How fast does IQ decline? Can you do anything about it?
At some point in their studies, most psychology students have seen a somewhat depressing figure that looks something like this:
Gf = Fluid intelligence
Gc = Crystallized intelligence
Gv = Visual processing
Gs = Processing speed
Glr = Long term storage and retrieval
Ga = Auditory processing
Gsm = Short term memory
This data resembles what is found in the Wechsler-Bellevue standardization samples. The Wechsler-Bellevue was the first test to provide this information, and indicates that both verbal and performance peak at around age 22! According to their data, people as young as 40 have performance IQs 20% lower than 16 year olds.
The data is also similar to the standardization samples of the WAIS-R, which also indicates that performance IQ drops after age 22, but that people reach their verbal IQ peak around age 30.
The idea that university students may never again have the same level of performance IQ is a little sad, but thankfully that’s probably not the case.
The issue with many studies on the relationship between age and IQ, which includes all the studies mentioned so far, is that the data is cross-sectional. Why is that a problem? We can’t assume that lower results at age 70 than at age 25 are due to participants being older, because it’s possible that the people born 70 years before the study have always had lower IQs. The data supports this hypothesis and is well known as the Flynn Effect. To get an idea of how powerful this effect is, Ulric Neisser analyzed available standardization samples and estimated that if individuals who took the test in 1932 were transported forward in time to take the test in 1997, their average score would be 80 — 1.33 standard deviations below the mean!
Knowing that the cross sectional data is likely giving us a pretty large overestimation of the decline of intelligence with age, we can take a look at what the longitudinal data tells us and hope for a better picture.
One of the largest and most famous studies of this kind is the Seattle Longitudinal Study conducted by K. Warner Schaie. The following graph uses data from that study:
As can be seen in the figure, Schaie’s study indicates that the steep decline in intelligence usually only starts after age 60, with only perceptual speed declining seriously before 60. The study showed that verbal ability is stable for most people until age 70, after which the decline is still very slow!
So can we trust these more optimistic results and move on? Unfortunately, longitudinal IQ studies come with their own problems. Firstly, while the effect may be small, there can often be practice effects when the same test is taken multiple times by the same participants. Secondly, in the case of many of the studies that have actually been performed like the Seattle study, there is self-selection and this results in a clear over-representation of middle class people with health care plans (since they recruited through health care systems). Finally, the data comes from people who lived to the age of 80, and IQ is strongly correlated with health and longevity! Each of these factors likely causes the SLS to underestimate the decline in intelligence through the lifespan.
So we know the actual decline for most types of intelligence doesn’t start as early as 22, but probably starts before age 60. Is there anything you can do to avoid losing intelligence as you age?
I struggled to find an IQ boosting method with widely reviewed/accepted studies behind it. For the most part, the articles I found said that cognitive training or programs that claim to boost IQ have not been validated, largely due to issues with reliability, issues with the stability of the“increased” IQ score, and the shortcomings of using data from a scale that isn’t a ratio scale (where the same score increase can have different meanings at different points along the scale). Some studies indicate that for both children and adults music can have the effect of boosting executive functions and as a consequence, certain measures of IQ. However, even these studies are met with a large degree of skepticism.
To end on a slightly more optimistic note, there are other factors that could be causing declines in IQ over time to be overestimated. Older individuals are often slower (recall that Gs fell fastest of any intelligence measure) and that could be trickling over and having an effect on other test components with speeded elements. Older individuals are also particularly weak on certain memory components. Lapses in short term memory during a test can result in scores that underestimate certain abilities. So the bright side here is that perhaps for the most part you will only lose memory capabilities and speed, but technically have other areas of intelligence decline slower than one would expect based on the data.
-260522465
Sources:
Mackintosh, N J. (2011). IQ and Human Intelligence, 2nd Edition
Neisser U (1997). “Rising Scores on Intelligence Tests”. American Scientist 85: 440–7.
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3950413/
http://www.pickthebrain.com/blog/why-you’re-only-getting-better-with-age/
http://www.highiqpro.com/iq-cognitive-health-aging/the-5-factors-of-intelligence-over-the-lifespan
