One of them is https://sci-hub.se/https://pubmed.ncbi.nlm.nih.gov/16123251/ https://sci-hub.se/https://pubmed.ncbi.nlm.nih.gov/16123251/ but there's a few more.

One of them is https://sci-hub.se/https://pubmed.ncbi.nlm.nih.gov/16123251/ https://sci-hub.se/https://pubmed.ncbi.nlm.nih.gov/16123251/ but there's a few more. Anyway, it seems obvious it's culture dependent. Given a person with usual western education and a person who's never dealt with shapes and abstract straight lines in their life - would you expect them to score similarly if they're both very intelligent people? These tests heavily rely on geometric puzzles which already make an assumption about education. The tricky question is how much does that affect differences you'd see between random people tested for comparison.