>hard sciences Yes, i have often said (including here on HN) that we should refer to them as the easy sciences, or less incendiary, the concrete sciences. The

>hard sciences Yes, i have often said (including here on HN) that we should refer to them as the easy sciences, or less incendiary, the concrete sciences. The question that lingers in my mind about the replication issue is perhaps the empirical process itself is ill suited to psychology. I think the replication crisis might be a crisis of process as much as it is of incentives. Empiricism seeks repeatable phenomena, same inputs, same outputs. In our psychological, virtual worlds, we have different inputs, different outputs, but are looking for similar processes. (Your example is a good one, in that the underlying process is a common experience but the lack of replicability of the study does not invalidate the process, just the input-ouput.) Maybe we need another model, or another layer, of empiricism in psychology.. I would characterise it as a sort of computational model. I am reminded of the study where the neurologists try to reverse engineer a computer program by looking at a computer.[0] I can't help but wonder how psychologists would approach the same problem. [0] cannot find it at the moment, but admittedly do like the analogy in that it is suggestive of the difficulty of understanding a complex system outside of its abstraction layer.