Leahy & Carey (2020)
The authors describe a study where pre-schoolers and chimpanzees perform poorly when needing to select the cup under which there is a certain reward as well as a study where pre-schoolers tended to perform poorly in a partial ignorance test, telling the researcher that they knew what object was under the box. This is taken as evidence for a lack of representation of uncertainty.
Is it possible that shooting a guess in these cases is not due to the lack of representation of uncertainty, but an attempt to reduce uncertainty by making choices that will result in more information about the environment? It seems counterintuitive that an ability to represent uncertainty observed in infants would disappear completely for a moment. Personally, I find it more plausible that pre-schoolers rather go through a developmental stage where they value innately more exploring the world and making actions to reduce uncertainty over certain rewards.
Stahl & Feigenson (2024)
Are there other experiments that can support the conclusion that children have different representations for improbability and impossibility? Though intuitively I agree with the statement, I don't think the methods are well thought through.
It is likely that if a never-before encountered object magically appears from the thin air, the object itself becomes more interesting. Therefore the children are more likely to remember the "Blick" in the impossible condition. If "Blick" is just a (albeit rare) object that the child saw in the gumball machine, it simply may not be as interesting, and its name not worth noting.