I'm still skeptical tbh. The scientific method in psychology seems like it could only ever work in a very probabilistic manner since, unlike physical phenomena,

I'm still skeptical tbh. The scientific method in psychology seems like it could only ever work in a very probabilistic manner since, unlike physical phenomena, humans predict very different to the same stimulus depending on their own unique circumstances. For example: People react very different to being offered a bacon pizza depending on (at least) the time (not for breakfast thanks), location (funerals are right out), their pizza topping preferences, their faith, whether they've already just eaten, how much they trust the person offering the pizza, if they're currently in a group or not (is there enough for everyone?), whether they've ever had any traumatic experiences with pizza or not, if any other foods may be available, whether they have dinner with other people planned, their current dietary restrictions, whether they're ill right now or not, whether they know you are doing pizza-based experiments on them, etc etc etc etc. All of these might radically alter the response you get. It might be possible to get enough information about someone to make a reasonable guess, but you can never know if your model is complete enough. Even if the same inputs occur twice, the output might still be different based on something you cannot reasonably know. So I think it would be very difficult to construct a predicable stimulus/response model for individuals. Groups might be easier because the differences sometimes average out, but quite often the internal communication leads to feedback loops that disturb any predictions you were trying to make about their behavior. Btw there is a third option beyond "scientific knowledge" and "faith", which is simply "we do not know and may never know" without ever getting to fill in that gap. Things like "what, if anything, created the universe" fall in that category as we cannot observe such a thing. Given some information about the position of a particle, we can never be sure of its speed per the Heisenberg uncertainty principle. Accurate models of people could be similar: it might simply be that they cannot be meaningfully reduced to simple formulas.