Computerized Cognitive Intervention in Cognitively Normal Very Elderly Individuals

Background: The elderly population, and especially the oldest-old (those aged 85 and older) and old-old (those aged 75 and older), are the fastest growing segments of the U.S. population, increasing the need for disease-modifying treatments for Alzheimer’s disease (AD) and other age-related forms of cognitive decline. There is significant evidence that modifiable, nonpharmaceutical factors and interventions like cognitive activity and cognitive training may slow the course of AD and cognitive decline. However, little is understood about how cognitive training may translate into improved cognitive functioning, as a potential strategy for preventing decline. To the best our knowledge, this has never been studied in the very elderly. This study examined the effectiveness of a computerized cognitive training program (CCT program) CogniFit Personal Coach, and an active control games program (games program), in cognitively healthy individuals aged 80 and older. Three hypotheses were examined (1) compared to the games participants, CogniFit Personal Coach participants are expected to demonstrate greater positive change in overall cognitive function (a global cognitive composite) immediately following training; (2) compared to the games participants, CogniFit Personal Coach participants are expected to demonstrate greater positive change in the specific cognitive functions of memory, executive functioning/attention, and language, immediately following training; and (3) those with less education (as determined by a median split) will benefit more from participating in cognitive training, especially those using the CogniFit Personal Coach, compared to those with more education. Methods: Sixty-nine older adults were randomized to the CCT program (n=39) or an active control (games) program (n=30). Participants completed a baseline neuropsychological assessment, and were then asked to train for 20 minutes using their program every other day, for 24 total training sessions. After completion of training, participants again completed the neuropsychological assessment. The primary outcome measure consisted of a global cognitive composite, and the secondary outcomes were specific cognitive outcome measures (memory, executive functions/attention, and language), comprised of the means of Z-scores of their respective tests (follow-up scores use baseline coefficients to calculate Z-scores for the follow-up composite scores). Results: Linear mixed models demonstrated no significant interaction of program and time (from baseline to follow-up) on the global and specific cognitive composite scores reflecting that the two groups did not differ in the change of cognition from before to after treatment. Further, no significant main effects of time on overall cognitive functioning (the global cognitive composite) or on the specific cognitive domains were found for the overall sample, though scores in each factor did improve non-significantly, and memory improved at a trend level. Additional analysis found that those with less education (no college degree) were found to improve significantly on the global cognitive measure and language functioning compared to those with more education (college degree) in both the CCT and games programs. Discussion: This study demonstrates that there was no beneficial effect of the CCT program compared to the games program for overall cognition or specific cognitive domains in individuals aged 80 and older who are cognitively healthy. However, the findings do suggest that cognitive training of any kind, even at a less challenging level of an active control, may be beneficial for those without college degrees. The findings may demonstrate that improvement in cognition by an active cognitive training program for individuals who are cognitively normal might not be an effective strategy, or alternatively, that in the oldest old, perhaps due to less cognitive and brain reserve, CCT has no effect. A variety of personal, state, and training program variables likely influence the efficacy of cognitive training, and future research will be crucial to improve understanding of the relationships between cognitive training, cognitive functioning, and modifier factors