Hemispheric dominance in HVC is experience-dependent in juvenile male zebra finches

Abstract

Juvenile male zebra finches (Taeniopygia guttata) must be exposed to an adult tutor during a sensitive period to develop normal adult song. The pre-motor nucleus HVC (acronym used as a proper name), plays a critical role in song learning and production (cf. Broca's area in humans). In the human brain, left-side hemispheric dominance in some language regions is positively correlated with proficiency in linguistic skills. However, it is unclear whether this pattern depends upon language learning, develops with normal maturation of the brain, or is the result of pre-existing functional asymmetries. In juvenile zebra finches, even though both left and right HVC contribute to song production, baseline molecular activity in HVC is left-dominant. To test if HVC exhibits hemispheric dominance prior to song learning, we raised juvenile males in isolation from adult song and measured neuronal activity in the left and right HVC upon first exposure to an auditory stimulus. Activity in the HVC was measured using the immediate early gene (IEG) zenk (acronym for zif-268, egr-1, NGFI-a, and krox-24) as a marker for neuronal activity. We found that neuronal activity in the HVC of juvenile male zebra finches is not lateralized when raised in the absence of adult song, while normally-reared juvenile birds are left-dominant. These findings show that there is no pre-existing asymmetry in the HVC prior to song exposure, suggesting that lateralization of the song system depends on learning through early exposure to adult song and subsequent song-imitation practice.

Keywords: Broca’s area; Language learning; Lateralization; Song learning; Zenk.

Figure 1

Parallels between human and songbird vocal production and auditory perception pathways. (a) Schematic of the vocal perception pathway in songbirds. The primary auditory pathway (dark gray arrows) ascends through a midbrain nucleus (MLd) and a thalamic nucleus (Ov) to the primary auditory region Field L2. L2 relays information to secondary auditory regions L1 and L3, which innervate the NCM and CMM, the higher-order processing regions. These areas provide input to the premotor control center HVC. Modified, with permission, from Chirathivat, et al.. (b) Circuit diagrams showing analogous regions in the human and songbird auditory perception pathway. The human auditory pathway also includes a midbrain (Icc) and a thalamic (MGv) nucleus, and the primary auditory cortex (A1) projects to higher-order auditory processing (Wernicke’s) and motor control (Broca’s) areas. Modified, with permission, from Chirathivat, et al., with additional sources: Bolhuis, et al., Jarvis. (c) Schematic of the vocal production system in songbirds. The primary motor pathway (black arrows) descends from the HVC, to the RA, directly to motor neuron nuclei that control syringeal and respiratory muscles. The anterior forebrain pathway (light gray arrows) is a motor control loop that includes cortical nuclei (HVC and LMAN), basal ganglia (Area X), and a thalamic nucleus (DLM). Modified, with permission, from Brainard and Doupe. (d) Circuit diagrams (round arrowheads indicate inhibitory synapses) showing analogous regions in the human and songbird vocal production systems. The LMAN performs an analogous function to Broca’s area in humans, while the RA performs a function similar to the laryngeal motor cortex (LMC). The HVC is functionally analogous to Broca’s area, as it receives auditory input and projects to the primary motor pathway and the anterior forebrain loop. However, comparisons of gene expression profiles and cell-type markers indicate that the HVC may be anatomically homologous to the LMC^,–. Modified, with permission, from Brainard and Doupe. Abbreviations: MLd dorsal part of the lateral mesencephalic nucleus, Ov nucleus ovoidalis of the thalamus, NCM caudomedial nidopallium, CMM caudomedial mesopallium, Icc inferior colliculus, MGv ventral medial geniculate nucleus, A1 primary auditory cortex, DLM medial part of the dorsolateral anterior thalamus, RA robust nucleus of the arcopallium, LMAN lateral magnocellular nucleus of the anterior nidopallium, LMC laryngeal motor cortex.

Figure 2

Auditory stimuli for playback experiments. (a) Experimental timeline with mean age (dph) at separation from father, isolation, and playback experiment. Playback experiments were performed within the sensitive learning period for normally-reared zebra finches, and consisted of 30 min of playback (silence, white noise, or conspecific song), followed by 30 min of silence before perfusion. (b) An overlay of the amplitude plots of the song stimulus (black) and white noise control stimulus (gray), showing matching amplitude and temporal envelope of both stimuli. Spectrograms of the song (c) and noise stimulus (d). Note the differences in spectro-temporal characteristics.

Figure 3

HVC activity is bilateral in song-isolated juvenile male zebra finches. (a) Boxplot showing interquartile range (box) and median (black line) of lateralization scores for song-isolated juvenile (57 dph) zebra finches that were exposed to conspecific song (n = 7), rhythmic white noise (n = 7), or silence (n = 6). Lateralization scores were calculated from the number of Zenk-immunopositive neurons per square millimeter in each hemisphere ([L − R]/[L + R]). Lateralization scores were not different from zero for any stimulus group (one-sample t-tests, p > 0.05), indicating that spontaneous HVC activity is bilateral in juveniles raised without an adult song tutor. (b,c) Neuronal activity (mean number of Zenk-immunopositive neurons per mm²) in the left (black) and right (gray) HVC. Error bars represent SEM. Song-isolated juvenile male zebra finches (b) show no difference in activity between the left and right HVC (F_{(1, 17)} = 1.228, p = 0.283) regardless of the stimulus presented (F_(2,17)= 0.758, p = 0.410). For comparison, normally-reared juvenile male (56 dph) zebra finches (c) show higher activity in the left HVC than the right HVC (n = 6 silence, n = 7 novel, n = 6 tutor) independent of stimulus exposure (reproduced, with permission, from Moorman et al.). (d) Neuronal activity in the left and right NCM. Error bars represent SEM. Song-isolated juvenile male zebra finches show a significant difference in activity between stimuli in the left (F_(2,15) = 4.989, p = 0.023; song > silence, p = 0.022), but not in the right NCM. (* indicates p < 0.05).