[Do dissociations between implicit and explicit memory disappear, when reliability of the tests is comparable? An example].

Implicit memory measures tend to be less reliable than explicit memory measures. Consequently dissociations can arise as a consequence of differential reliabilities of the tasks. In the present study the effect of a manipulation of levels of processing in an implicit and an explicit memory task were investigated in a sample of 200 students after establishing comparable reliabilities of the memory tasks. A fragmented object-naming test was used as an indirect test and a recognition task was used as a direct test. The reliabilities of these measures were comparable in all conditions. A main effect of levels of processing, but no interaction between type of test and levels of processing, was found. These findings reinforce the necessity of taking into account the reliability of memory measures for the interpretation of dissociations between explicit and implicit memory.