Is IQ fixed for life?

What the evidence shows

Stable, in rank order

The 1994 mainstream statement signed by 52 researchers said IQs gradually stabilize during childhood and generally change little thereafter (Gottfredson, 1997). One striking test of that comes from Scotland. In the Lothian Birth Cohort 1921, people took the same intelligence test at age 11 and again at age 90, and the two scores correlated .54, or .67 after correction for range restriction (Deary et al., 2013). Children who scored high tended to stay high, but the link was far from perfect.

Not fixed for individuals

The 1996 APA task force reported that in one long-term study the average change between ages 12 and 17 was 7.1 points, and some children changed by 18 points (Neisser et al. 1996). Every score also carries measurement error, with clinical manuals reporting a confidence interval of around plus or minus 5 points (Wechsler, 2008). And retesting alone raises scores by about 4 points on average (Hausknecht et al., 2007), so not every change is a change in ability.

Education moves it

A 2018 meta-analysis of studies with more than 600,000 participants found that each additional year of education raises IQ by about 1 to 5 points, and called education the most consistent, robust and durable method yet found for raising intelligence (Ritchie & Tucker-Drob 2018).

Whole populations have moved

Average measured IQ rose through the 20th century, about 2.8 points per decade across 271 samples in a 2015 meta-analysis (Pietschnig & Voracek, 2015). Norwegian data comparing brothers show that the rise, its turning point and the later decline reflect environmental factors, not changes in which parents had children (Bratsberg & Rogeberg, 2018).

Why the myth persists

Partly because heritability is misunderstood. Twin studies find that heritability of general cognitive ability rises from 41% at age 9 to 66% at age 17 (Haworth et al., 2010), and people often read "heritable" as "unchangeable." The APA task force called that assumption a common error (Neisser et al., 1996), and the 1994 statement stressed that high heritability does not mean intelligence is unaffected by environment or fixed at birth (Gottfredson, 1997).

How to read a claim that IQ can or cannot change

• Is it about rank order (who scores higher than whom) or about the level of a person's score?
• Is a change bigger than the few points of measurement error and practice?
• Did the gain last, or fade once a program ended?

What is actually true

• IQ is fairly stable from later childhood on, especially in rank order.
• Individual scores move by several points, sometimes many more.
• Schooling raises measured intelligence; some short-term gains, such as from early programs, fade.

Read more in can your IQ change, intelligence across the lifespan, environment and intelligence and the Flynn effect guide. Related myth: brain training games.