6 tweets · March 2022 · 15 likes · 0 retweets · read on Twitter
the marshmallow test is the simplest case study on Moloch—can you not defect against different selves across time? (in the context of "if I don't do it, someone else will, and if they do and I don't, I lose" which is the molochian trap) (—paraphrasing @jgreenhall)
<insert necessary caveat re marshmallow test being complex in terms of what it's actually measuring, especially if applied statistically> hotelconcierge.tumblr.com/post/113360634…
but if we screen off The Desire To Pass Tests and look at situations that involve intrinsic structures of delayed gratification or something we might find that the marshmallow test measures something more like "self-trust" than like "will"
although ofc maybe the actual strategies kids use to succeed at it aren't self-trust-based even though I think that practicing such things would probably cause increased self-trust, at least narrowly, even if the strategy used is distraction
unless you end up generating opponent processes which mess you up later the advantage of the marshmallow test is that the reward and the cost are in the same units: marshmallow so the only tradeoff is time/selves, not some messier thing that can get out of balance
right yeah sorry it's not just The Desire To Pass Tests—the marshmallow test also measures trust in adults/etc, to some degree but yeah, low external trust scenarios make internal trust harder, perhaps!