What if you read the dramatic claim that playing a certain video game could reduce your risk of dementia by 25%? I hope you’d be skeptical. You might think that it was a ploy to increase the video game’s sales. Whenever you read a claim like this, you should ask to see the evidence.

In the case of a brain game called Double Decision, there actually IS evidence for such a claim. Let’s read about it through the CNN story about the research.

“A certain type of brain training appears to prevent or delay dementia by some 25% in people older than age 65, according to new research.”

You can see examples of the Double Decision game in the photos.

The game shows you a vehicle in a particular setting. Then it flashes a small black and white Route 66 sign. To score points, you have to make two decisions, First, you’ll see two cars and you have to decide which of them you just saw. Then, you have to point out where in the periphery the Route 66 sign showed up. As the game progresses, it gets harder. The car becomes harder to identify. And the Route 66 sign starts to be more difficult to find among a bunch of distractors. You can watch a preview of the game here.

How do we know that this game helped prevent dementia? Here’s how CNN described the study:

[The study] tested three types of cognitive training on more than 2,800 volunteers with an average age of 74. All were free of dementia at the start and lived independently in six communities around the United States. A fourth group who received no training served as a control.

“A real strength of the study is this was a really representative population — 25% of the participants were minorities,” Albert said. “So we can truly say the findings generalize to the entire US population.”

One group was focused on memory, learning techniques for remembering word lists, text materials and details of stories. A second group underwent training focused on reasoning, such as solving problems and identifying patterns that could help with daily life.

The third group played the game Double Decision.

Volunteers were trained in person twice per week for 60 to 75 minutes per session over five weeks. At the end of the first year, about half of the people in each cognitive training group underwent an additional “booster” of four one-hour sessions. [In the booster group,] another four hours of training was also done at the end of the third year of the study, for a total of 22.5 hours.

a) So if you’re keeping track, this is a design with 7 groups. Can you name them all?

Here’s how the results are described by CNN:

… when investigators compared the three groups with their Medicare records 20 years later, they found it was only the [Double Decision game] that contributed to a 25% reduction in dementia diagnoses compared with the control group.

That benefit, however, was only for a subset of the volunteers…“The 25% reduction in risk for dementia was only in people who had the original training on [Double Decision] and then the booster sessions. If you didn’t have the booster sessions, you didn’t benefit,” [author] Albert said.

b) Sketch a bar graph of these results. Your graph should have seven conditions on the x-axis and a clearly labeled y-axis. Which group should stand out this bar graph?

c) Evaluate this study’s external validity. In the quote above, the author claims that, becaues the study included 25% minorities, it is a representative sample. What do you think? (Keep in mind that for external validity, it’s “how” not “how many” that matters).

d) Assume seniors were randomly assigned to the seven groups. Can this study support the claim that playing Double Decision lowers your dementia risk? Apply the three criteria of covariance, temporal precedence, and internal validity.

You can access the original empirical article here.