When I wrote my final research paper, I just dove into it wanting to learn more about how we measure brain function and memory loss, particularly in elderly people, but I came to realize how limited some of these cognitive tests are. I never fully thought of how many problems I’d run into with the tests we are still using today, which date back to the ’90s. One of the biggest things I took away is how unfair some of these tests are around language and culture. Many of the most widely used cognitive tests exist only in English or Spanish, leaving a large portion of the population behind. That felt wrong to me, like why should someone be diagnosed with the wrong cognitive level just because the test wasn’t made with them in mind? The stats portion of the paper was a learning curve for me. I never thought I’d be using Excel so much, let alone learn formulas for bell curves and ways to look at z-scores. That showed me that even when the numbers on the surface looked all right, some of these tests were filtering out important numbers to make themselves look better.

One of my most interesting findings was that certain tests have no resemblance to each other, even though they’re all supposed to be measuring the same quality. The “Mini Mental State Exam” was the one I ended up using as the control, because it was the most accurate and had a perfect normal distribution. Two of the cognitive tests, “Mini-Cog” and the “Wechsler Adult Intelligence Scale”, were far from the control, even though they had some of the best scores. Their bell curve were completely skewed, which means that their data ranged way in either direction, and that’s a red flag. It made me wonder: If these tests are producing varying scores, how are they still being used to diagnose conditions as serious as Alzheimer’s? Overall, I did not have an easy time finding enough tests that fit what I was looking for, particularly ones that were focused on older adults. That was frustrating because that essentially did prove my point, that there’s just not enough of a focus on age-specific testing, and older people are getting lumped in with various ages. That’s not fair considering that these results have the potential to influence their future care and treatment.

If anything, this paper taught me how crucial it is to question systems we assume are good because they’ve been around for long periods, but just because a test is “well known” or “commonly used” doesn’t mean it’s accurate or fair. I do believe we need to shift how we think about memory and aging, and I do believe part of that shift is in ensuring the tools we lean on are current and honest. What I learned from this experience was not only to do more research regarding cognitive tests, but also the importance of looking deeper into data at face value.