A recent study by Goodrich-Hunsaker and Hopkins showed that amnesic patients with hippocampal damage performed above the recommended cutoff scores on immediate and delayed recognition of the Word Memory Test (WMT), but were significantly impaired on the multiple-choice, paired associate, and free-recall subtests.
The authors suggest that the hippocampal damage may be the culprit for such impairment. So how do they explain the above cutoff scores on the immediate and delayed recognition subtests? They suggest that the familiarity, simplicity, and repeated presentation of the items may have something to do with it.
A bit of an aside but related to my specific area of interest; I'd bet my meager stipend that if these patients were hooked up to an EEG machine during NREM sleep we'd observe a significant reduction in sleep spindle activity and SWS. Sleep spindles and SWS have been found to be associated with declarative memory consolidation so it'd only seem natural to see such a reduction.
Goodrich-Hunsaker NJ, & Hopkins RO (2009). Word memory test performance in amnesic patients with hippocampal damage. Neuropsychology, 23 (4), 529-34 PMID: 19586216