A three-minute smartphone game can detect a subtle cognitive mechanism behind depression

A new study published in the Proceedings of the National Academy of Sciences provides evidence that people with depression experience a measurable shift in their expectations for pleasure, making it harder for them to enjoy rewarding activities. Scientists found that a simple smartphone game can detect this shift in just three minutes, offering a potential new way to quickly assess the severity of the illness.

Depression is a widespread condition affecting millions of people worldwide. Despite its prevalence, doctors mostly rely on symptom checklists rather than specific measurements of underlying cognitive processes to diagnose the condition. A major feature of depression is anhedonia, a symptom where individuals lose the ability to experience pleasure from normally enjoyable activities.

Behavioral economists and psychologists suggest that how a person experiences a reward depends on their decisional reference point. A decisional reference point acts as an internal baseline of expectations that determines whether an event feels positive or negative. For instance, finding a single slice of pizza feels like a positive reward if a person expects to have no food at all. If the person expected a large five-course meal, that same slice of pizza might feel like a disappointment.

Prior research links the processing of these expectations to the anterior cingulate cortex, a brain region involved in decision-making and emotion. Brain imaging shows altered activity in this area among people with depression. In some cases, doctors treat severe depression using deep brain stimulation, which sends electrical impulses to the pathways connecting the anterior cingulate cortex to other brain areas.

A team of researchers from NYU Langone Health, led by Paul Glimcher and Dan Iosifescu, suspected that a disruption in this reference point might cause the symptoms of anhedonia. They designed a study to measure whether people with depression have an unusually high reference point and whether they struggle to adjust these expectations when their environment changes.

To test these ideas, the research team recruited 120 adults. This group included 50 individuals clinically diagnosed with major depressive disorder and 70 healthy adults without psychiatric conditions. Each participant engaged in two separate computer-based behavioral tasks designed to measure their internal expectations and how those expectations shifted over time.

The first assessment was a virtual foraging game in which participants collected digital apples from animated trees. The design relies on the marginal value theorem, an ecological concept that explains how animals decide when to leave a patch of food. The theory suggests an animal weighs the energy gained from staying at a depleting food source against the time and energy required to travel to a new one.

To translate this to a human experiment, the goal was to collect as many apples as possible in a set time limit to earn a small cash payout. On each turn, a participant could choose to keep harvesting the current tree or wait through a travel delay, either three seconds or ten seconds, to move to a new tree. Every time a participant harvested the same tree, the yield diminished by an average of twelve percent.

Eventually, the diminishing reward falls below a person’s decisional reference point, prompting them to abandon the tree. Healthy adults tend to keep harvesting until a tree provides about four or five apples. People with major depressive disorder stopped harvesting much earlier. On average, they abandoned the tree when it yielded about eight or nine apples.

This behavior suggests that the baseline expectation for a reward is approximately fifty percent higher for individuals with depression compared to healthy participants. Interestingly, this difference did not mean the individuals with depression performed worse financially. The researchers noted that differences in total earnings between the two groups were not statistically significant, as both groups earned an average of twenty-seven dollars. The difference lay entirely in how they valued the diminishing returns, as patients required more apples to feel that staying at the tree was a positive choice.

Because the individuals with depression required a higher yield to feel rewarded, the task effectively identified their condition. The authors noted that this behavioral game could accurately separate patients from healthy adults using only three minutes of gameplay. In addition, the individual quitting points strongly matched the severity of a person’s depression as measured by traditional clinical interviews.

The second task investigated how participants adapted their reference points to changing environments. Participants received an endowment of five dollars and were asked to bid on thirty common snack foods using an economic bidding process. This step measured their baseline willingness to pay for a reward.

Following these initial bids, the participants entered an adaptation phase. They viewed and rated either their ten favorite snacks or their ten least favorite snacks over the course of three hundred trials. This repetitive exposure was intended to shift their internal reference point by immersing them in a highly rewarding or poorly rewarding environment.

After the adaptation phase, the researchers asked the participants to bid on the original thirty snacks again for ninety trials. When healthy adults spent time rating highly valued snacks, their internal reference point went up, and their subsequent bids on average snacks temporarily dropped. After several rounds of bidding, the healthy participants successfully relaxed their expectations and returned to their original baseline bids within a couple of minutes.

The individuals with depression initially shifted their bids in response to the adaptation phase, just like the healthy group. However, their expectations exhibited an inflexibility over time. The patients did not return to their baseline bids during the testing period, showing a rigid response to the changing reward environment.

As study co-author Aadith Vittala noted in a press release, “Patients with depression do not seem to be able to adapt their expectations normally as conditions change, which gives us a hint about what is wrong mechanistically in their brains.” The inability to adjust their reference point back to a normal state provides evidence of a broken cognitive mechanism that might maintain feelings of anhedonia.

While these findings present a novel way to measure depression, the study has some limitations and potential for misinterpretation. One limitation is the grouping together of potentially different subtypes of depression. Major depressive disorder encompasses a wide range of symptoms, and it is possible that these reference point abnormalities only apply to a specific subgroup of people who experience severe anhedonia. Readers should not interpret the study to mean that every individual with depression processes rewards in the exact same way.

The current research also does not determine whether these expectation deficits exist in other psychiatric conditions. Future research will need to explore whether an altered reference point is unique to depression or if it appears in other disorders characterized by a loss of pleasure. Scientists also hope to test these measurements in larger, more diverse populations to see if the game could be used outside of clinical settings. Another direction for future research involves testing whether physical or cognitive effort could help regulate the reference point in patients.

Identifying this inflexible reference point opens new doors for targeted interventions. “This looks to us like a therapeutic target, and we are already checking to see if a behavioral therapy or a drug might be able to fix this reference point stickiness,” Vittala explained in the press release. The authors suggest that measuring this cognitive mechanism could help doctors match patients with the most effective treatments remotely, saving time and reducing the need for repeated clinic visits.

The study, “Decisional reference point pathology: A cognitive mechanism for and a correlate of major depressive disorder in humans,” was authored by Aadith Vittala, Lulu Wu, Dongni Yan, David Liebers, Elizabeth Tell, Xiaotong Song, Damon Dashti, Kenway Louie, Candace Raio, Dan V. Iosifescu, and Paul Glimcher.