Should we require a cognitive fitness assessment for certain government leadership positions?
At the start of Donald Trump's second presidential term, he was 78 years, 220 days old, and at the start of Joe Biden's term, he was 78 years, 61 days old, making them the oldest and second-oldest people, respectively, to assume the office of the presidency. During the bitter campaign season of the 2024 presidential election, the candidates' advanced ages were a central feature of debate and discussion, ultimately culminating in Joe Biden's disastrous debate performance and later his withdrawal from the race entirely. Many voters were left wondering whether the most powerful military on the planet, or the health and safety of hundreds of millions of people should be managed by an individual with questionable cognitive capacity. While a cognitive fitness requirement for office sounds like a useful safeguard, there are numerous complexities that make it unlikely any such requirement will come to fruition. (In case you were wondering, the youngest president was Teddy Roosevelt, who was 42 years, 322 days old when he assumed office)
Where is the bar? Norm Referencing Versus Criterion Referencing.
Norm Referencing
In clinical neuropsychology, we use large datasets from people of all ages and demographic characteristics to gauge where a patient's performance on a test falls in relation to their peers. In other words, if you are a 58-year-old Asian American female with an Associate's degree education, we have test data from people with those characteristics to use as a yardstick to compare to your score on a test. This allows us to tell if your score is better, the same, or worse than your peers, and we use many test scores to tell if certain cognitive abilities are worse than they "should be," for an individual with your demographic characteristics. In clinical neuropsychology, "the bar" is how your peers are scoring on various tests. This way of looking at the test data is called "norm referencing" because it involves comparing a patient's data to a "normative sample."
Norming a test based on age also has its drawbacks when a cognitive fitness test is used to judge a candidate's fitness for office. In the cases of Joe Biden and Donald Trump, even if their memory or executive functioning scores were right on the mark for a 78-year-old, do 78-year-olds as a group, really have the cognitive fitness to responsibly execute the responsibilities of a President? Another conundrum: Let's say a presidential candidate was 48 years old, but gave the cognitive fitness performance that was normal for a 78-year-old. As a clinician, this would make me very concerned that something was very wrong with this 48-year-old's brain, and that they did not possess the cognitive fitness to be a President. However, if we gave the same test to a 78-year-old presidential candidate who got the exact same score as the 48-year-old with probable dementia, we would probably conclude that candidate is cognitively intact. In short, even though these candidates got the exact same score on the test, we drew very different conclusions about their cognitive fitness, based on their age.
Criterion Referencing
Rather than judging a person's performance against some normative group of interest, criterion referencing involves judging a person's performance against a minimum level or minimum score (called the criterion). Most of us have encountered criterion referencing in school, where we are judged against a grading scale and we need to get a certain number of questions right on a test in order to earn an "A," "B," "C," etc. In other words, the number of questions you need to answer correctly to earn a grade of "A" is the same for everyone, regardless of what other people's scores on the test are. In school, the grading scale is the criterion. Criterion referencing is also used in certain professional situations where achieving a minimum level of performance is needed. For example, in certain fields that are highly safety-sensitive with low margin of error and potential for tragic consequences (think airline pilots or surgeons), criterion referenced assessments have already been used to protect the public from impaired professionals. Think about how an airline pilot needs to be able to fly a plane safely (the criterion) regardless of their age or other demographic characteristics.
By extension, having a criterion referenced assessment for a presidential candidate has significant advantages over a norm referenced assessment because the minimum level of performance for the office is (or should be) the same, whether a candidate is 48 years old or 78 years old. The problem is, what is the criterion exactly? For pilots, the criterion could be ability to perform a safe maneuver, or ability to diagnose and fix an in-flight technical issue in a flight simulator. The problem for Presidential candidates is that there is no "flight simulator" for the office of the Presidency, and there is no specific and objective cognitive skill or skills (the criterion) that would allow us to predict with any appreciable accuracy a candidates cognitive fitness for office.
Norm Referencing - With a Twist
Sometimes if the criterion is hard to identify or hard to test for, norm referencing can be used in a way that helps us characterize someone's cognitive fitness for a specific role like the Presidency. Rather than using age and other demographic characteristics to define a normative group, some researchers and clinicians (including in this office) use variables like occupation to define a normative sample. For example, let's take our surgeon from the previous example, and rather than comparing her to other people of her age, gender, and education, let's compare her to a mixed gender sample of other surgeons of different ages and different ethnicities. This approach allows us to see where a test taker falls in relation to other people who are performing the same kind of work, but in a way that does not depend as much on age. If a surgeon is testing in the bottom 2% of their surgeon peers on some test, then a case might be made that their skill set is deficient in some way. However, norm referencing loses its usefulness in situations where there isn't enough data against which to compare a person's scores. While there may be tens of thousands of surgeons to compare against our hypothetical surgeon, as of this writing there have been only 45 distinct individuals who have ever served as a sitting US President (interestingly, apart from Donald Trump, Grover Cleveland is the only other president to have served two non-consecutive terms in office as both the 22nd and 24th presidents). On top of that, since most of these 45 individuals are deceased, we wouldn't be able to get any test data from them, which further shrinks the size of our sample. On top of that, the cognitive demands of the office of the presidency have undoubtedly changed over the last 250 or so years making the idea of cognitive fitness a kind of moving target.
Summary
Having the two oldest presidents in our nation's history serve back-to-back terms has certainly raised the issue of cognitive fitness for office. As valid as this concern is, challenges related to basic assessment principles and statistics make the developmental of any kind of formal and objective assessment of cognitive fitness a remote possibility, at least in the foreseeable future. For now, voters themselves will need to decide whether a candidate possesses or lacks cognitive fitness for office.
