What is the MMSE test? The classic 30-point cognitive screening exam

The MMSE test, short for Mini-Mental State Examination, is a brief 30-point cognitive screening exam that clinicians use to check for possible cognitive impairment. In about 5 to 10 minutes, its 11 items sample orientation, registration, attention and calculation, recall, language, and the ability to copy a figure. Psychiatrists Marshal Folstein, Susan Folstein, and Paul McHugh published it in the Journal of Psychiatric Research in 1975, and for decades it was the most widely used bedside cognitive screener in medicine. This article explains what the MMSE asks, how the 30-point scale is interpreted, who publishes it today, what it can and cannot detect, and how it differs from a test of reasoning ability.

The original 1975 paper described the MMSE as a "practical method for grading the cognitive state of patients for the clinician." The examiner reads a fixed set of questions and simple tasks aloud, and the patient responds verbally or with pencil and paper. A comprehensive Cochrane review describes it as an 11-item assessment covering attention and orientation, memory, registration, recall, calculation, language, and the ability to draw a complex polygon.

A classic clinical reference, the textbook Clinical Methods, breaks the 30 points down this way:

• Orientation (10 points): questions about the current time and the current place. This is the largest single block of points on the exam.

• Registration (3 points): immediately repeating back newly presented information, a check on whether new material is being taken in at all.

• Attention and calculation (5 points): a short mental task that requires holding information in mind and working with it.

• Recall (3 points): remembering the registered material after a brief delay, the exam's core short-term memory check.

• Language (8 points): a set of quick language functions performed on request.

• Constructional ability (1 point): copying a polygon figure, the exam's only visuospatial drawing item.

The attention and memory items tap the same territory as classic "working memory" measures, which hold and manipulate information over seconds. Readers who want a deeper look at that family of tasks can see our article on the digit span test.

Every correct response earns points, and a perfect performance yields 30. According to the Cochrane review by Creavin and colleagues, the conventional cutoff is 24, with lower scores indicating increasing cognitive impairment. In their meta-analysis of community studies, a cutoff of 24 detected dementia with a "sensitivity" (the share of true cases flagged) of about 85 percent and a "specificity" (the share of healthy people correctly passed) of about 90 percent.

A single fixed cutoff hides a lot, though, because MMSE scores vary with age and education among people with no diagnosis at all. In a landmark JAMA study, Crum and colleagues analyzed MMSE results from 18,056 adults in the Epidemiologic Catchment Area surveys. The median score fell from 29 among adults aged 18 to 24 to 25 among those 80 and older. Education mattered even more: the median was 29 for people with at least nine years of schooling, 26 for those with five to eight years, and 22 for those with zero to four years. A healthy older adult with little formal education can land below the standard cutoff, while a highly educated person in early decline can still score above it. Tombaugh and McIntyre's comprehensive review reached a similar verdict: scores are affected by age, education, and cultural background, though not by gender, so clinicians are advised to interpret results against appropriate norms rather than a single universal threshold.

Population performance on cognitive tasks also drifts across decades, which is one more reason norms collected in one era should be applied cautiously in another. Dr. Warne discusses that drift in the video above.

The MMSE is commercial test material, which surprises many people who remember photocopying it freely. PAR, Inc. publishes the current edition, the MMSE-2, authored by Marshal Folstein and Susan Folstein. According to the publisher, the MMSE-2 comes in three forms: a Brief Version taking about 5 minutes, a Standard Version taking 10 to 15 minutes, and an Expanded Version taking about 20 minutes, with norms spanning ages 18 to 100.

The route to that arrangement generated real controversy. As law professor Robin Feldman and physician John Newman document in the Stanford Technology Law Review, the authors retained copyright after the 1975 publication but did not enforce it for roughly 25 years, during which the MMSE spread through textbooks, pocket cards, and websites. They transferred the copyright to a company called MiniMental LLC in 2000, which granted PAR an exclusive worldwide license in March 2001. Enforcement began in earnest around 2010, with per-test fees and removal of the exam from textbooks and online resources, and in 2011 PAR requested a takedown of an open-access alternative called the Sweet 16 over similarity concerns. The current StatPearls clinical reference notes that what was once the gold standard is now declining in use, partly because of these copyright restrictions and costs. Clinicians who want to administer the MMSE legitimately license it through PAR.

The MMSE is a screening test. It flags people who may need a full workup, and it offers a quick way to track gross change over time. Tombaugh and McIntyre found its reliability and construct validity satisfactory, with high sensitivity for moderate-to-severe cognitive impairment. That is the job it was built for, and it does that job in minutes.

Its limits are just as well documented. Sensitivity drops for mild impairment, and the same review found the content highly verbal, with language items so easy that they contribute little to detecting subtle deficits. The Cochrane authors concluded that the MMSE contributes to a dementia diagnosis in low-prevalence settings but "should not be used in isolation to confirm or exclude disease." Crum and colleagues made the same point in their norms paper: scores identify current cognitive difficulties and should not be used to make formal diagnoses. A diagnosis of dementia rests on clinical history, informant reports, neurological examination, and often a full neuropsychological battery.

There is also a "ceiling effect," meaning the test is too easy to distinguish people at the upper end. Most cognitively healthy adults score at or near 30, so the MMSE cannot rank ability among people who pass it. A score of 30 says the screen found no red flags; it says nothing about whether someone's reasoning is average or exceptional. Clinicians who need a fast screen with a visuospatial component sometimes pair brief instruments, and our article on the clock-drawing test covers one common companion measure.

The Montreal Cognitive Assessment, or MoCA, is the MMSE's most prominent modern rival, and it was designed specifically to catch what the MMSE misses. In the original validation study, Nasreddine and colleagues reported that the MMSE detected only 18 percent of patients with "mild cognitive impairment" (a decline noticeable in testing but short of dementia), while the MoCA detected 90 percent. For mild Alzheimer's disease the figures were 78 percent versus 100 percent, though the MMSE showed somewhat higher specificity. The StatPearls reference adds that the MoCA covers more cognitive domains with greater sensitivity and specificity, and it has largely displaced the MMSE in many clinics. For a full breakdown of that instrument, see our article on the MoCA test.

The practical difference is item difficulty. The MoCA includes harder tasks, so healthy adults no longer cluster at a perfect score and early decline becomes visible. Both remain screeners, however, and neither replaces a diagnostic evaluation.

A screening test asks whether something may be wrong. An intelligence test asks how strong someone's reasoning is relative to other healthy adults. The MMSE was never designed to do the second job: its items are deliberately easy, and a perfect score is the expected outcome for a healthy adult. Anyone hoping to learn about their intellectual strengths from an MMSE score will learn nothing beyond the absence of red flags.

Measuring reasoning ability requires items that challenge people across the full range. The Reasoning and Intelligence Online Test, developed by RIOT IQ with psychometrician Dr. Russell T. Warne, was built for adults 18 and older and uses 15 subtests across six cognitive indices: verbal reasoning, fluid reasoning, spatial ability, working memory, processing speed, and reaction time. It takes about 52 minutes and reports scores on the familiar mean-100, standard-deviation-15 IQ scale. Like any online measure, it does not replace an individually administered diagnostic evaluation, and it is intended for healthy adults rather than for detecting impairment. If you are curious where your reasoning ability falls, you can take the RIOT IQ test at riotiq.com.

It depends on age and education. The conventional cutoff is 24, but median scores in large community samples range from 29 in young, educated adults down to 22 in adults with fewer than five years of schooling. A clinician interprets 24 against the appropriate reference group, never in isolation.

About 5 to 10 minutes for the classic version. The publisher reports that the MMSE-2 Brief Version takes about 5 minutes, the Standard Version 10 to 15 minutes, and the Expanded Version about 20 minutes.

No. It is a screening exam for cognitive impairment. Its items are easy by design, and most healthy adults score at or near the 30-point ceiling, so it cannot measure intelligence or rank ability among healthy people.

Not legitimately. The MMSE is copyrighted, and PAR holds an exclusive worldwide license to publish and distribute it. Unofficial copies circulating online may be outdated or altered, and self-administration without a trained examiner is not how the test is validated.

The MMSE-2, published by PAR in 2010, is the current edition from the original authors. It keeps the familiar format and adds Brief and Expanded forms alongside the Standard version, with norms for ages 18 to 100.

A score of 30 means the screen found no evidence of gross impairment. Because of the test's low ceiling and limited sensitivity to mild deficits, subtle problems can still exist. Anyone with persistent memory concerns should seek a fuller evaluation regardless of an MMSE result.

1. Folstein, M. F., Folstein, S. E., & McHugh, P. R. (1975). "Mini-mental state": A practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research, 12(3), 189-198. pubmed.ncbi.nlm.nih.gov

2. Tombaugh, T. N., & McIntyre, N. J. (1992). The Mini-Mental State Examination: A comprehensive review. Journal of the American Geriatrics Society, 40(9), 922-935. pubmed.ncbi.nlm.nih.gov

3. Crum, R. M., Anthony, J. C., Bassett, S. S., & Folstein, M. F. (1993). Population-based norms for the Mini-Mental State Examination by age and educational level. JAMA, 269(18), 2386-2391. pubmed.ncbi.nlm.nih.gov

4. Creavin, S. T., Wisniewski, S., Noel-Storr, A. H., et al. (2016). Mini-Mental State Examination (MMSE) for the detection of dementia in clinically unevaluated people aged 65 and over in community and primary care populations. Cochrane Database of Systematic Reviews, 2016(1), CD011145. pmc.ncbi.nlm.nih.gov

5. Nasreddine, Z. S., Phillips, N. A., Bédirian, V., et al. (2005). The Montreal Cognitive Assessment, MoCA: A brief screening tool for mild cognitive impairment. Journal of the American Geriatrics Society, 53(4), 695-699. pubmed.ncbi.nlm.nih.gov

6. Feldman, R., & Newman, J. C. (2013). Copyright at the bedside: Should we stop the spread? Stanford Technology Law Review, 16(3), 623-655. pmc.ncbi.nlm.nih.gov

7. Martin, D. C. (1990). The mental status examination. In H. K. Walker, W. D. Hall, & J. W. Hurst (Eds.), Clinical methods: The history, physical, and laboratory examinations (3rd ed., Chapter 207). Butterworths. ncbi.nlm.nih.gov

8. PAR, Inc. (n.d.). MMSE-2: Mini-Mental State Examination, 2nd Edition. parinc.com

Hero photo: a person completing printed paperwork with a pen at a desk. Photo by Helloquence via Wikimedia Commons / Unsplash, CC0 (https://creativecommons.org/publicdomain/zero/1.0/), cropped.