Vocal Experimentation in the JuvenileSongbird Requires a Basal Ganglia CircuitBence P. O¨lveczky1,2, Aaron S. Andalman1, Michale S. Fee1*1 McGovern Institute for Brain Research, Department of Brain and Cognitive Sciences, Massachusetts Institute of Technology, Cambridge, Massachusetts, United States ofAmerica, 2 Harvard Society of Fellows, Harvard University, Cambridge, Massachusetts, United States of AmericaSongbirds learn their songs by trial-and-error experimentation, producing highly variable vocal output as juveniles. Bycomparing their own sounds to the song of a tutor, young songbirds gradually converge to a stable song that can be aremarkably good copy of the tutor song. Here we show that vocal variability in the learning songbird is induced by abasal-ganglia-related circuit, the output of which projects to the motor pathway via the lateral magnocellular nucleusof the nidopallium (LMAN). We found that pharmacological inactivation of LMAN dramatically reduced acoustic andsequence variability in the songs of juvenile zebra finches, doing so in a rapid and reversible manner. In addition,recordings from LMAN neurons projecting to the motor pathway revealed highly variable spiking activity across songrenditions, showing that LMAN may act as a source of variability. Lastly, pharmacological blockade of synaptic inputsfrom LMAN to its target premotor area also reduced song variability. Our results establish that, in the juvenilesongbird, the exploratory motor behavior required to learn a complex motor sequence is dependent on a dedicatedneural circuit homologous to cortico-basal ganglia circuits in mammals.Citation: O¨lveczky BP, Andalman AS, Fee MS (2005) Vocal experimentation in the juvenile songbird requires a basal ganglia circuit. PLoS Biol 3(5): e153.IntroductionThe acquisition of complex motor sequences, such asswinging a golf club or playing the piano, can be thought of asreinforcement learning. This learning process requires theexploration of a range of motor actions and the concomitantevaluation of the resulting performance, reinforcing motorprograms that lead to improved outcomes [1]. Similarly,juvenile songbirds explore a large range of vocalizations bycontinuously varying their song [2], utilizing auditory feed-back to improve their performance [3]. Thus, song learningencompasses the two ingredients of reinforcement learning:exploratory motor behavior, and performance evaluation.In the songbird, two main neural pathways are involved insong production and song learning (Figure 1A). The ‘‘ motorpathway’’ controls the vocal motor program through thehierarchical organization of several premotor nuclei [4]. Akey nucleus in the motor pathway is the robust nucleus of thearcopallium (RA), which projects to brainstem nuclei con-trolling the vocal and respiratory muscles [5]. During singing,RA neurons in adult birds generate a highly stereotypedsequence of bursts [6,7], which appear to be driven byprecisely timed inputs from a higher premotor vocal area,nucleus HVC [8]. RA also receives input from the ‘‘ anteriorforebrain pathway’’ (AFP), a circuit homologous to the basalganglia thalamo-cortical loops [9,10] that may be involved incontrolling motor behavior and stereotypy in mammals [11].Lesions of the AFP in juvenile zebra finches have devastatingeffects on song development, whereas the same manipula-tions in adults have few short-term consequences for songproduction [12,13].While the critical importance of the AFP for song learninghas been established, its specific role remains unknown [14]. Ithas been proposed that the AFP may be involved incomparing the auditory feedback of the bird’s vocal outputwith a stored auditory template of the desired song—anevaluation process that could provide a corrective signal tothe motor pathway needed for learning [15]. However, recentresults showing that the firing patterns of neurons in thelateral magnocellular nucleus of the nidopallium (LMAN) ofadult birds are insensitive to distorted auditory feedback havecalled this idea into question [16,17]. Here we test thealternative hypothesis that, in juvenile songbirds, LMAN isinvolved in generating vocal variability [18]—the otherimportant ingredient of reinforcement learning.ResultsOur approach was to transiently inactivate LMAN injuvenile zebra finches (n = 7 birds, see Materials andMethods), and observe whether and how their songs wereaffected. Birds were briefly head-restrained, and injections ofa sodium channel blocker, tetrodotoxin (TTX, 30 nl, 50 lM),were made in LMAN in both hemispheres, inactivating thenucleus (see Figures S1 and S2). After injections, birds werereturned to a sound-isolated chamber, where they typicallybegan to sing after 0.5–1.5 h. In all birds probed, LMANinactivation resulted in an immediate loss of acousticvariability across song renditions. The effect was particularlydramatic in birds at an early stage of song development(approximately 55 d post hatch [dph]) because these birdsReceived February 4, 2005; Accepted March 1, 2005; Published March 29, 2005DOI: 10.1371/journal.pbio.0030153Copyright: Ó 2005 O¨lveczky et al. This is an open-access article distributedunder the terms of the Creative Commons Attribution License, which permitsunrestricted use, distribution, and reproduction in any medium, provided theoriginal work is properly cited.Abbreviations: AFP, anterior forebrain pathway; AMPA, a-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid; AP5, 2-amino-5-phosphonovalerate; dph, dayspost hatch; LMAN, lateral magnoc ellular nucleus of the nidopallium; MMAN, medialmagnocellular nucleus of the nidopallium; NMDA, N-methyl-D-aspartate; RA,robust nucleus of the arcopallium; TTX, tetrodotoxinAcademic Editor: Wolfram Schultz, University of Cambridge, United Kingdom*To whom correspondence should be addressed. E-mail: [email protected] Biology | www.plosbiology.org May 2005 | Volume 3 | Issue 5 | e1530902Open access, freely available onlinePLoSBIOLOGYnormally exhibit greater song variability (Figures 1B, 1C, andS3; Audios S1–S4).To quantify song variability, experiments were carried outin slightly older birds with less sequence and acousticvariability (n = 6 birds; age range, 59–72 dph) (Figure 2).This allowed us to reliably identify song syllables, the basicacoustic units of zebra finch song, across song renditions(Figure 2A). The variability score (V)—a measure reflectingthe acoustic variability of a syllable across renditions