Researchers analyzed Swedish administrative records to test whether cognitive ability is linked to real actions for the public good. The work was published in The Economic Journal.
The first analysis involved more than 1.2 million men and about 3,000 women. All of them took a standard cognitive ability test at military conscription between 1969 and 1997. Their scores were matched against tax data on charitable donations for 2012-2015, voter lists for the 2009 European Parliament elections and a national vehicle database.
People with high scores more often performed all three types of actions. Among men with the highest results, the probability of donating was more than five times higher than among men with the lowest. A one-standard-deviation increase in the test score (a measure of the spread of values around the mean) corresponded to a relative increase in the probability of donating of 40 percent, of voting of 31 percent and of owning an eco-friendly car of 14 percent. Women showed similar trends for voting and charity.
Logical and verbal reasoning were most strongly linked to prosocial behavior; they are considered good indicators of general fluid intelligence, that is, the ability to reason abstractly and solve new problems.
However, a statistical association does not prove causation. People with high scores may grow up in families or areas where participating in public life is customary. To account for this background, the researchers conducted a second analysis on 5,786 pairs of twins. Twins usually share an environment and genetics, so the comparison helps filter out extraneous factors.
The smarter twin in a pair voted and donated to charity more often. But the size of the link fell by about half: one standard deviation of intelligence accounted for a 26 percent increase in the probability of donating and 13 percent for voting. The link with eco-friendly cars also fell to 13 percent. This means family factors explain part of the picture, but not all of it.
A third analysis tested whether the results are explained by lifestyle. Education explained a notable share of the link; schools may instill prosocial values. Income was not the main mediator: intelligent people did not donate and vote simply because they have more spare money. After adjusting for education and income, the link between intelligence and voting and charity remained, while the link with eco-friendly car ownership became statistically insignificant.
Finally, about 600 men completed a questionnaire in which they rated their own altruism. Higher intelligence was associated with greater self-rated altruism: one standard deviation of intelligence accounted for 0.16 standard deviations of altruism. These self-reports are consistent with the behavior in the registries, although they do not allow the mechanism to be pinpointed.
The authors emphasize limitations: the data relate mainly to Swedish men born in 1951-1979, and prosociality was measured by only three specific actions. People with low scores may show cooperation differently, for example by helping neighbors. Future work should examine how empathy and the ability to recognize emotions interact with general intelligence.