Dave Munger at Cognitive Daily has this post on faking amnesia and whether the act of faking actually does distort memory.

A common defense in murder cases is "focal retrograde amnesia": the defendant claims to have simply forgotten what occurred around the time of the crime (perhaps due having consumed too much alcohol or other drugs). In fact, "amnesia" is claimed in as many as 45 percent of murders. Psychologists know that this sort of amnesia is actually quite rare, so it's very likely that most, if not all of these defendants are faking amnesia.

*                *                *

But in testing for faking amnesia, a secondary phenomenon seemed to be appearing. Several studies found that people who were asked to pretend to have amnesia were later unable to recall as much as people who told the truth from the start.

*                *                *

There were a couple problems with these studies.... So a team led by Xue Sun designed a study to address these problems.

*                *                *

In other words, pretending to have amnesia doesn't hurt your memory, but rehearsing the correct answers improves it. There's still a separate question about how all this applies to the real world. A criminal probably does rehearse his or her alibi. How does this rehearsal affect memory for the actual crime? It's hard to say, and also difficult to envision a study that would address this question. Perhaps our readers have some ideas.