Nonadjacent Dependencies and Syntactic Structure of Chimpanzee Action During a Natural Tool-Use Task

Published: 25 March 2024| Version 1 | DOI: 10.17632/xdtnxzdchj.1
Contributors:
Elliot Howard-Spink,
,
,
,
,

Description

The code and data in this repository supports the manuscript: Howard-Spink et al. (2024) "Nonadjacent Dependencies and Syntactic Structure of Chimpanzee Action During a Natural Tool-Use Task". The data contains over 8000 actions used by eight wild chimpanzees at Bossou, Guinea, when engaging with nuts and stone tools at an Outdoor Laboratory. To learn more about how data was collected and processed, see the methods section of the associated manuscript. To run the code, please download a version of RStudio (https://www.rstudio.com/) (This code was written in RStudio Version 1.2.5001). Total installation time ~ 10 minutes. Codes were written and tested on RStudio (RStudio (v1.2.5001); Running R v.4.2.1.) on operating system macOS Monterey 12.3.1. For further information or questions about presented data or scripts, contact the corresponding author at: elliot.howardspink@outlook.com

Files

Steps to reproduce

The outdoor laboratory (OL), Bossou, is a maintained clearing in the rainforests of Guinea (See Matsuzawa et al, [2011] The Chimpanzees of Bossou and Nimba). At the OL, nuts and stone tools are left out for wild chimpanzees, who visit the site as part of their daily ranging, and engage in natural nut-cracking behaviours. The chimpanzees are recorded engaging in such behaviours, and have been for several decades. This dataset includes descriptions of sequences of actions used by chimpanzees during nut cracking, from all recordings taken in the 2011-2012 field season (December-February). Sequences of actions were coded using an ethogram of 34 actions (each contained in data column 'behavior'), and 6 possible objects (each contained in data column 'Modifier 1'; see supplementary information to main manuscript for full ethogram). Actions were coded both in respect to their function, e.g. 'grasp' and their focal object, e.g. 'nut' to produce a holistic description of the behaviour (in this case, 'grasp nut'). Action coding began whenever chimpanzees began interacting with a nut, nut fragment, or stone tool. Where the onset of the behaviour was observed, sequences began with a 'start' codon. Action coding ceased whenever chimpanzees dropped all relevant objects to engage in another behaviour, or rest for longer than one minute. Where the cessation of behaviour was observed, sequences were terminated with an 'Endbout' codon. Where chimpanzees moved out of sight during the behaviour, sequences were terminated with a 'not visible' codon. These sequences represent incomplete fragments of behaviour. As such, any sequence which did not begin with a 'start' and end with an 'end' codon is incomplete. Sequences are marked as complete or incomplete within the data. Instances of individual object play were included in the data, however social play between more than one individual was not included. Sequence data can be found in multiple formats, each contained in the subfolder 'Data': Event_Codes_Original - This contained the sequence data, separated into fragments. Each csv file is an uninterrupted, observed sequence of tool behaviour collected from an individual. EachNut - This contains the sequences of actions used by individual chimpanzees when cracking open individual nuts. These sequences are taken from 'Event_Codes_Updated'. Concatenated - This contains concatenated sequences (end to end) for each individual. Sequences are concatenated in the order they are observed (See script 1). MI Estimation - MI estimations from sequence data; see associated scripts for methods of MI estimation. Markov contains generated Markov sequences and their MI estimates. For more information on reproducing data, see associated manuscript int he description. All required R scripts are provided.

Institutions

Universite de Geneve, University of Rochester, Chubu Gakuin Daigaku Chubu Gakuin Daigaku Tanki Daigakubu, University of Oxford, California Institute of Technology

Categories

Animal Behavior, Syntax, Animal Tool Use, Animal, Sequential Decision Making, Behavior (Neuroscience), Computational Psychology, Primate, Tool Use, Ethology, Computational Biology, Great Ape, Information, Comparative Psychology, Paleoanthropology, Primatology, Evolutionary Anthropology, Action and Motor Control

Funding

Natural Environment Research Council

NE/L002612/1

Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung

PCEFP1_186832

Ministry of Education, Culture, Sports, Science and Technology, Japan

#12002009

Ministry of Education, Culture, Sports, Science and Technology, Japan

#16002001

Ministry of Education, Culture, Sports, Science and Technology, Japan

#20002001

Ministry of Education, Culture, Sports, Science and Technology, Japan

#24000001

Ministry of Education, Culture, Sports, Science and Technology, Japan

#16H06283

Japan Society for the Promotion of Science (JSPS)

Core-to-core CCSN and U04-PWS

Ministry of Education, Culture, Sports, Science and Technology, Japan

15K00204

Ministry of Education, Culture, Sports, Science and Technology, Japan

19K21824

Japan Society for the Promotion of Science

JP17H06381

Wolfson College, University of Oxford

Boise Trust

Clarendon Fund

Licence