How does Early Life Stress Affect Children’s Ability to Learn About Rewards?


We know (and it’s been discussed on this blog!) that people who have experienced early life stress, such as child abuse, neglect, or extreme deprivation (such as institutional care during infancy) are more likely to develop behavioral and mental health problems. For example, they are more likely to be socially isolated, commit crimes, and develop psychological disorders such as anxiety and depression. But what explains these links? These problems often don’t emerge right away, but instead become evident in later childhood or adolescence, or even adulthood. This pattern of “sleeper effects” suggests that early life stress might disrupt aspects of brain development that support key emotional and cognitive processes. These processes would normally act as building blocks to set people on a path towards positive social relationships and mental health.

But what aspects of development are affected by stress?

Previous research indicates that early life stress affect children’s ability to control or regulate their emotions—for example, children who have experienced a lot of stress seem to have more difficulty containing negative emotions such as anger, which can lead to acting out and alienating  peers; or anxiety, which can contribute to uncontrollable worry about bad things that might happen in the future.

But emotion regulation might not be the whole story. New research suggests that general learning mechanisms, especially the ability to update one’s knowledge when circumstances change, may also be affected by early stress exposure.

The ability to learn and update associations between one’s own actions and the outcomes that result from them—what psychologists call instrumental learning—is fundamental to our abilities to function socially in the world. A very simple example of instrumental learning would be learning that when I ring the doorbell, someone comes to the door. But ringing a doorbell doesn’t ALWAYS result in someone coming to the door (i.e., maybe no one is home). So links between our actions and outcomes depend on the context.

For another example, say I’m telling my friend about the last marathon I ran, giving her a mile-by-mile recap. My friend might be really interested at first, so I’d form a positive association between my chosen conversation topic and her enthusiasm. But eventually she might start to get bored (I can be pretty long-winded when talking about running). Hopefully I’ll notice this, my association between me talking and her reaction will change, and I’ll wrap up my recap. But if I repeatedly fail to pick up on signals that my conversation partner is becoming bored with what I’m talking about, she might start taking more rain checks on our coffee dates. The point is, the circumstances around us, including other people’s reactions to our behavior, are continually changing, and it’s good for us to be able to recognize these changes and adjust our behavior accordingly. If not, we’ll have trouble developing healthy social relationships!

A few recently published studies suggest that stress-exposed children and teenagers have trouble with exactly this type of learning. In one study that my colleagues and I conducted, teenage participants—half of whom had been physically abused by their caregivers in early childhood—viewed pictures of everyday objects (like a belt or boot). Each time they saw a picture (see picture below for example of the task), they had the option to either press a button or do nothing. If they pressed the button, they would either win points or lose points, and this outcome depended on which picture was on the screen—some pictures led to a reward and others to a loss. If they didn’t press the button, nothing happened. Halfway through the task, we switched things up so that some of the pictures that had led to a reward now led to a loss and vice versa (akin to when my friend becomes bored with what I’m talking about). Now participants needed to change their responses if they wanted to continue earning points. It turns out that teens who had been physically abused had much more trouble changing their responses than their peers who had not been abused. If these learning difficulties carry over into everyday social situations, we can see why people exposed to early life stress end up with more social and mental health problems.

Diagram of a task that children are asked to do so that scientists can study how children learn about rewards.

Diagram of a task that children are asked to do so that scientists can study how children learn about rewards.

While teens completed this task, we were also measuring which areas of their brains were active. When abused teens saw pictures that led to reward, regions of the brain that help us learn associations between our actions and outcomes were less active. Interestingly, researchers have found similar patterns of reduced brain activity when reward is at stake in adults who have experienced a range of different early life stressors, and in psychological disorders such as depression.

Another recent study provides further evidence that reward learning is disrupted in people who experienced maltreatment. Here, researchers examined performance on a similar learning task in abused and non-abused teenagers. In this task, participants repeatedly chose between two images, only one of which would usually lead to reward. Teenagers who had been abused by their caregivers in early childhood learned to associate pictures with reward more slowly than teens who had not been abused. The researchers also used mathematical modeling to show that abused teens seemed to act as if rewards happened randomly rather than being linked to specific pictures, and did not use the feedback they received during the task to guide their decisions as much as non-abused teens (for instance, if you have usually been getting a reward when you choose picture X, you would be more likely to choose picture X later if you were using feedback to inform your choices).

Put together, these research findings suggest that early adversity could affect how people learn to obtain rewards in their lives.

Early adversity might have these effects because stress disrupts the development of key brain regions that help us associate specific events or actions with positive or negative outcomes. Children exposed to early stress might therefore have trouble learning how to achieve positive outcomes in their lives, like doing well in school or making friends.

As a result, they might encounter fewer positive experiences even after the adversity has ended, and end up with higher risks for mental health problems like depression. Because these learning difficulties don’t go away when the stress ends, this pathway helps explain the sleeper effects of early life stress that I mentioned earlier.

To recap, adolescents with a history of early life stress seem to have difficulty with at least two aspects of learning: changing what they do when circumstances change, and learning which events or actions lead to rewards.

If early life stress disrupts something as fundamental as basic learning, is there any hope for these kids? Yes! In fact, these studies suggest new ways we could think about designing interventions to help kids who’ve experienced early adversity. For example, carefully designed computer games could help teach children to pay attention to rewards in their environment and identify the events or actions that predict these rewards. Other interventions could target children’s abilities to deal with changing circumstances. In fact, programs like Big Brothers and Big Sisters, which appear to improve outcomes in at-risk kids, might already work this way by exposing children to new environments and new people. Boosting children’s learning abilities in these ways might be an effective way to improve social and mental health outcomes. Although we should also strive to prevent children from being exposed to deprivation, abuse, or neglect in the first place, new research on how stress exposure affects learning can lead to more ways to help kids who have already experienced early adversity.

About the Author: Dr. Madeline Harms

labphoto (1).jpeg

I completed my PhD training in Developmental Psychology at the University of Minnesota’s Institute of Child Development and am currently a post-doctoral fellow at the University of Wisconsin-Madison. Starting in September 2018, I will be a visiting assistant professor at Gustavus Adolphus College in St. Peter, Minnesota. I investigate how reward learning changes across development, and the influences of social experience and stress on the ways in which children, adolescents, and adults perceive and respond to potential risks and rewards in their environment. I use a combination of computer-based tasks, questionnaires, interviews, and brain imaging methods to explore these questions.