Kazarovytska, F., Árnadóttir, K., D’Ottone, S. A., Halabi, S., Clarke, E. J. R., Sharma, S., Heidrich, V., & Imhoff, R. (in press). Do people across the world want to remember positive ingroup histories? Journal of Personality and Social Psychology. Preprint doi: doi.org/10.31234/osf...

Across seven countries, we find striking cross-national differences in what people want to preserve in collective memory. In some countries, primarily want to remember events in which the ingroup behaved morally. In other countries, there is a weaker preference for these ingroup-favoring histories.

So, the visually accessible MC ratio on gender cues is typically enhanced. For categories like race, an implication would be that an increased frequency of people visually read as "biracial", the salience of the category dimension might attenuate...

Yes! It may also be part of the puzzle why age is less reliably used. In the information ecology, it has no clear category boundaries and a flat meta contrast ratio -- unless it is institutionally created (e.g., in schools). On the contrary, gender is actively done and emphasized: costumes, make-up

Forgot to tag @chrispetsko.bsky.social - apologies!

Heidrich, V., Flade, F., & Imhoff, R. (in press). Face the difference: Meta-contrast as an affordance to spontaneous social categorization. Journal of Personality and Social Psychology.
osf.io/preprints/ps...

This idea started in my 2019 ESCON keynote (with Felicitas Flade) and came to life thanks to the incredible talent of Verena Heidrich who turned this vague idea into a paper I am really proud of.

Across 5 pre-registered experiments, we found:
People don’t just categorize more when meta-contrast is high —
They prefer to use the dimension with the stronger relative meta-contrast when several are available.

We call the clarity of those clusters the meta-contrast ratio (kind of like an ANOVA F value or Cohen's d).
It’s like how the same mean difference feels “bigger” when the groups themselves are less variable.
Low within-group variance = stronger affordance to categorize along that line.

Imagine the features that define a category (like skin color for race or skin texture for age). If these cues fall into two tight, non-overlapping clusters, categorizing is easy.
If they’re messy and (almost) overlapping, not so much.