Timothy Salthouse, PhD, a noted cognitive psychologist at the University of Virginia, has shown that giving a test once is not enough to get a precise account of an individual's mental function.

Repeating tests over a short period of time might give more accurate cognitive scores, thus refining diagnostic workups.

This study appears in the July issue of Neuropsychology, which is published by the American Psychological Association (APA).

Salthouse conducted 2 studies (90 participants in the first, 1600 in the second) in which he gave 16 cognitive and neuropsychological tests to subjects divided into groups of 18-39, 50-59 and 60-97 years old.

Salthouse found that the difference between an individual's score on the same test given three times over two weeks was as wide as the difference between the scores of people in different age groups.

Think of it as like getting test scores from an individual who acted like 18 the first time, 50 the next test date, and 97 years old by the final testing day.

Salthouse's findings bring up questions about the value of just one test.

"I don't think many people would have expected that the variability would be this large, and apparent in a wide variety of cognitive tests - not simply tests of speed or alertness," says Salthouse.

Physicians usually give tests that include vocabulary, word recall, spatial relations, pattern comparison, etc. to understand normal function and diagnose impairment.

Experts use the scores to give a diagnosis, to make a distinction between diagnoses or note changes in the level of functioning.

Test scores fall within standardized cutoffs, these cutoffs then determine the treatment, insurance, etc.

However, the inconsistencies of one-time test could make it difficult to determine whether an individual is

• really impaired
• getting better or worse

Instead of indicating normal "short-term fluctuation."

As a result, Salthouse arrived at the notion that "everyone has a range of typical performances, a one-person bell curve."

Salthouse approaches his notion like a sports statistician, explaining that "a given test will  net a performance somewhere along this curve, the way a hitter's good and bad days are included in his seasonal statistic."

Some people's scores would figure more closely to their average. On the other hand, classification based on one-time testing may yield inaccurate results for people with high internal variation.

Salthouse says that perhaps it's time to recognizer cognitive abilities as a "distribution of many potential levels of performance." Many psychologists are making the same claim, they it has come up in their own studies.

He suggests using what he calls the measure burst procedure which "bases understanding on several parallel assessments within a relatively short period."

This procedure may likely yield better results since there will be a better basis for adjusting individual change.

Salthouse, "More will have to be learned about this phenomenon and the conditions under which it operates." 

Multiple tests take more time and they cost more. However, Salthouse argues that they may be necessary to "distinguish short-term fluctuation from true ability level."

New test norms and true test equivalents would have to be developed.

Lastly, Salthouse believes that "measures of within-person variability" might prove to be a valuable diagnostic marker by itself. Salthouse gives an example about something and he and his fellow cognitive psychologists are discussing: whether wilder fluctuations within one person's test scores are an early warning of mental decline. 

Source: APA.org

Source: APA.org