Talks about _gestaltism_ as a "label applied to question-oriented psychology". Gestalt, as I understand it, is stuff that is irreducable. N&S put it in opposition to behaviorism, but behaviorists are sort of anti-reduction too, in their refusal to reduce the cognitive threads that tie together stimulus and response. Hmm. Maybe I have a futuristic (or wrong?) understanding of behaviorism.
Ho ho, how many times will I laugh at jovial jabs at the overuse of college sophomores in psychology experiments.
Conditionals are logic, I guess. But you can also look at them as sets of distinct scenarios.
Does forcing subjects to speak (symbolize) their thoughts alter the validity of the experiment?
Is speaking really symbol production? (p2012c3) Maybe we speak things to store them in short term memory.
N&S talk about how "true defense" of a theory lies in its "power to explain the behavior." We often separate behavior plausibility from biological plausibility. I wonder how important biological plausibility is in psychology.
Maybe what people do is try stuff randomly, see what works, associate it with salient parts of the percept (how does that happen?) and then statistically steer themselves towards approaches more likely to work.
They chose the data points that confirm their ideas. Here's an idea for a science journal: Orphan Data. It contains papers on the data left out by scientists.
The range of tasks they choose is {Solving logic problems,...,Writing computer programs}???? Not a terribly wide range.
To show that we use symbol manipulation in logic I'd think that you'd have to show that we use symbol manipulation in other places too. Otherwise, maybe you've just chosen a subset of human behavior that is explainable by your theory in a self-serving way. That's not great science. Maybe you haven't personally shifted the domain in a self-serving way, but possibly the field has as a whole.