Background: Li & Kanazawa (2016) confirmed the already established relationship between socialization with friends and psychological wellbeing, yet also found that in highly intelligent people (measured with the Peabody Picture Vocabulary Test, PPVT) the relationship is reversed. They explained their findings with the Savanna Theory of Happiness, which claims that the human brain has difficulty dealing with situations that did not exist in the ancestral environment, and since living without regular contact with friends qualifies as such a situation, it decreases subjective wellbeing. Moreover, the more intelligent the individual, the more successfully their brain can cope with evolutionarily novel problems, hence the weaker the effect of a lack of friends on life satisfaction.
Aim: Our primary aim was to replicate Li and Kanazawa’s finding that more intelligent people are more satisfied with life when meeting friends less frequently and to critically examine its explanation, the Savanna Theory of Happiness. Our secondary aim was to examine the differential effects of fluid and crystallised intelligence (Gf and Gc) on the relationship between wellbeing and social interactions.
Method: We recruited volunteers from the Hungarian chapter of the high IQ society Mensa (N=193). Every participant had an age-standardised IQ score above the 98th percentile (i.e. above 130) on a fluid intelligence test (the Raven’s Advanced Progressive Matrices, RAPM). In an online survey participants answered questions about their ego-centred networks. They subsequently filled the Hungarian adaptation of WHO’s Well Being Index (WBI-5) as well as a computerised adaptive vocabulary test.
Results: We could not replicate Li and Kanazawa’s finding: people in our sample of high IQ individuals experienced higher wellbeing with a larger network of friends (r = .36, p <.0001). Moreover, the relationship is not affected by crystallised intelligence (vocabulary size, Li and Kanazawa’s original measures of intelligence) either: the partial correlation between network size and wellbeing was .33 (p <.0001) when vocabulary was controlled for.
Conclusions: While Li and Kanazawa used a test of crystallised intelligence (the PPVT), they interpreted their finding along the construct ‘general intelligence’. Our sample was selected with a test of fluid intelligence (the RAPM). While both the PPVT and the RAPM load highly on g, they arguably measure different broad abilities. In particular, while tests of crystallised intelligence (Gc) like the PPVT measure reasoning with already acquired knowledge, tests of fluid intelligence (Gf) tap the ability to solve problems in novel situations. Therefore, with the explanatory focus of the Savanna Theory of Happiness on dealing with novelty, the RAPM is arguably better suited to test the theory, and our results seem to contradict it.
