Lumosity, one of the leading companies in the controversial “brain training” industry, is having its credibility tested once again.

Researchers at the University of Pennsylvania recently conducted a controlled, randomized test involving 128 young adults (71 males and 57 females). Participants were asked to play five 30-minute brain-training Lumosity games each week for 10 weeks.

Specifically, the study looked at whether or not brain training had any impact on users’ risk versus reward decision-making.

At the end of the test, researchers said they found no evidence for relative benefits of cognitive training with respect to changes in decision-making behavior or brain response, or for cognitive task performance beyond those trained.

In other words, the only improvements noted were that participants got better at the games they were playing, a result that’ll hold true with virtually any game you play given enough practice.

This isn’t the first time Lumosity has come under fire. Last year, the Federal Trade Commission fined the company $50 million for deceptive advertising (but later reduced it to $2 million because the company couldn’t afford the large fine).

Lumosity issued the following statement to Ars Technica via e-mail:

Looking at the link between cognitive training and risk-reward decision-making is a novel approach—most people don’t associate brain training with decision making or risk sensitivity—and at Lumosity, we encourage taking an innovative approach to research...

However, it’s a giant leap to suggest this study proves cognitive training is “no better than video games at improving brain function”: in fact, the study has a much narrower scope, focusing on risk sensitivity in young adults. There remain many open questions in the field—how, why, and in what circumstances cognitive training is efficacious—and so painting in such broad strokes potentially undermines this important, ongoing research area. We remain committed to supporting quality research, regardless of the outcome: every study can be built on, and they all move us closer to answering open questions—in turn, improving the quality of products available.