There is a common assumption that accurate recognition exclusively reflects explicit-memory processing. However, Voss and Paller seem to have turned this notion on its head. In their study they found that implicit-memory processes can in fact guide responses in an explicit recognition test.
Kaleidoscope images were used to either divide or keep intact the human subject's attention. These same images were then discriminated from visually similar foils during forced-choice recognition testing.
They found higher accuracy for images studied with divided attention than for images studied with no attentional diversion. Surprisingly, "guess" responses were even more accurate than "know" responses. Moreover, metamemory was disrupted during the divided attention condition. ERPs of 200-400ms were observed for correct guesses which have been attributed to implicit-memory processing in past studies.
The authors suggest that implicit-memory processing to recognition may have been enhanced due to reduced potential for explicit-memory processing. They conclude by stating that their study "provides an unprecedented demonstration of the distinctive nature of overt recognition derived from unconscious memory, a phenomena that we describe as ‘implicit recognition.’"
Voss JL, & Paller KA (2009). An electrophysiological signature of unconscious recognition memory. Nature neuroscience, 12 (3), 349-55 PMID: 19198606