Agent-based modelling in animal research – interview with Iris Boumans and Eddie Bokkers

Congratulations Iris Boumans and Eddie Bokkers to the publication of two papers in Physiology and Behaviour! These papers present work using a relatively new behaviour research approach. What is Agent-Based Modelling and how is it useful in the study of farm animal behaviour?

Agent-based modelling is a computer simulation method in which agents (representing pigs in our case) are individually programmed to have their own goals and can make their own decisions, within the limits of certain rules. Especially interesting in these simulation is the interaction between agents and the environment, which can affect the decisions agents make. In this way behaviour can emerge that was not programmed on forehand. Due to conflicts at a feeding place, for example, pigs can adapt their feeding behaviour and aggressive interactions might occur. Or a stimulus-poor environment can trigger redirected behaviour in an individual, such as tail biting, which can develop into a large scale tail biting outbreak in a group. These kind of simulations provide us more insight in the underlying mechanisms of pig behaviours, the effect of the environment and pig characteristics on this behaviour, and how this affect the performance and welfare of pigs. This can be very useful in the study of farm animal behaviour to understand how housing and management conditions affect animals and to learn how we can use behaviour as indicator for performance and animal welfare. The advantage of using simulation models such as we did is that many variations of situations can be explored without the need of doing animal experiments.

Feeding_motivation_pigs_140106 view- groot4

In your first paper (, you present the development of the model, and in the second paper ( you use it to study feeding patterns, social interactions and growth in pigs. Tell us more about what you actually did and what you found!

Large variation in feeding behaviour can be observed among pigs: some have large meals where others ‘nibble’ more, some eat fast where others eat slow, and some eat more in the morning where others eat more in the afternoon. What is causing this variation, and what can it tell about the performance of the animals? The aim of our studies was to understand this variation in feeding behaviour of pigs and the implications of these various feeding patterns for their productivity and welfare. In the first paper we explain how we developed an agent-based model to understand the mechanisms underlying feeding and social interaction behaviour. We used the model to explore the complex interaction between physiological processes (metabolic, hormonal and growth processes) and the social context (social facilitation, competition and behavioural strategies). We found that high levels of competition can be recognised by changing meal-based patterns (e.g. meal frequency), which can either increase or decrease due to varying behavioural strategies of pigs (approach or avoid). Especially increased meal frequency was associated with increased aggression, although both an increased and decreased meal frequency caused problems, such as a reduced feed intake and reduced growth. The group size at which competition levels reduced productivity and welfare could be predicted from the average daily feeding time of pigs. In the second paper we further explored the variation in feeding, social interaction and growth patterns in pigs. In literature, studies found contradicting results on relations between dominance rank of pigs, feeding types (e.g. meal eaters vs. nibblers) and growth patterns. These contradictions can be explained by variation in pig characteristics, such as growth potential and coping style. Model results showed that variation in characteristics among group-housed pigs can benefit animal welfare by reducing aggression at group level, but can also make some pigs more susceptible to competition, such as low-ranking pigs and pigs with a passive coping style. Pigs that are socially constrained in a competitive situation can be identified by a changed 24 hour feeding pattern with reduced feed intake in the morning.

This is quite a new approach, and so knowledge is still rather limited. Is it possible to say something already about the potential practical applications of this work? What is needed before this research can be put to practical use?

Our work shows that feeding patterns can tell us much more about pig welfare and productivity then was thought so far. With currently ongoing technological developments in pig farming, such as automatic feeders and individual recognition, pig behaviours can be monitored individually and real-time on farms. Recognising behavioural signals, preferably at an early stage, can help to prevent problems and improve productivity, health and welfare in pigs. At present, however, interpretation of these behaviours and early detection of deviating behavioural patterns is still challenging due to the large variation within and among individuals. With our model we gained a deeper understanding of variation in feeding behaviour and found feeding patterns that can serve as indicators for reduced growth, aggression and social constraints. We learned, for example, that it is important for these indicators to analyse patterns on individual level and within 24 hours, instead of at group level and daily averages as is currently common. Furthermore, we should not only look at one component of a behaviour, such as feed intake, but also at other components of feeding behaviour, such as feeding time, meal frequency and feeding rate. Changes in feeding behaviour have been associated with several welfare issues, such as health problems and tail biting behaviour. For future studies, it would be interesting to study the relation between feeding patterns.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s