The survey completely overlooks a very important possibility: The participants know their own ignorance of the exact circumstances of each situation, and so th

The survey completely overlooks a very important possibility: The participants know their own ignorance of the exact circumstances of each situation, and so they use their judgement of the parent's character to fill in the gaps. For example, when the parent leaves for an "immoral" reason, it reflects badly on their character, so the participant would have less reason to believe that they've thought about the child's safety. On the other hand, when a parent whom the participant has judged to be of good character leaves the child, they assume that the parent them-self has thought through the safety implications and has judged it to be safe. Since the parent would have more information than the participant, the participant is trusting their judgement, and is able to determine that the situation is safer. In essence, the stories told to the participants actually do have different amounts of information in about the child's safety.