opment (Akutagawa and Konishi 1994; Johnson and Bottjer1994). Therefore LMAN may well influence both rapid(synaptic) and slower (trophic) time scale processes thatcontribute to the physiology of the motor pathway and thestructure and plasticity of song.

A C K N O W L E D G M E N T S

We thank A. Arteseros, L. Bocksai, and K. McManaway for technicalassistance, E. Ihle and S. C. Woolley for behavioral scoring, and A. J. Doupefor thoughtful comments on this manuscript.

G R A N T S

This work was supported by a Howard Hughes Medical Institute (HHMI)Predoctoral Fellowship to M. H. Kao; the MacArthur Foundation, NationalAlliance for Research on Schizophrenia and Depression, the Sandler FamilySupporting Foundation, the Steven and Michele Kirsch Foundation, andNational Institute of Mental Health Grant MH-55987 to A. J. Doupe; and theMcKnight Foundation, the Klingenstein Fund, the HHMI Biomedical ResearchSupport Program grant, a Searle Scholars Award, a Sloan Research Fellow-ship, and National Institute of Deafness and Communication Disorders GrantR01 DC-006636-01 to M. S. Brainard.

R E F E R E N C E S

Akutagawa E and Konishi M. Two separate areas of the brain differentiallyguide the development of a song control nucleus in the zebra finch. ProcNatl Acad Sci USA 91: 12413–12417, 1994.

Alvarez-Buylla A. Mechanism of neurogenesis in adult avian brain. Experi-entia 46: 948–955, 1990.

Ashmore RC, Wild JM, and Schmidt MF. Brainstem and forebrain contri-butions to the generation of learned motor behaviors for song. J Neurosci 25:8543–8554, 2005.

Bottjer SW. Developmental regulation of basal ganglia circuitry during thesensitive period for vocal learning in songbirds. In: Behavioral Neurobiol-ogy of Birdsong, edited by Zeigler HP and Marler P. New York: New YorkAcademy of Sciences, 2004, p. 395–415.

Bottjer SW, Miesner EA, and Arnold AP. Forebrain lesions disrupt devel-opment but not maintenance of song in passerine birds. Science 224:901–903, 1984.

Bottjer SW, Roselinsky H, and Tran NB. Sex differences in neuropeptidestaining of song-control nuclei in zebra finch brains. Brain Behav Evol 50:284–303, 1997.

Brainard MS. Contributions of the anterior forebrain pathway to vocalplasticity. In: Behavioral Neurobiology of Birdsong, edited by Zeigler HPand Marler P. New York: New York Academy of Sciences, 2004, p.377–394.

Brainard MS and Doupe AJ. Interruption of a basal ganglia-forebrain circuitprevents plasticity of learned vocalizations. Nature 404: 762–766, 2000.

Brainard MS and Doupe AJ. Postlearning consolidation of birdsong: stabi-lizing effects of age and anterior forebrain lesions. J Neurosci 21: 2501–2517, 2001.

Cooper BG and Goller F. Physiological insights into the social-context-dependent changes in the rhythm of the song motor program. J Neurophysiol95: 3798–3809, 2006.

Cynx J. Experimental determination of a unit of song production in the zebrafinch (Taeniopygia guttata). J Comp Psychol 104: 3–10, 1990.

Cynx J and Von Rad U. Immediate and transitory effects of delayed auditoryfeedback on bird song production. Anim Behav 62: 305–312, 2001.

de Cheveigne A and Kawahara H. YIN, a fundamental frequency estimatorfor speech and music. J Acoust Soc Am 111: 1917–1930, 2002.

Doupe AJ and Kuhl PK. Birdsong and human speech: common themes andmechanisms. Annu Rev Neurosci 22: 567–631, 1999.

Doya K and Sejnowski TJ. A computational model of avian song learning. In:The New Cognitive Neurosciences, edited by Gazzaniga MS. Cambridge,MA: MIT Press, 2000, p. 469–482.

Dunn AM and Zann RA. Undirected song in wild finch flocks: context andeffects of mate removal. Ethology 102: 529–539, 1996.

Foster EF and Bottjer SW. Lesions of a telencephalic nucleus in male zebrafinches: influences on vocal behavior in juveniles and adults. J Neurobiol 46:142–165, 2001.

Glaze CM and Troyer TW. Temporal structure in zebra finch song: impli-cations for motor coding. J Neurosci 26: 991–1005, 2006.

Hessler NA and Doupe AJ. Singing-related neural activity in a dorsalforebrain-basal ganglia circuit of adult zebra finches. J Neurosci 19: 10461–10481, 1999a.

Hessler NA and Doupe AJ. Social context modulates singing-related neuralactivity in the songbird forebrain. Nat Neurosci 2: 209–211, 1999b.

Immelmann K. Song development in zebra finch and other Estrildid finches.In: Bird Vocalisations, edited by Hinde RA. London: Cambridge, 1969, p.61–74.

Jarvis ED, Scharff C, Grossman MR, Ramos JA, and Nottebohm F. Forwhom the bird sings: context-dependent gene expression. Neuron 21:775–788, 1998.

Johnson F and Bottjer SW. Afferent influences on cell death and birth duringdevelopment of a cortical nucleus necessary for learned vocal behavior inzebra finches. Development 120: 13–24, 1994.

Kao MH, Doupe AJ, and Brainard MS. Contributions of an avian basalganglia-forebrain circuit to real-time modulation of song. Nature 433:638–643, 2005.

Kittelberger JM and Mooney R. Acute injections of brain-derived neurotro-phic factor in a vocal premotor nucleus reversibly disrupt adult birdsongstability and trigger syllable deletion. J Neurobiol 62: 406–424, 2005.

Kittelberger JM and Mooney R. Lesions of an avian forebrain nucleus thatdisrupt song development alter synaptic connectivity and transmission in thevocal premotor pathway. J Neurosci 19: 9385–9398, 1999.

Leonardo A and Konishi M. Decrystallization of adult birdsong by pertur-bation of auditory feedback. Nature 399: 466–470, 1999.

Lombardino AJ and Nottebohm F. Age at deafening affects the stability oflearned song in adult male zebra finches. J Neurosci 20: 5054–5064, 2000.

Merrill EG and Ainsworth A. Glass-coated platinum-plated tungsten micro-electrodes. Med Biol Eng 10: 662–672, 1972.

Morrison RG and Nottebohm F. Role of a telencephalic nucleus in thedelayed song learning of socially isolated zebra finches. J Neurobiol 24:1045–1064, 1993.

Nordeen KW and Nordeen EJ. Projection neurons within a vocal motorpathway are born during song learning in zebra finches. Nature 334:149–151, 1988.

Nordeen KW and Nordeen EJ. Auditory feedback is necessary for themaintenance of stereotyped song in adult zebra finches. Behav Neural Biol57: 58–66, 1992.

Nordeen KW and Nordeen EJ. Long-term maintenance of song in adultzebra finches is not affected by lesions of a forebrain region involved in songlearning. Behav Neural Biol 59: 79–82, 1993.

Nottebohm F, Stokes TM, and Leonard CM. Central control of song in thecanary, Serinus canarius. J Comp Neurol 165: 457–486, 1976.

Olveczky BP, Andalman AS, and Fee MS. Vocal experimentation in thejuvenile songbird requires a basal ganglia circuit. PLoS Biol 3: e153, 2005.

Perkel DJ. Origin of the anterior forebrain pathway. In: Behavioral Neurobi-ology of Birdsong, edited by Zeigler HP and Marler P. New York: NewYork Academy of Sciences, 2004, p. 736–748.

Reiner A, Perkel DJ, Bruce LL, Butler AB, Csillag A, Kuenzel W, MedinaL, Paxinos G, Shimizu T, Striedter G, Wild JM, Ball GF, Durand S,Gunturkun O, Lee DW, Mello CV, Powers A, White SA, Hough G,Kubikova L, Smulders TV, Wada K, Dugas-Ford J, Husband S,Yamamoto K, Yu J, Siang C, and Jarvis ED. Revised Nomenclature foravian telencephalon and some related brainstem nuclei. J Comp Neurol 473:377–414, 2004.

Scharff C. Chasing fate and function of new neurons in adult brains. CurrOpin Neurobiol 10: 774–783, 2000.

1454 M. H. KAO AND M. S. BRAINARD

J Neurophysiol • VOL 96 • SEPTEMBER 2006 • www.jn.org

on August 14, 2006

jn.physiology.orgD

ownloaded from

Scharff C and Nottebohm F. A comparative study of the behavioral deficitsfollowing lesions of various parts of the zebra finch song system: implica-tions for vocal learning. J Neurosci 11: 2896–2913, 1991.

Sossinka R and Bohner J. Song types in the zebra finch (Poephila guttatacastanotis). Z Tierpsychol 53: 123–132, 1980.

Sutton RS and Barto AG. Reinforcement Learning: An Introduction. Cam-bridge, MA: MIT Press, 1998.

Tchernichovski O, Mitra PP, Lints T, and Nottebohm F. Dynamics of thevocal imitation process: how a zebra finch learns its song. Science 291:2564–2569, 2001.

Troyer TW and Doupe AJ. An associational model of birdsong sensorimotorlearning. II. Temporal hierarchies and the learning of song sequence.J Neurophysiol 84: 1224–1239, 2000.

Vu ET, Mazurek ME, and Kuo YC. Identification of a forebrain motorprogramming network for the learned song of zebra finches. J Neurosci 14:6924–6934, 1994.

Vu ET, Schmidt MF, and Mazurek ME. Interhemispheric coordination ofpremotor activity during singing in adult zebra finches. J Neurosci 18:9088–9098, 1998.

Williams H. Choreography of song, dance and beak movements in the zebrafinch (Taeniopygia guttata). J Exp Biol 204: 3497–3506, 2001.

Williams H and Mehta N. Changes in adult zebra finch song require aforebrain nucleus that is not necessary for song production. J Neurobiol 39:14–28, 1999.

Yu AC and Margoliash D. Temporal hierarchical control of singing in birds.Science 273: 1871–1875, 1996.

Zann RA. The Zebra Finch: A Synthesis of Field and Laboratory Studies.Oxford, UK: Oxford, 1996.

Zevin JD, Seidenberg MS, and Bottjer SW. Limits on reacquisition of songin adult zebra finches exposed to white noise. J Neurosci 24: 5849–5862,2004.

1455LMAN LESIONS PREVENT CHANGES TO SONG VARIABILITY

J Neurophysiol • VOL 96 • SEPTEMBER 2006 • www.jn.org

on August 14, 2006

jn.physiology.orgD

ownloaded from