Love this but – a caveat that jumps out at me (that I’m sure is addressed in the paper) is: online.
online “Grouping Game”
There is a large amount of knowledge scaffolding needed to reach a point to being able to join in an _online_ grouping game.
But I haven’t read the paper and am judging on first blush but it’s a marker I’m putting on so I can reference it as I read to see if my initial concern holds or falls away.
“A total of 1480 subjects were recruited from Amazon Mechanical Turk”
49 countries can participate so the pool of 1480 is out of a potential total of 49 countries. Tests were n=2 (divergent) vs n=50 (convergent)
How diverse is the education in these countries (both formal and cultural) such that each of the 1480 participants had a profit motive to participate in the study?
Or in a more generic sense: Which countries might be considered “Westernized”?
US Minor Islands
US Virgin Islands
French Southern Territories
It is an fantastic study idea and I love the results regardless of which population might be used. But diverse populations may be an overstated feature of the study as I did not see a caveat mentioned in the paper. Maybe it’s such a common caveat that it was hiding in a word or a a checkbox in one of the supplementals.
If so, I’d like to know what it’s called so I don’t type all this kind of stuff again for a future study
Caveat about the diversity: it was done on Mechanical Turk, which automatically reduces pool to 49 countries. But it’s still fascinating even if it was done in a single country, giving names of blobs and that at samples of 50, common answers emerge. Here’s the countries MTurk operates in but because it was 1480 randomized from that pool there’s no knowing which. Are these mostly Westernized? Does it being MTurk users looking for money skew the results? You can decide. But it’s fascinating regardless.