Exploring automated and manual approaches to describing chimpanzee behaviour : pose estimation and algae fishing tool use variation

Tool use can highlight a species’ ability to problem solve, plan, and their understanding of cause-and-effect relationships, all of which are considered cognitively challenging. Of non-human animals, chimpanzee tool use remains one of the most extensively described. When using tools for extractive foraging, chimpanzees flexibly employ large repertoires in response to various aspects of the problem-space: For example, specifics of the task, tool properties, and individual characteristics. The ability to respond flexibly by adjusting tool use behaviour (technique) suggests that there is a benefit to doing so. However, the real-world impact of behavioural flexibility on tool using, for example the yield retrieved, is not often measured. In this thesis, I use camera trap video data to explore a recently described tool use behaviour: Algae fishing by Moyen-Bafing chimpanzees. Camera traps allow the study of multiple, unhabituated individuals but creates large video data archives from which behaviours of interest must be extracted. To streamline identification of videos of interest, I developed a pose estimation model (‘DeepWild’) that automatically tracks wild chimpanzees and bonobos. I then apply manual coding methods to describe behavioural variation in algae fishing using a systematic multi-level framework. I show a chimpanzee’s tool use technique varies dependent on the pool, individual identity and tool length. Finally, I explore how technique choice impacts the yield chimpanzees’ retrieve, and as a result on an individual’s access to nutrition. Alongside an open-source pose estimation model, I provide a holistic investigation of a chimpanzee tool use behaviour, providing evidence for chimpanzees’ ability to respond flexibly to a dynamic problem space to increaseaccess to a high-value food item. In doing so I contribute to our wider understanding of non-human tool use and suggest that chimpanzee algae fishing may represent a model task for investigations of behavioural variation and its underlying mechanisms.

Authors

Publication Details

Journal
University of St Andrews
Published
2026-03-27
DOI
https://doi.org/10.17630/sta/1576
Primary Topic
Primate Behavior and Ecology
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Exploring automated and manual approaches to describing chimpanzee behaviour : pose estimation and algae fishing tool use variation

Charlotte Wiltshire
University of St Andrews
Primate Behavior and Ecology
article

Exploring automated and manual approaches to describing chimpanzee behaviour : pose estimation and algae fishing tool use variation

Charlotte Wiltshire
article en

Abstract

Tool use can highlight a species’ ability to problem solve, plan, and their understanding of cause-and-effect relationships, all of which are considered cognitively challenging. Of non-human animals, chimpanzee tool use remains one of the most extensively described. When using tools for extractive foraging, chimpanzees flexibly employ large repertoires in response to various aspects of the problem-space: For example, specifics of the task, tool properties, and individual characteristics. The ability to respond flexibly by adjusting tool use behaviour (technique) suggests that there is a benefit to doing so. However, the real-world impact of behavioural flexibility on tool using, for example the yield retrieved, is not often measured. In this thesis, I use camera trap video data to explore a recently described tool use behaviour: Algae fishing by Moyen-Bafing chimpanzees. Camera traps allow the study of multiple, unhabituated individuals but creates large video data archives from which behaviours of interest must be extracted. To streamline identification of videos of interest, I developed a pose estimation model (‘DeepWild’) that automatically tracks wild chimpanzees and bonobos. I then apply manual coding methods to describe behavioural variation in algae fishing using a systematic multi-level framework. I show a chimpanzee’s tool use technique varies dependent on the pool, individual identity and tool length. Finally, I explore how technique choice impacts the yield chimpanzees’ retrieve, and as a result on an individual’s access to nutrition. Alongside an open-source pose estimation model, I provide a holistic investigation of a chimpanzee tool use behaviour, providing evidence for chimpanzees’ ability to respond flexibly to a dynamic problem space to increaseaccess to a high-value food item. In doing so I contribute to our wider understanding of non-human tool use and suggest that chimpanzee algae fishing may represent a model task for investigations of behavioural variation and its underlying mechanisms.

University of St Andrews
Life below water
Openalex Percentile: Top 7%
Primate Behavior and Ecology
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.