Stronger amygdala-control network connectivity predicts impulsive choices in older adolescents

A new study published in Human Brain Mapping sheds light on the brain mechanisms that influence how adolescents weigh immediate versus delayed rewards. The research focused on dynamic brain connectivity and found that stronger functional links between the left amygdala and the brain’s cognitive control network predicted a greater preference for immediate rewards—but only in older adolescents faced with large monetary decisions.

This pattern did not appear in younger adolescents or young adults. The findings provide evidence that unique patterns of brain function during late adolescence may drive heightened reward-seeking tendencies, especially when the stakes are high.

Delay discounting refers to how much a person devalues a reward depending on how long they have to wait for it. For example, someone might prefer receiving $100 today over $150 in six months. This behavior is often used as a measure of impulsivity or self-control. The lower a person’s ability to wait, the steeper their delay discounting curve. Researchers use this measure to examine decision-making, especially in contexts like addiction, risk-taking, and adolescence, a period known for rapid changes in the brain’s reward and control systems.

Adolescents are known to be more sensitive to rewards compared to children and adults. Brain imaging studies have shown that regions like the amygdala and ventral striatum become more responsive to potential rewards during adolescence. At the same time, the regions responsible for executive control—mainly within the frontoparietal network—are still maturing.

This imbalance, where reward systems are fully online but self-regulatory systems are still developing, has been proposed as one explanation for the increase in risky or impulsive behavior during this period of life. The current study set out to examine how the communication between these two systems—the emotional salience system (amygdala) and cognitive control network—might help predict individual differences in delay discounting behavior across development.

Free daily newsletter

“Adolescence is a time of major changes at both behavioral and brain levels,” said study author Gaelle Doucet, director of the Brain Architecture, Imaging and Cognition Lab at Boys Town National Research Hospital.

“Increased risk taking and reward sensitivity are among these changes; however, their neural origins remain unclear. In this context, we wanted to investigate the neural activity of specific brain regions involved in emotional regulation and executive function, that typically play a role in these cognitive functions and how their involvement may change throughout adolescence and early adulthood. Such findings could help us understand better why some teenagers take more risk than others.”

The research team used data from 448 participants aged 10 to 21 who were part of the Human Connectome Project – Development. Participants were divided into three age groups: younger adolescents (10–13 years), older adolescents (14–17 years), and young adults (18–21 years).

Each participant completed a delay discounting task, in which they chose between a smaller amount of money available immediately and a larger amount available after a delay. This was done across two reward sizes—$200 and $40,000—with six different time delays ranging from one month to ten years.

The choices allowed researchers to calculate each individual’s “area under the curve” (AUC), a common metric in delay discounting research. A lower AUC indicates steeper discounting, meaning a stronger preference for immediate rewards.

To examine brain function, resting-state fMRI data were collected and analyzed using a sliding window approach. This method breaks up the brain scan into short time segments and calculates functional connectivity for each one. The researchers focused on connectivity between the amygdala and 52 regions of the cognitive control network.

Importantly, the researchers used dynamic functional connectivity (dFC), a method that captures how connections between brain regions fluctuate over time, rather than assuming they are stable throughout a scan. This allowed them to ask not just whether two regions are connected, but how that connection varies—and whether that variability matters for behavior.

Overall, participants in all age groups showed a stronger preference for immediate rewards in the $200 condition compared to the $40,000 condition. But age made a difference. Younger adolescents were significantly more likely than older adolescents and young adults to favor immediate rewards in the high-value condition. This supports previous work suggesting that sensitivity to reward peaks early in adolescence and gradually declines.

However, when it came to predicting individual differences in delay discounting behavior from brain connectivity, a more specific pattern emerged. Only among older adolescents (aged 14–17) did dynamic functional connectivity between the left amygdala and the cognitive control network significantly predict behavior on the task. Specifically, those with stronger connectivity were more likely to choose the immediate reward in the $40,000 condition.

No such association was found for the right amygdala. Nor did the connectivity metrics predict behavior in younger adolescents or young adults. This suggests that during a particular developmental window—late adolescence—the dynamic interaction between emotion-processing and cognitive control systems may play a larger role in shaping reward-based decision-making, especially for large potential gains.

“Our findings confirmed a difference in sensitivity to large monetary reward between early and late adolescence, with younger adolescents preferring smaller but immediate reward compared to older adolescents or adults,” Doucet told PsyPost. “We further revealed that older adolescents aged 14 to 17 years old showed a unique relationship between reward sensitivity and brain activity in specific regions related to emotion regulation and executive function. This relationship was not present in younger adolescents or young adults. This suggests that reward preference in older adolescents may be linked to a particular configuration of these brain regions which seem to go away as adolescents grow up.”

The finding that only the left amygdala was involved aligns with prior research indicating that it may play a stronger role than the right amygdala in tasks that involve executive function, decision-making, and reading emotional information. The left amygdala also tends to mature earlier and may be more sensitive to cognitive influences than its right-hemisphere counterpart.

Additionally, the most predictive connections in the study involved the medial and lateral prefrontal cortex, which are central components of the brain’s cognitive control system and have been repeatedly implicated in goal-directed behavior, self-control, and subjective valuation. A smaller number of connections also involved the inferior parietal lobule, which has been linked to numerical reasoning and mental computation—skills that may come into play during decisions involving tradeoffs over time.

One possible explanation for the age-specific pattern lies in the development of dopamine systems. During adolescence, dopamine receptor availability and connectivity within the prefrontal cortex change rapidly. Some researchers have proposed that these changes temporarily lead to a mismatch between elevated dopamine levels and still-developing executive control circuits, making this a particularly sensitive period for reward-seeking and impulsivity.

Older adolescents may be at a “tipping point,” where their prefrontal regions are becoming more connected and dynamic, but not yet fully mature. This could create a situation where brain systems involved in evaluating rewards and exerting control are especially active—but not always in ways that support long-term decision-making.

“It is important to keep in mind that even typical adolescence is associated with higher risk taking which is thought to be a normal part of development,” Doucet said.

As with all research, there are some limitations to consider. The study focused solely on the amygdala and did not include other important reward-related regions like the nucleus accumbens. Although the amygdala plays a key role in emotional salience and reward evaluation, a broader network of subcortical structures is also likely involved in delay discounting behavior.

Future studies could examine other types of rewards as well as real-world risk-taking behaviors. “While we used a test that estimated sensitivity to monetary reward, it will be important to test responses to other types of reward (social acceptance, substance use),” Doucet explained. “One of the long-term goals is to understand the mechanisms behind risk taking and higher reward sensitivity in adolescents that may lead to more dangerous behaviors. This knowledge could lead to a better insight on why some young people will turn to alcohol and drug use.”

The study, “Dynamic Functional Connectivity Between Amygdala and Cognitive Control Network Predicts Delay Discounting in Older Adolescents,” was authored by Attakias T. Mertens, Callum Goldsmith, Derek J. Pavelka, Jacob J. Oleson, and Gaelle E. Doucet.