Forsøksdyr: Neurobiology of social learning and cognition in rodents

Godkjenningsdato 13.10.2020

Godkjenningsperiode 01.11.2020-31.10.2024

1. Purpose or the application
To lay the foundation for studying the cellular mechanisms of observational learning and social processing in rodents. To achieve this goal, we shall first establish behavioral paradigms for sensoriomotor observational learning as well as paradigms for studying neuronal population dynamics during social interactions in rats.

2. Harm to animals
These studies require implanting both chronic stimulating and recording electrodes as well as adeno-associated virus (AAV) and optic fibers. Some animals will undergo one surgery, while others will have a maximum of two. The rats may experience moderate short-term pain in the first days following drive implantation, a surgical procedure requiring general anesthesia and analgesia.

3. Expected benefits
The field of neuroscience lacks a mechanistic account of how we learn to imitate the actions of the others. The value in these experiments is that they will establish much-needed behavioral paradigms for studying the neurophysiology of basic forms of observational learning and social comprehension, using rodents as models. Currently, social learning in rodents is studied primarily through aversive vicarious learning paradigms, which typically involve a "demonstrator" animal receiving a footshock and an obsesrver animal who learns e.g. that a tone or light predicts a fearful shock by observing the demonstrator. However, the rodent work lacks a robust form of non-aversive sensorimotor learning, which focuses on the neural mechanisms underlying behavioral imitation rather than learning by fear. This project will, we believe, produce at least one paradigm for studying the neural systems and cellular mechanisms underlying imitative social learning.

4. Number of animals requested & type:
160 rats.

5. How to uphold the 3 R's:

Replacement: For direct investigation of the link between cellular activity patterns in freely-behaving, socially interacting subjects, there is presently no substitute for the use of animals. However, scientific advance can be improved enormously, and the need for animal experiments can be reduced, by close interaction between experimental and computational approaches to neuroscience. Although computational modelling represents only a minor part of the activity in our research group, our work is guided by computational models and we will have close interactions with computational neuroscientists over the course of the project, both within and outside the Centre.

Reduction: A maximal amount of information is sampled from each animal, and measures are taken to maximize the chances of getting useful data from every single experimental subject (e.g. sterile surgery conditions, healthy animals). We have also recently implemented next-generation recording probes in our experiments, which allow us to obtain statistically meaningful datasets in fewer animals than in previous years.

Refinement: We aim to ensure that conditions in the animal house are the best possible. Implanted animals are housed in large cages (50×50×50 cm) and are handled frequently. We will also make use of enriched environments, which consist of large (2m x 1m) cages full of objects to which the animals are exposed regularly before the start of experiments.