Short answer: the replication crisis affected psychology by showing that some published findings were less reliable than they appeared, especially when studies had small samples, flexible analyses, weak incentives for replication, or strong pressure to publish surprising results.
Psychology became one of the central fields in the replication debate because many of its studies involve human behaviour, which is variable, context-sensitive, and difficult to measure cleanly. A result found in one group of participants may not appear in another. A subtle experimental cue may matter more than expected. A statistical result may look convincing until someone tries to reproduce it.
What went wrong
Several problems came together. Small studies can overestimate effects. Selective reporting can hide null results. Researchers may try many analyses and report the most attractive one. Journals may prefer surprising findings over careful confirmations. The result is a literature that can look cleaner and more certain than the underlying evidence really is.
This does not mean psychology is fake. It means parts of the field had to become more honest about uncertainty, incentives, and method. The crisis was painful because it challenged published work, but it also pushed psychology toward stronger practices.
What changed
Many psychologists now use larger samples, preregistration, open data, registered reports, improved statistical practice, and multi-lab replication projects. These reforms do not solve every problem, but they make it harder for weak findings to become accepted simply because they are interesting.
For readers, the lesson is simple. Do not treat one psychological study as a law of human nature. Look for replications, effect sizes, transparent methods, and whether the claim has survived serious testing. The best psychology is not less scientific because it admits difficulty. It is more scientific because it confronts it.