Experimental smartphone game quickly identifies patients with depression

An experimental diagnostic tool in the form of a computer game was able to quickly identify patients with depression based on anhedonia, a key feature of the disease, a new study shows.

Present in about 70 percent of patients with major depressive disorder (MDD), anhedonia is loss of the ability to enjoy normally pleasureful things. It is known to shift the point at which people decide whether an activity brings them pleasure.

Published online May 18 in the Proceedings of the National Academy of Sciences, the new report describes a low-cost game that reliably identifies patients with MDD in as little as three minutes. The game, which is based on neurobiological studies, can be played remotely on a smartphone, and its diagnostic accuracy compares well with the best existing tests, which are typically used across a number of in-person clinic visits, say the study authors.

Led by NYU Langone Health researchers, the new work found that people previously diagnosed with major depression by standard tests stopped taking pleasure in the game's main activity 50 percent sooner than healthy subjects.

Our behavioral game gives us clues to what is happening in the brains of patients with depression, which we hope will let us identify them as reliably as finding heart disease by taking someone's blood pressure."

Paul W. Glimcher, PhD, co-senior study author, chair of the Department of Neuroscience and director of the Institute for Translational Neuroscience, NYU Grossman School of Medicine

Decision point

Landmark theories going back decades argue that the pleasant or unpleasant quality of something (a reinforcer) depends on a person's expectations. For example, if no food is expected, then a single pizza slice may be positively reinforcing (pleasureful).

The current findings argue that at least in a large subset of patients, MDD involves a pathological disruption of a person's expectations, shifting the reference point at which the person decides whether something is pleasing. With this shift, activities that would be positively reinforcing for a healthy person are experienced negatively. The behavioral game takes advantage of a key study that first linked anhedonia with the anterior cingulate cortex, a brain region in which such expectations are thought to be set.

The researchers asked 120 game players—50 diagnosed with major depression and 70 who were not—to compete to collect the most apples falling from digital trees. Researchers use such foraging tasks because reward-seeking circuitry, especially regarding something that looks like food, has been engrained in the mammalian brain by evolution.

As a player encountered a tree, foraging yielded fewer apples with each round of harvesting, and the researchers tracked when each player gave up on a tree and moved to the next one. The game measures the point at which a participant no longer finds a tree's apple yield rewarding. On average, healthy people stuck with a tree until the yield dropped to five, but people with MDD typically left a tree much earlier (before the yield dropped to eight or nine apples, depending on the severity of their depression—a nearly 50 percent increase in the decisional reference point).

A second study activity asked participants to bid on snacks and sought to measure how their decisional reference point responded to changes in their environment (e.g., changes in their options). Previous work showed that when healthy people are asked to rate how much they are willing to pay for each snack in a small group of their previously indicated favorites (high value), and then to bid on a larger random group of snack foods (lower value), their reference points rise and their bids dip, with all items on the random list seen temporarily as less valuable.

After healthy participants bid on their favorites and then went through several rounds of bidding on the random snack list, these reference points relaxed back to levels seen before the participants bid on favorites. In contrast, the researchers found that reference points for MDD patients did not go back to their original setting after the same sequence.

"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," said Aadith Vittala, an MD/PhD student in Glimcher's lab and the study's co-first author. "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."

"Depression is increasingly thought of as an umbrella term that may include several distinct conditions," added co-senior author Dan Iosifescu, MD, a professor in the Department of Psychiatry at NYU Grossman School of Medicine. "Measuring reference points may help us identify a specific subtype of depression linked to anhedonia, clarify its disease-causing brain computations, and tailor treatments. And we may be able to do this remotely by asking patients, rather than traveling repeatedly for in-person visits, to spend a few minutes per week playing a smartphone game that lets us quickly adjust their treatment."

Along with Dr. Glimcher and Dr. Iosifescu, study authors from the Institute for Translational Neuroscience were co-first study author Lulu Wu, as well as Dongni Yan; David Liebers, MD; Elizabeth Tell; Xiaotong Song; Damon Dashti; Kenway Louie, MD, PhD; and Candace M. Raio, PhD. The study was funded by National Institute of Mental Health grant R21MH126197.