We assume that the replication crisis in psychology mainly derives from poorly formulated study problems and hypotheses. We propose four variants of study problem’s formulation together with hypotheses aimed at solving these problems. (1) A paradox problem occurs when the indisputable empirical data contradicts the solid knowledge (logic, laws, etc.). The main way of resolution is to hypothesize a reason for such paradox. Empirical confirmation of such hypothesis might lead to the revision of existing knowledge. A few unresolved paradoxes in psychology are described. (2) A search problem occurs when a phenomenon is known to exist, but we do not know how to measure it and therefore its properties. Developing a method aimed at solving the problem can help. In this unit we consider direct and indirect ways of research. (3) A puzzle problem occurs when there is a need to incorporate an accidentally discovered and unexpected phenomenon into existing knowledge. Usually the solution of such problem broadens the existing knowledge. (4) A verification problem occurs when we need to verify if a phenomenon exists or this is just an artefact. Verification problem is also related to hypotheses’ validation. We recommend to clearly specify study problems and hypotheses while preparing a scientific paper.