There are two types of memory: explicit (declarative) memory, which concerns facts and experiences, can be recalled by conscious effort and reported verbally, whereas implicit (nondeclarative) memory, which involves skill learning, cannot. However, new findings by Park and colleagues indicate that this dichotomous distinction might be oversimplistic. In a double-blind study, they show that midazolam — a drug that was thought to impair only explicit memory — impairs implicit memory in healthy volunteers.

Early neuropsychological studies showed that patients with amnesia were impaired on explicit, but not implicit, memory tasks, which was thought to indicate that the two types of memory use different regions of the brain. However, this was called into question when later studies showed that some patients with amnesia also had deficits in a form of implicit memory known as contextual cuing. Contextual cuing is an implicit process that results in improved performance in spatial configuration tests, such as a visual search task, when display patterns are repeated.

Results from neuropsychological studies are unlikely to be conclusive because the patterns of brain damage between patients can vary considerably. In this new study, 27 healthy participants were tested for implicit memory using a visual task after they had been injected with either saline or midazolam. Each participant was tested under both conditions, so that each served as his or her own control. The participants performed better over time after both treatments, which is consistent with the literature on skill learning. The drug did not affect performance in the first trial, but the participants showed less facilitation for repeated displays in subsequent trials after injection with midazolam. They seemed to have difficulty learning the association between pieces of information, and so took a significantly longer time to identify the targets under midazolam when the repeated configurations cued the location of the targets.

Based on these findings, the authors suggest that conscious accessibility should not be the only criteria for defining different memory systems. Instead, the information processing requirements of a task, such as associative processing (or binding) of cues and context, might be an alternative framework for separating different types of memory performance. If memories require binding — whether explicit or implicit — then they are subject to amnesia.

Whether this new conception of memory is correct has yet to be determined. However, it is clear that the distinction between explicit and implicit memory is not as clear cut as it first seemed.