Stages of an R&D project

Before embarking on a research project to understand a given situation, you need to know whether scientific work already exists on the subject. If so, you need to know what precisely has been studied, how, and what the results are. This first step enables you to decide whether or not a new research project is relevant and necessary.

This bibliographical research stage can also be used to acquire additional and complementary knowledge about the species, the individual or the environment.

Once the problem has been definitively defined and the hypotheses proposed, you can set about constructing the protocol and experimental design. In order to understand a behavior, its variability, causes and consequences, or even its functions, it is essential to determine which traits and parameters are to be studied, how they are to be observed, and the study framework. The construction of the protocol is a crucial and sensitive stage, requiring a great deal of rigor and anticipation in order to avoid disturbing either the animals or the environment. This stage generally requires considerable time for elaboration and calibration, to ensure that data is collected reliably and accurately.

An ethogram (defined as a list of behaviors from the species' behavioral repertoire) is then created, tailored to the problem and to various characteristics such as the species studied, the age of the animals, the observation context, the behaviors of interest, etc. The calibration phase then allows us to test the experimental protocol, including the ethogram, under real experimental conditions, to check the relevance of the data collected, the feasibility of the experiment or its adjustment. The calibration phase is then used to test the experimental protocol, including the ethogram, under real experimental conditions, in order to verify the relevance of the data to be collected, the feasibility of the experiment or its adjustment.

Now that all the experimental design has been created and validated, it's time for field observations.

Is it worth mentioning that data collection must be carried out by ethologists and/or anyone with a thorough understanding of the protocol and observation of the species concerned?

In fact, although broad behavioral categories can be found, each species (not to say each animal) is different, so their behaviors can be complex to apprehend and understand. If behaviours are poorly observed, noted or transcribed, then data analysis will be at best distorted, if not impossible.

As a result, it is imperative that observers are perfectly trained. Conducting experiments and observing data can be like a game of hide-and-seek, where observers have to blend into the environment so as not to disturb the animals as they go about their business. Observations can take place in any environment, under any weather and climatic conditions, over varying lengths of time.

Another factor to bear in mind is the representativeness, repeatability and reproducibility of the data. Indeed, enough animals must be observed to be able to draw conclusions at the level of the entire group of animals of interest to the study. Secondly, observations must be repeated several times to ensure that the results obtained are representative (and do not reflect an isolated situation, for example). Finally, depending on the scale of the study, experiments must be reproducible, i.e. they must be repeatable on other sites (such as different farms, for example). This is particularly important when, for example, you want to generalize the use of a protocol. In all cases, these elements are fundamental to ensuring the robustness of the data and, consequently, the quality of the results.

Once you've completed the fieldwork phase, you'll enter the lesser-known or little-known territory of data processing and analysis.

While some data can be processed directly, others require a few additional steps, as in the case of biochemical samples taken in the field (e.g. hormone samples), which then need to be assayed in the laboratory.  

Other types of data are not directly usable as they stand. This is, for example, the classic case of filmed experiments. Data must be extracted from these films by coding them using appropriate software. Once these steps have been completed, you're at the gateway to the world of statistics!

Statistics are often perceived as a black box by many people, and can be frightening for some, but a real playground for others. In any case, statistics are the sine qua non for obtaining study results. Statistical tests, models and graphs are then produced using appropriate software.

At last, you've reached the stage you've been waiting for: theresults !

This almost final stage should not be underestimated. Although the results may be very clear and in line with the original hypotheses, they can often be more complex than they appear, and may even point to paths other than the original ones. Interpretation of the results must therefore be carried out with caution and hindsight, to avoid drawing hasty or erroneous conclusions. Knowledge of the scientific literature on the subject can be used to support or compare the results obtained.

Once the research is completed, the results are generally disseminated through the writing and publication of scientific articles. They may also be presented at national or international conferences.

Bankiva will support you throughout the design and implementation stages of your R&D work in ethology and animal welfare.

Bankiva is recognized by the French Ministry of Research as an approved body for the Research Tax Credit (CIR).