Type 2 diabetes alters brain circuits involved in reward processing, study finds

A new study published in The Journal of Neuroscience has found that type 2 diabetes can alter how the brain processes spatial and reward-related information. Using a rodent model of diabetes, researchers showed that neural activity in a brain region called the anterior cingulate cortex shifted away from representing reward locations and instead became more focused on anticipating them. The findings raise questions about how metabolic disorders may impact cognition and motivation.

Type 2 diabetes is a chronic metabolic disorder that disrupts blood sugar regulation and is linked to widespread health problems, including increased risk for dementia and depression. Previous research has shown that people with diabetes often experience subtle cognitive difficulties, including problems with working memory and attention. However, scientists still have a limited understanding of how the disease disrupts specific brain circuits that support these functions.

“Type 2 diabetes is the number one modifiable risk factor for Alzheimer’s disease and the condition also has cognitive impacts on its own that are not well understood,” said study author James M. Hyman, an associate professor of psychology at University of Nevada Las Vegas and director of the HivE Lab.

To investigate this, the researchers turned to a well-established animal model of diabetes. They used a chemical compound called streptozotocin to induce chronic hyperglycemia in rats, mimicking the prolonged high blood sugar levels seen in human type 2 diabetes. They then trained the rats to perform a spatial working memory task known as delayed alternation. In this task, rats learned to alternate between left and right turns in a T-shaped maze to receive a reward—a drop of sugar-free chocolate milk—after a short delay.

During task performance, the researchers recorded the activity of neurons in two brain regions: the anterior cingulate cortex, which is part of the prefrontal cortex and plays a central role in goal-directed behavior and reward processing, and the hippocampus, which is critical for memory and spatial navigation. Tiny electrodes implanted in the rats’ brains allowed the researchers to measure the firing patterns of individual neurons and analyze how these patterns changed in response to different parts of the maze and different stages of the task.

Free daily newsletter

The researchers found that, on the surface, diabetic and control rats performed similarly on the task. Both groups completed the same number of trials, had similar accuracy on short-delay trials, and moved through the maze at comparable speeds.

However, subtle behavioral differences emerged when the researchers looked more closely. Diabetic rats spent less time at the reward location than healthy rats. While control animals often paused after receiving the reward, a pattern known as the “post-reinforcement pause,” diabetic rats quickly moved on, suggesting a reduced sensitivity or response to receiving a reward.

When the researchers examined neural activity in the anterior cingulate cortex, they observed a number of differences between diabetic and control rats. Although the overall firing rates of individual neurons were similar between the two groups, diabetic rats showed higher spatial information content in their neural activity. In other words, their neurons were more precisely tuned to specific locations in the maze. Notably, many more neurons in the diabetic rats behaved like “place cells,” which fire when an animal is in a specific location.

However, these place cells were not evenly distributed across the maze. In diabetic rats, they were heavily concentrated in areas leading up to the reward zone rather than at or after the reward location itself. This suggests that the diabetic rats’ brains were focusing more on anticipating the reward rather than responding to its receipt. In contrast, control rats had neurons that were more evenly spread across the maze and especially prominent at the reward location, reflecting both reward anticipation and receipt.

“We did not expect to find changes to reward processing,” Hyman told PsyPost. “We thought there would be differences related to memory processes only, but interestingly we found this key circuit between the hippocampus and anterior cingulate cortex (which we thought was primarily a memory circuit), is actually strongly involved in reward activity. This is also a key circuit that is altered early in the progression from healthy aging to mild cognitive impairment (the first step on the road to dementia).”

At the population level, the researchers found further evidence of altered processing in diabetic rats. In control animals, groups of neurons in the anterior cingulate cortex showed distinct activity patterns that clearly differentiated between left and right turns in the maze, especially at the reward location. This pattern was weaker in diabetic rats, indicating that their neural ensembles did not encode reward-related spatial information as strongly.

The study also looked at how neural activity was coordinated with rhythmic brain waves known as theta oscillations, which are involved in memory and navigation. In control rats, neurons that were synchronized with theta rhythms—particularly those coming from the hippocampus—were especially important for coding reward information. In diabetic rats, this reward-related coding by theta-synchronized neurons was muted, even though the number of synchronized cells was similar across both groups. This suggests that while the structure of neural communication remained intact, the information being transmitted was altered.

Importantly, the researchers found that diabetic rats still retained information about the expected value of rewards. When analyzing firing patterns that predicted task events regardless of maze location, they found similar activity between groups. However, the key difference was that reward location and reward expectation, which were tightly linked in control rats, became dissociated in diabetic animals. This separation could explain why diabetic rats failed to linger at the reward site even though they anticipated receiving a reward.

“Type 2 diabetes leads to changes in how rewards are processed in our brains,” Hyman explained. “While these changes don’t impair cognitive performance, they do show that the brain is functioning differently.”

“This has two main impacts: 1) Decreased reward responses with type 2 diabetes could help explain why it is so difficult for patients to stick to lifestyle changes, such as proper diet and exercise. Their brains just don’t respond to normal rewards like they should. 2) Alzheimer’s has a decades-long prodromal phase where the effects are hidden by the brain’s ability to reroute processes and solve problems differently, known as cognitive or neural compensation. This masks underlying pathology and is currently not detectable.”

“If we can detect when patients are unknowingly using cognitive compensation, it might help us identify people in their 40s and 50s who are on their way to getting Alzheimer’s disease in their 70s. Hopefully, we could intervene and prevent the disease from ever appearing.”

The study does have some limitations. The research was conducted in rats, and while these models are useful for studying disease mechanisms, results do not always translate directly to humans. Additionally, the behavioral task focused on short-delay trials, so it remains unclear whether the observed effects would be more pronounced under more demanding memory conditions.

“This data all comes from rodent models, so all necessary caveats are in play,” Hyman noted. “Also, these animals only had one symptom of type 3 diabetes, chronic hyperglycemia. This is usually accompanied by obesity and other metabolic problems. For this study, we wanted to isolate hyperglycemia.”

The researchers point to several possible mechanisms behind the altered brain activity in diabetic rats. One possibility is the destruction of glucose-monitoring neurons in the anterior cingulate cortex by streptozotocin, which may disrupt how this region processes reward. Another possibility involves changes in a brain metabolite called myo-inositol, which is elevated in diabetes and has been linked to altered brain connectivity and cognitive decline.

Future research could explore how these changes develop over time and whether similar patterns are present in human patients. The findings also suggest new questions about how diabetes treatments—including lifestyle changes and medication—might help preserve or restore healthy brain function. As diabetes rates continue to rise globally, understanding how the condition impacts the brain will become increasingly important for developing effective prevention and intervention strategies.

“We hope to identify the hidden signatures of cognitive compensation so that we can know better what to look for in patients,” Hyman explained. “Since once Alzheimer’s disease is diagnosed, there is no cure, the hope is that early intervention might be able to halt the disease.”

“The big problem seems to be from untreated diabetes, so it is important to have annual blood tests to detect diabetes,” he added. “And if diagnosed, it is important to monitor your blood sugar levels and aim to avoid large fluctuations.”

The study, “ACC reward location information is carried by hippocampal theta synchrony and suppressed in a Type 2 Diabetes model,” was authored by Guncha Bhasin, Emmanuel Flores, Lauren A. Crew, Ryan A. Wirt, Andrew A. Ortiz, Jefferson W. Kinney, and James M. Hyman.