To ensure the comprehensive nature of the database, we have included research outputs from a variety of sources. This includes: publications from all relevant and available journals and book chapters, including preprints and registered reports. We also included unpublished masters and PhD thesis, when these were available online. We primarily used Google Scholar to search for the relevant literature, but cross-checked with other online sources to ensure our literature search captured everything that can be accessed online (e.g., Web of Science, Scopus). We collected research outputs that were published (or are available online) between 01/01/1950 and 01/02/2024. This large range allowed for a vast amount of literature to be collected. Once all the research output was collected and categorized, we excluded observational studies, and only considered experimental studies for the database.
What kind of information did you include in the database?
The database is divided into two spreadsheets. The first one includes all relevant information on the study itself. Each sub-study (or experiment) within a publication is recorded separately. For each study we include data on: facility name, facility ID, study ID, study alternative ID, title, authors, abstract, publication year, journal/publication source, keywords, data availability, number of participants, number of participants from all facilities, participant IDs, testing conditions, coding method, recording method, cognitive domain, category, subcategory, task name, task specification, apparatus/materials, rewards, testing duration, pretesting conditions, comments. The second spreadsheet records information on the participants in each study. The spreadsheets are linked and draw relevant information from each other. Each participant is inserted separately, and we include information on: facility name, facility ID, study ID, alternative study ID, participant ID, participant name, date of birth, age, study year, sex, origin, rearing, facilities lived in, group, subspecies, task name, task duration, task performance, latency/trial to success, success/fail criteria, comments.
Yes! Once we have completed our database on captive chimpanzees, we would like to extend our database to include studies on wild chimpanzees as well. The next step will be to expand the database to include other primate species, focusing on the ones that have, historically, been tested the most (e.g., bonobos, gorillas and orangutans).
Phase two will build upon the data collected in phase one to generate targeted data on the effect of previous research experience on captive chimpanzee cognition, behaviour and performance in research tasks. To do so, we will use the database to identify groups of chimpanzees, and individuals, with a spectrum of experience in research tasks. This will allow us to form precise predictions according to our documentation of their previous research participation. Each target individual/group will then be presented with the same novel task battery. We will control for other influencing factors as much as possible before testing, and afterwards in the analysis (e.g., groups will be matched by sex, age, rearing, subspecies, as well as enclosure types, exposure to humans, certain personality traits, etc).
We hypothesise that like humans, previous experience will have an effect on chimpanzee behaviour, and their performance in research tasks. If this hypothesis is validated, it would warrant a reappraisal of the current approach which generalises findings from one group of captive chimpanzees to a species-wide level. Findings from each group would have to be interpreted on a case-by-case basis, and it is likely that we will be limited in the extent of the generalisations that can be made from specific groups of captive chimpanzees. The ARI database will provide a tool for future researchers to help them identify the most scientifically appropriate group for their specific questions, and will allow for adjustments to expectations to be made before testing. This will hopefully strengthen the robusticity of insight drawn from future studies that test captive apes.