You can use existing error catalogs or develop your own.
To use an existing error catalog:
Find a defect in the list
Ask whether the software under test could have this defect
If it's theoritically possible the program could have the defect, ask how you could find the bug
Ask how plausible it is that the bug could be in the program and then how serious the failure would be if it was there
If appropropriate design a test or series of tests for bugs of this type