A methodology to assist in distinguishing deficient memory storage from compromised retrieval operations is presented for use with the Logical Memory (LM) and Visual Reproduction (VR) subtests of the Wechsler Memory Scale — Revised (WMS-R; Wechsler, 1987). A 21-item five-option multiple choice recognition test for LM and a cuing technique for VR are presented, followed by descriptive data based on a sample of 94 psychiatric inpatients (mixed diagnoses) and a sample of 99 brain-injured (BI) patients. The BI sample performed worse than the psychiatric sample on all of the LM and VR measures, with VR-CUED showing the highest degree of discrimination. In both groups, many subjects initially acquired much more information than they reported during the standard free recall format. The supplemental scoring format and reference data reported herein provide clinically useful information regarding a person's ability to store and access new information.