A Pig Model for Behavioral Neuroscience


The lab develops the domestic pig as a large animal model for cognitive and behavioral neuroscience to further our research in the mPFC-HC system and to advance the comparative approaches available in the field. While pigs have been studied extensively in biomedical research, and are often the preferred clinical model, less work has been done in the area of cognitive neuroscience. Notably, the pig is a rapidly growing large animal model for the human brain, in part, due to their large gyrencephalic cerebrum (cortical folding).

As you’ve probably heard, pigs are very smart. In the lab, pigs easily learn to perform complex cognitive tasks with demands identical to humans and other primates. From our point of view, it is equally important that pigs can also perform the spatial tasks that are most commonly used in rodent models. This is because much of our knowledge about the neurobiological mechanisms of learning and memory have been obtained in rodents performing spatial tasks. However, questions still exist as to whether the neurobiological mechanisms of memory found in rodents extend to other species and other tasks. The pig can help bridge the divide by studying the neurobiological mechanisms in an individual subject that performs both the human and rodent tasks following neurobiological manipulations (e.g., DREADDs) or during electrophysiological recordings (e.g., depth electrodes).

The Porcine Neuroscience Facility (PNF)

To address these issues we established the Porcine Neuroscience Facility at FIU that incorporates hoof stock housing facilities, a state-of-the-art surgical suite, custom behavioral and physiological rigs (US Patent 10,251,722, 2018), and appropriate histology set ups. Below is a saggital view of one of our pig brains after a transcardiac perfusion and fixation (Draper et al., 2018, SfN).

Touchscreen Cognition in Pigs

To test pigs on human memory tasks, we built a large touchscreen apparatus that runs the exact same tasks used in human and primates using slightly modified pre-exisiting PsychoPy (Python) or PsychToolBox (MATLAB) scripts. In the example below, the pig is performing a conditional associative memory task designed by our collaborators (MaDLab) designed to compare activations in striatal and mPFC-HC memory systems during learning and memory (Hamm et al., 2019, Cell Reports).

Spatial Memory in Pigs using a Large Automated T-Maze

A large amount of our knowledge about the neurobiological underpinnings of learning and memory have been obtained in rodents performing spatial tasks, which are very often variations of a T-maze. The T-maze has provided enormous insight into memory. For example paradigms testing place vs. response learning provided clear evidence that there are multiple memory systems that follow different rules in the striatal and HC systems. Importantly, electrophysiological studies have used T-maze tasks to show that there are representations of time and space in the HC. We aim to test the pig brain, particularly the mPFC-HC system, for representations of time and space, which required us to build a large automated T-maze.

In the video below, you can see one of our pigs in the custom automated T-maze built for these studies. In the task the pig is performing a classic delayed spatial alternation task (DSA). In DSA, the correct choice (right or left) should be the opposite of the last choice. For example, if the last choice was left, the correct choice is to go right. Performing this task well requires memory. The task allows us to change the difficulty by increasing the length of time the pig must remember their last choice. While, rats can be successful at DSA up to about 1 minute, the pig can be successful after more than 4 minutes delay. Additionally, the path of the pigs (from detailed video tracking) suggests that the pig makes a decision at the T-junction early in training, as presumed throughout the rodent literature. However, as training progresses, the decision is evident as soon as pig leaves the start area, suggesting the decision is made well before the T-junture. Although this may be specific to smart pigs, we suspect it’s true of rodents as well. Regardless, these pigs show remarkable DSA performance, making them an ideal model for probing the neural mechanisms of memory in a large gyrencephalic brain.

As such, we’re currently using wireless recordings and custom depth electrodes in this maze to address questions of spatial and temporal neural representations in the pig mPFC-HC system.

Return to Research Page