The cultural construction of "executive function"

IQ Research
1

Cross-cultural research in intelligence can get very complicated. One challenge is that basic tasks used to measure cognition are often not as universal as they may seem to people in Western countries. A new article in PNAS News explores this.

Article 1
Article 1890×875 206 KB

The authors administered executive functioning (EF) tasks to four samples of children, ages 3-18: British children, Kunene children (in Angola and Namibia) in school and those with little contact in school, and Tsiname children in the Bolivian rainforest whose schooling is very ineffective. The different cultural groups, levels of education, and ages will make it easier for any differences to detect.

The results showed strong evidence that EF tasks are not as universal in their development and age progression as many psychologists believed. A good example is the Dimensional Change Card Sort task, which asks children to sort cards based on one characteristic (e.g., color of objects on the card) and then to shift to sorting cards based on a different characteristic (e.g., number of objects on the card). Almost every British child could do this from a young age, but the Tsiname and unschooled Kunene children struggled much more with the task. What is most interesting is that the Kunene children with exposure to school did about as poorly as the other non-British children at age 5, but improved on the task until age 10, when they performed it as well or better than British children.

Article 2
Article 2806×788 82.6 KB

On a verbal fluency task, the major difference was between British and non-British children. Starting at age 6, British children could name more objects in a given category (e.g., animals) in 2 minutes than the Tsiname or Kunene children. Still, all three groups show improvement in this task as they age.

Article 3
Article 31102×567 55.9 KB

Another interesting result happened when children were administered a task called Luria’s game in which they are taught two simple hand gestures. After they learn to imitate the gestures, children are asked to make the opposite gesture in response to the gesture the adult makes. Again, this task was far easier for British children than the other groups (although the Tsiname children performed as well as the schooled Kunene group).

What is most interesting for intelligence researchers is the result of the forward and backward digit span tasks, which often appear on intelligence tests. On the forward digit span tasks, very few of the non-British children could ever recall in order more than 4 single-digit numbers spoken to them. Backward digit span was even more difficult, some children failed the task completely (even when asked to recall only 2 digits in reverse order).

Article 4
Article 41531×701 75.8 KB

These results show that cognitive development can have different trajectories in different cultures and environments. Based on this one study, it is not possible to say why these differences develop. But it does show that tasks developed in Western contexts that value cognitive “games” and rules may not be intuitive to people in other parts of the world. Using such tasks in cross-cultural research demands caution.

Original post: https://x.com/RiotIQ/status/1946662827168780776?s=20

Link to paper: https://www.pnas.org/doi/epdf/10.1073/pnas.2407955122

2

The digit span results are particularly striking for IQ testing because forward and backward digit span appear on the WAIS and other major tests. If non-British children can’t even do forward span beyond 4 digits regardless of age, that suggests the task is measuring cultural familiarity with arbitrary number games rather than pure working memory capacity. The DCCS findings are also important, showing that cognitive flexibility as measured by Western tasks develops completely differently depending on schooling exposure. This doesn’t mean these kids lack executive function, it means our tests measure culturally specific skills developed through formal education rather than universal cognitive capacities.

3

What’s fascinating is that schooled Kunene children eventually caught up to British kids on DCCS by age 10, showing these skills can be learned but aren’t naturally developing milestones. The verbal fluency gap makes sense given British kids are exposed to far more vocabulary and categorization exercises through books and school. For intelligence testing, this is a huge problem because tests like the WAIS include digit span and other tasks that clearly advantage people from schooled, Western contexts. We’ve been calling these “culture fair” measures when they’re actually heavily culture loaded. This study basically shows that what we measure as executive function is partly universal capacity and partly learned cognitive strategies from formal education.