Talk2NatureParrots / Vocal learning and sociality study
Budgerigar vocal and social records are not event-aligned
Source year 2023 · Note reviewed 2026-10-14
In a 48-bird study, older adults showed reduced vocal diversity over the experiment and older-age flocks had weaker/less dense affiliative ties; measured vocal plasticity and convergence were similar across age classes. The deposit pairs derived call features and social observations at flock/date/block scale, not at synchronized call-event scale.
What we reviewed · Primary methods and dataset-fit review; files remain metadata-only
Reviewed the primary article's animal housing, experimental outline, audio recording and video recording Methods through PMC-indexed full text and a coauthor-hosted article PDF; the official Dryad version history, file inventory and README data dictionary; Dryad reuse guidance; and the official NMSU facility locator. Direct PMC page access encountered CAPTCHA. No Dryad files, raw audio, video, supplement or code were downloaded or executed. The README describes the interobserver workbook, but it was not inspected and agreement was not recalculated.
Not yet independently reviewed by an avian bioacoustics or social-behavior specialist.
Study at a glance
- Study question
- Does adult age relate to contact-call learning and social integration? The study addresses vocal learning and sociality, not call translation.
- Animals and design
- 48 adult male budgerigars formed 12 novel flocks of four birds: six young-adult and six older-adult flocks. The experiment ran from December 2020 through April 2021.
- Acoustic sessions
- Each bird had four 55-minute baseline sessions and 16 sessions after joining a flock. Calls were recorded individually in a sound-isolation chamber; nearby flockmates were visible but not acoustically connected.
- Social observations
- Flock video was collected in two 80-minute sessions per recording block. Early-morning videos were sampled continuously for social interactions during a 40-minute window after a 30-minute settling period.
- Public deposit
- The February 2024 Dryad version is 68.14 MB and includes contact-call feature tables, social-interaction coding, proximity data, an interobserver-assessment workbook, example video and R scripts. The file inventory does not list original call WAV files.
- Rights and reuse
- The article lists CC BY 4.0. Dryad's reuse guide says its datasets use CC0; that dataset policy does not clear the article or separately included software. No deposit files were downloaded or re-distributed.
Reading location: Primary PMC index, Methods 2(a)–(d): Animal housing, Experimental outline, Audio recording and Video recording; Dryad version/file inventory and README; Dryad reuse guide; NMSU building list for city-level study-site locator.
What this does — and doesn’t — establish
The experiment formed 12 novel flocks of four adult males in two age classes. Individual contact calls were collected in separate 55-minute sessions with each focal bird in a sound-isolation chamber, visually but not acoustically connected to flockmates. Social behavior was filmed in separate 80-minute sessions of these newly formed experimental flocks, not home flocks; early-morning video was sampled for affiliative and agonistic interactions after a 30-minute settling period, while proximity was assessed from snapshots. Dryad's 2024 deposit lists derived call features and social tables with overlapping flock/date/block fields but no original call WAV files or shared call/event timestamps. It cannot support event-level sound-to-behavior prediction. The 12-flock design and age treatment constrain generalization. The README documents an interobserver workbook with a subset recoded by a second observer; the workbook and R analysis scripts were not inspected, and no agreement statistic or reproduced result is claimed.
What we can build on
A study can collect sound and behavior in the same experiment without producing synchronized audio–behavior labels. Require a shared clock and source-file linkage between the sound and observed behavior; flock/date overlap alone is not event alignment. Treat this Dryad deposit as a protocol and schema comparator, not the first Talk2Nature context benchmark.
See the related tools and reading paths →
The next reading task
If aggregate vocal-learning/sociality analysis is considered, inspect the interobserver assessment, define the analysis unit, audit exact group/date/block joins and missingness, and obtain independent scientific review before any model. For call-level context prediction, seek rights-cleared synchronized recordings with independent event annotations and access to original audio.
Go to the sources
- Budgerigar aging, vocal learning and social interaction study ↗
Moussaoui et al. (2023), Proceedings of the Royal Society B. Abstract and targeted Methods passages reviewed from the primary PMC index; direct PMC page open encountered CAPTCHA. The article lists CC BY 4.0. The full supplement and raw audio were not reviewed or acquired.
- Budgerigar vocal-learning and social-interaction data ↗
Dryad deposit updated February 5, 2024. Its README describes derived contact-call features and metadata, separate flock-video social-interaction and proximity coding, an interobserver workbook, example interaction video and R scripts. The file inventory lists no original call-audio recordings. No files were downloaded.
- NMSU Animal Care Facility location ↗
New Mexico State University's building list places the Animal Care Facility in Las Cruces, New Mexico. Used only to generalize the study site to a city-level atlas marker.
- Dryad dataset reuse and CC0 policy ↗
Dryad's official reuse guide states that Dryad datasets are published under CC0 and recommends scholarly citation. This is repository-wide policy guidance, not a substitute for reviewing included software or other linked artifacts separately.
Original editorial note by Talk2Nature contributors. AI-assisted, source-grounded synthesis; not a peer review or an independent replication. Linked works retain their own rights.