Experimental validation of a minimal model for birdsong production

SANTIAGO BOARI; YONATAN SANZ PERL; ANA AMADOR; GABRIEL MINDLIN

Huerta Grande, Córdoba

Congreso; XXIX CONGRESO ANUAL DE LA SOCIEDAD ARGENTINA DE INVESTIGACION EN NEUROCIENCIAS; 2014

Sociedad Argentina de Investigación en Neurociencias

Songbirds are known as an animal model for studying sensorimotor vocal learning and control of a complex behavior such as singing. The song is generated from neural instructions that command the respiratory system and the syrinx (vocal organ of birds), with striking similarities to what is observed in humans. Interestingly, there are neurons in a songbird´s brain telencephalic nucleus HVC that respond selectively to the auditory presentation of the bird´s own song (BOS). In this work we used a synthetic song that has been achieved by modeling the functioning principles of the biomechanical periphery. Recent experimental evidence (Amador et. al, 2013) showed that this low-dimensional model of the syrinx presented biological relevance, by comparing firing properties in the evoked response of HVC neurons to BOS and SYN renditions. We have since then developed a computational program that automatically generates a synthetic song using as input only a recorded bird´s song. In this work we present the stereotactic device built in our laboratory to perform extracellular recordings in anesthetized zebra finches (Taeniopygia Guttata), with which we measured the evoked neuronal response to several auditory stimuli: Bird´s own Song (BOS), Synthetic Song (SYN), Reverse song (REV) and Conspecific song (CON). The experimental evidence gathered by this protocol could serve as a basis to assess the state of validation of the low-dimensional model of the songbird´s vocal organ.