My point of view is based on this very paywalled article.
In science a study doesn’t just become bunk just because its original conclusions no longer hold, or because a journalist says so.
The data is solid and is widely replicated. All the guardian article attacks is the analysis and conclusions.
And it is correct to do so, as that is how scientific discourse works. The journalist was however incorrect in saying it had been debunked.
Mischel is indeed still taught and students are taught to think critically about it.
Modern pedagogy uses it as a solid reference point both in terms of development and also in the limitations of empirical data.
Psychology is not like other sciences. It’s very nature is questioning established shibboleths and you’d be crazy to think that prevalent thought hadn’t moved on from a study done more than 50 years ago.
The guy who made the original hypotheses disagree with you. The issue is that data are not solid and are not widely replicated. And especially, the point is overstated.
The guy who made the original hypotheses also literally says that marshmallow test is bad proxy for the trait under question.
Groundbreaking research is groundbreaking research regardless of how well it holds up over time. Obviously it’s been supersceded but that just demonstrates the vitality of the field.