Long-Term Effectiveness of Spaced-Retrieval Memory Training for Older Adults with Probable Alzheimer's Disease

ABSTRACT

Ten older adults with probable Alzheimer's disease (AD) were trained to recall everyday objects using the spaced-retrieval technique. Five persons had participated in a prior spaced-retrieval program (Cherry & Simmons-D'Gerolamo, 1999, Clinical Gerontologist, 20, 39–63). The authors retested these original participants to measure the long-term effectiveness of spaced-retrieval training. Five new participants were included for comparison purposes. During training, participants selected a target object from an array of items at increasingly longer retention intervals. To enhance the effectiveness of spaced-retrieval training, two original and two new participants were given a target object orientation task prior to training. Positive effects of prior experience occurred for most of the original participants. The orientation task enhanced memory for the target objects. Implications for memory remediation in late adulthood are considered.

The authors thank Diane Kelly, Director of St. Francis House Adult Day Care Center, a program of Our Lady of the Lake Regional Medical Center, for providing space for testing. The authors are also grateful to Verly Young for her help with participant recruitment. They thank two anonymous reviewers for helpful comments on an earlier version of this manuscript. Data for two standard procedure control participants (S1, S1-R, S2) and two target object orientation task participants (S7, S7-R, S8) were presented by the second author at the Tenth Annual Student Convention in Gerontology and Geriatrics in Savannah, GA (1999).

Notes

^a Mini-Mental State Examination (MMSE; Folstein, Folstein, & McHugh, 1975).

^b Geriatric Depression Scale (GDS; Sheikh & Yesavage, 1986).

^c Vocabulary Score (Short-Form of the WAIS Vocabulary test; Jastak & Jastak, 1965).

^d Forward Digit Span (FDS) and Backward Digit Span (BDS) from the Wechsler Adult Intelligence Scale (Wechsler, 1955).

^e Free and cued recall data were scored according to a strict (s) and lenient (l) criteria.

^f Comparison Memory Measures (taken from Cherry et al., 1999).

Digit span tests were scored by giving full credit for sequences where the two trials were correct and half credit if only one trial per sequence length was correct (an adaptation of the original scoring procedure for this test).

²Further prompting and partial credit for the name-tag task was as follows. If the participant failed to promptly return the name-tag after the first verbal cue (4 points), (a) the verbal cue was repeated (3 points); (b) the experimenter had touched her own name-tag and repeated the verbal cue a third time (2 points); or (c) the experimenter had touched her own name-tag, repeated the verbal cue a fourth time, and removed her own name-tag (1 point). If the name-tag was not returned after all prompts, 0 points were awarded (cf. Cherry et al., 1999).

Note. FT denotes number of failed recall trials; TT denotes total number of recall trials attempted; PF denotes the proportion of failed recall trials (i.e., FT/TT); LD denotes longest retention duration successfully achieved within each session (in seconds).

³The proportion of failed recall trials was calculated to provide an index of error on the task that is comparable across participants who vary in the total number of trials attempted. Inspection of Table 2 reveals that this proportional measure of performance essentially mirrors the absolute number of failed recall trials. For this reason, we have largely confined the discussion to comparisons involving the number of failed recall trials.

Note. FT denotes number of failed recall trails; TT denotes total number of recall trials attempted; LD denotes longest retention duration successfully achieved within each session (in seconds).

Note. Entries reflect pretest/posttest scores. Scores of 0 indicate that the participant did not name the target object. Scores of 1 indicate that the target object was named. Scores of 0/1* reflect implicit memory for the target objects.

⁴Note that S5 and S7 each named the target object in the category exemplar generation pretest in all three sessions. Consequently, there was no opportunity for these participants to demonstrate implicit retention at their original testing, as memory for the trained object cannot be inferred unequivocally when the target item has been named in the category exemplar generation pretest. Data in Table 4 should be interpreted in light of this constraint.

Note. Entries reflect free recall/recognition scores for the target objects. Scores for S1, S3, S5, S7, and S9 are based on a total of 3 days of training and 1 separate day for final recall/recognition. For S1-R, S2, S3-R, S4, S5-R, S6, S7-R, S8, S9-R, and S10, there were 6 days of training and 2 separate days of final recall/recognition (on Sessions 4 and 7).

^a Administered at the end of each training session (within-session measure of explicit retention).

^b Administered after a 48-h delay of testing (across-session measure of explicit retention).

^c Administered at the beginning of Day 6 (for Sessions 1–3) and on Day 9 (for Sessions 4–6). The retention intervals for the jacket, bracelet, and carrot, in order, were 3, 5, and 7 days for both Sessions 1–3 and Sessions 4–6.