‘Gambling rats’ shed light on how traumatic brain injury affects decision-making

“Each of these experiments used a behavior called the Rodent Gambling Task, which distinguishes between optimal, suboptimal, and risky decision-making,” said lead author Cole Vonder Haar, assistant professor in the Department of Neuroscience at Ohio State. “We found that rather than purely increasing ‘riskiness,’ brain injury reduced sensitivity or understanding of the consequences of decisions.

The Rodent Gambling Task is based off a classic human neuropsychological assessment, the Iowa Gambling Task. In the Iowa Gambling Task, participants choose from four decks of cards to earn money and learn over time that two decks are “risky” (that is, high rewards, but large losses) and that two decks are “safe” and better overall ( that is, low rewards, but small losses).

While rats won’t work for money, they enjoy sugar pellets. Researchers set up a computer-controlled chamber to allow rats to choose among four ports, each of which had a probability of reward (from 1 to 4 sugar pellets) and punishment (5 to 40 seconds time out). By only allowing 30 minutes per day to earn pellets, rats should maximize pellets and minimize time outs.

Like the Iowa Gambling Task, some options are riskier than others. Specifically, there is one optimal choice, one suboptimal choice, and two risky choices (many pellets, big time outs). Rats were tested daily on this for many weeks, eventually generating 660,000 interactions discrete events across almost 160 rats (which was pared down to 109 for the current publication).

Vonder Haar began this research at West Virginia University during the global COVID-19 pandemic.

“This project was initiated during lab closures due to COVID-19. We had students doing lab work for credit, but no ability to get into lab. So, we decided we could pool all of the data we had collected to that point and ask broader questions that go beyond just one experiment,” Vonder Haar said.

Based on these experiments, researchers identified two key findings that may have implications for humans who have suffered TBI. First is the finding that, brain injury may not increase “riskiness,” but rather result in insensitivity to outcomes. These findings suggest that merely providing more salient information to patients may help guide and improve decision-making.

Second is the finding that, not all rats responded to TBI in the same way, which highlights this issue in the human population. Some individuals may suffer a single concussion and have long-term consequences, while others may be fine.

“While this is inherently difficult because we still do not understand why this occurs, having a rat model which mimics the human condition allows us to probe biological variables that may confer vulnerability or resilience to brain injury, and treatments targeting these, which may confer vulnerability or resilience,” Vonder Haar said.

More than 2.8 million people suffer a brain injury every year in the United States, which is 0.85% of the population annually, Vonder Haar said.

“We found that TBI strongly affected a subgroup of rats. Not only does this map onto the human condition, where some individuals suffer greater consequences, but it allows us to begin asking ‘why’ and determine the biological variables that may confer vulnerability or resilience to brain injury,” Vonder Haar said.

Funding for this project was provided by the National Institute of Neurological Disorders and Stroke; National Institute of General Medical Sciences; West Virginia University and Ohio State University.

# # #

Media Contact: Eileen Scahill, Wexner Medical Center Media Relations, 614-293-3737, Eileen.Scahill@osumc.edu

Leave a Comment