
Open access
Author
Date
2022Type
- Doctoral Thesis
ETH Bibliography
yes
Altmetrics
Abstract
Children share with juvenile songbirds the ability to learn through vocal imitation, in a process known as vocal learning. Song learning shares many similarities with human speech learning, at behavioral, neuronal, and genetic levels. One of those similarities is the key role of social interactions during vocal learning. Much importance in this process is ascribed to social interactions, and it is critical to know how they influence vocal learning to understand mechanisms underlying vocal.
One of the most used species in research is the zebra finch (Taeniopygia guttata), with the parallels with language learning and speech development between humans and zebra finches, make zebra finches an important model for studying mechanisms that are involved in vocal learning and productions. Interactions are so important that social deprivation causes song and language deficits (in both zebra finches and humans, respectively). The process of song learning in zebra finches has been studied for many years, but social interactions and how actually they impact this process have still many questions unanswered. There is still a need for more and improved of tools and approaches to better study the impact social interactions in song learning.
We devised a new tool to help dissect vocal interactions in animals. The system provides real-time control of vocal interactions between animals housed separately was developed. It consists of a system implemented on a field programmable gate array (FPGA), that allows asserting communication networks between a small group of animals. The system contains noise attenuation algorithms to ensure that mostly the desired vocalizations are transmitted. In experiments with zebra finches, we showed that the system can control vocal communications between individuals creates the possibility to enforce complex eavesdropping scenarios.
The system was incorporated into a setup study song learning in juveniles. This allowed us to perform experiments where juveniles could interact with different tutors and learn songs from them. Juveniles in the setup could interact with two tutors by siting on active perches (starting the interaction channels between the juvenile and each tutor). They were tutored for 15 days, 2 hours of tutoring each morning. Majority of the juveniles, in a condition were the only social interaction they would get was trough the setup, managed to learn syllables from one tutor. On a following up experiment, the same system was used but the social condition was changed. Juveniles and tutors had companions for most of the time, and the time in the setup is a small fraction of the day, during the tutoring window (19 days, 2-3 hours of tutoring per day). In this condition they did not learn syllables from the tutors, in part possibly due to the tutors singing less than it the first experiment.
Finally, we investigated how changes in social conditions and environment could impact the singing of birds. We analyzed song data acquired in previous experiments to investigate signing rate as a non-invasive indicator of the welfare condition of male zebra finches during changes of social and environmental conditions. Looking at the singing rate of male zebra finches when birds were moved from social contact to isolation (new environment or familiar environment), and to full social contact but in a new environment, we observed that change in social condition alone didn’t had impact on singing rate. However, when coupled with a change to a new environment, it negatively impacted the singing rate. we also saw that continued isolation in a new environment arises as a better solution than transient. To avoid cumulative impact in singing rate, transient isolation in the same environment is a better solution. Show more
Permanent link
https://doi.org/10.3929/ethz-b-000550764Publication status
publishedExternal links
Search print copy at ETH Library
Contributors
Examiner: Hahnloser, Richard H.R.
Examiner: Cook, Matthew
Examiner: Derégnaucourt, Sébastien
Publisher
ETH ZurichOrganisational unit
03774 - Hahnloser, Richard H.R. / Hahnloser, Richard H.R.
More
Show all metadata
ETH Bibliography
yes
Altmetrics