Analysing vocal complexity in relation to sociality in orcas of British Columbia: An application of long-term computational passive acoustics
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Orcas are both highly social and highly vocal animals. In coastal waters of the North-Eastern Pacific Ocean, the Northern Resident orca population is well monitored, providing a great opportunity to learn about their social and communicative behaviour. Here, we report a series of acoustic analyses that lead to the empirical assessment of factors that might impact vocal complexity. Automatically processing long-term passive acoustic data, we detected and classified calls to transcribe vocal activity. Detailed post-hoc analyses show that the detection model is imperfect, especially in detecting calls of low energy. Also, diarisation is not possible with this data and transcriptions might gather a mixture of several emitters. Taking these limitations into account, we measured communicative complexity considering the groups’ vocal production as a whole. Acoustic and visual cues also enabled the identification of specific groups with estimated numbers of individuals. Results highlight a positive correlation between vocal and social complexity, which could be due to the mere effect of having more potential emitters. Nonetheless, this brings a first demonstration of the non-trivial link between the number of emitters and complexity in the composition of sequences. We also demonstrate significant impacts of other proximate factors such as behaviour on vocal complexity measurements, and advocate for multi-factor considerations when evaluating communicative complexity. This work demonstrates the pertinence of joint efforts between passive acoustics, visual observations and machine learning to enhance the scale of behavioural studies and assess the validity of evolutionary hypotheses of communication systems.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it