Lilac-crowned Amazon (Amazona finschi) / Field bioacoustic context study

Lilac-crowned Amazon calls vary with context; meaning remains a question

Publication year 2016 · Note reviewed 2026-10-13

Montes-Medina, Salinas-Melgoza and Renton analyzed 75 hours of non-continuous field recordings from 61 free-living Lilac-crowned Amazons. They report 8,622 notes classified into 152 note-types; after excluding single-occurrence types, 8,571 notes formed 101 note-types emitted at least twice. The researchers associated calls with nine behavioral contexts, and common note-types occurred in several contexts with different proportions. Context was associated with acoustic features and note-type frequency. Separately, a nested permuted discriminant analysis distinguished 28 note-types while accounting for repeated notes from individuals (46.9% observed classification versus 9.6% expected under the paper's permutation null, P=0.001). That is note-type discrimination, not prediction of behavior from sound and not evidence that calls have human-language meanings.

What we reviewed · Full-text primary-source review

Reviewed the full publisher article for abstract, field methods, context definitions, acoustic measurements, statistical analyses, results, discussion, data availability, ethics and rights; reviewed the additional-file inventory. The research viewer failed or timed out on selected binary supplements. No article assets, audio, data or code were downloaded, opened, executed or redistributed; no independent reanalysis or specialist review was performed.

This Talk2Nature synthesis has not been independently reviewed by an avian-bioacoustics specialist.

Study at a glance

Study group: 61 free-living Lilac-crowned Amazons (Amazona finschi), identified at nest sites or inferred as potentially distinct individuals from recording day and distance for some off-site contexts.
Recordings: 75 hours of non-continuous recordings made at about 30 m with directional microphones and solid-state recorders; 44.1 or 48 kHz WAV, with 48 kHz material resampled to 44.1 kHz.
Observed contexts: Nesting, threat, alarm, foraging, perched, take-off, flight, landing and food soliciting. The study is observational; it did not test playback responses.
Acoustic units: Researchers classified 8,622 notes into 152 note-types. Acoustic-parameter analyses used 58 types meeting frequency and recording-quality criteria; the note-type pDFA used 28 types emitted by more than one individual.
Supplements and rights: Publisher lists a 116 kB XLSX of statistical-analysis data, context descriptions, a note-frequency table and sound examples. Article text is CC BY 4.0; the publisher says data made available with the article are CC0 unless otherwise stated. Supplement binaries and any file-specific exceptions were not checked.

Reading location: Frontiers in Zoology 13:40 (2016): Abstract; Methods—Vocal recording, Vocal analyses, Statistical analyses; Results—design characteristics and classification by note-type; Discussion; Availability of data and materials; Ethics approval and consent; Additional files; Rights and permissions.

What this does — and doesn’t — establish

This is a single-species field study with non-continuous recordings and researcher-assigned behavioral contexts, not a validated context classifier or modern ML benchmark. The article notes low sample sizes in some contexts and says playback evidence is needed to establish signal function. GLMM analyses excluded food soliciting; the pDFA classifies note types, not behavioral contexts, and its permutation comparison is not an untouched held-out-animal test. Some off-nest individual identities were inferred from day and distance. The paper does not report quantified independent context-label agreement. Supplement descriptions conflict with several linked file extensions; no data or example audio was opened, and the listed files do not describe a full 75-hour archive. The acoustic setup and wild species do not validate ordinary phone capture or household-parrot behavior.

What we can build on

Keep the observation, acoustic event, behavioral context and interpretation as separate records. A call can be associated with multiple contexts, and a note-type classification result is not a behavior prediction or translation. Use this paper to design labels and confound checks, not as evidence that Talk2Nature can infer a call's meaning.

The next reading task

Ask an independent avian-bioacoustics reviewer to assess the context definitions, individual assignment and nested pDFA. If a concrete reproduction need emerges, first retrieve the exact XLSX and supplement inventory through an authorized route, inspect columns and provenance, resolve the publisher's file-type mismatches, and verify the CC0 notice for each selected data file. Do not train or treat the sample sounds as a benchmark.

Go to the sources

Contextual flexibility in the vocal repertoire of an Amazon parrot ↗
Montes-Medina, Salinas-Melgoza and Renton. Frontiers in Zoology 13:40 (2016). Full-text field study of Lilac-crowned Amazon note types across nine behavioral contexts; context association and note-type discrimination are distinct outcomes.

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.