Aerobic exercise improves episodic memory in late adulthood: a systematic review and meta-analysis

Abstract

Background Aerobic exercise remains one of the most promising approaches for enhancing cognitive function in late adulthood, yet its potential positive effects on episodic memory remain poorly understood and a matter of intense debate. Prior meta-analyses have reported minimal improvements in episodic memory following aerobic exercise but have been limited by restrictive inclusion criteria and infrequent examination of exercise parameters. Methods We conducted a meta-analysis of randomized controlled trials to determine if aerobic exercise influences episodic memory in late adulthood (M = 70.82 years) and examine possible moderators. Thirty-six studies met inclusion criteria, representing data from 2750 participants. Results Here we show that aerobic exercise interventions are effective at improving episodic memory (Hedges’g = 0.28; p = 0.002). Subgroup analyses revealed a moderating effect of age (p = 0.027), with a significant effect for studies with a mean age between 55–68 but not 69–85. Mixed-effects analyses demonstrated a positive effect on episodic memory among studies with a high percentage of females (65–100%), participants with normal cognition, studies reporting intensity, studies with a no-contact or nonaerobic physical activity control group, and studies prescribing >3900 total minutes of activity (range 540–8190 min). Conclusions Aerobic exercise positively influences episodic memory among adults ≥55 years without dementia, with larger effects observed among various sample and intervention characteristics—the clearest moderator being age. These results could have far-reaching clinical and public health relevance, highlighting aerobic exercise as an accessible, non-pharmaceutical intervention to improve episodic memory in late adulthood.

An umbrella review of randomized control trials on the effects of physical exercise on cognition

“…Among those that assessed it, 7 found evidence of publication bias (7 9,20–22,24,26,32 out of 14) and, among them, only 3 adjusted the final effect (3 9,20,32 out of 7). Notably, one of the meta-analyses concluded benefits in episodic memory with physical exercise even after the publication-bias correction methods indicated a non-significant effect 32 .…”

“…Twenty-four meta-analyses 12,7,13,14,8,[15][16][17][18][19][20][21][22]9,[23][24][25][26][27][28][29][30][31][32] meeting the inclusion criteria were selected among the 2,000 records retrieved in the search (see Fig. 1).…”

“…Almost all of the 24 meta-analyses found a significant positive effect of regular exercise on cognition in healthy participants (median Cohen's d = 0.29), concluding that exercise may improve cognitive skills (22 12,7,13,14,[15][16][17][18][19][20][21][22]9,[24][25][26][27][28][29][30][31][32] out of 24). To avoid the influence of using non-target exercise programs or samples with health conditions, we analyzed the outcomes of primary studies only with target interventions and healthy populations.…”

The Molecular Effects of Environmental Enrichment on Alzheimer’s Disease

“…A meta-analysis comprising of 28 studies (2156 participants) conducted by Zhidong and his team found that physical exercise could improve working memory in older individuals [ 220 ]. Aghjayan et al also conducted a meta-analysis investigating the effects of physical activities in older adults and found that aerobic exercises positively affect episodic memory as well [ 221 ]. Similarly, the same effects can also be observed in elderly patients with AD, with Jia et al reporting that physical activities contribute to their improvement in cognitive functions [ 222 ].…”

The Effect of Exercise Intensity on Affective and Repetition Priming in Middle-Aged Adults

“…Mixed-effects analyses demonstrated the positive effect on episodic memory among studies with a high percentage of females (65–100%) and participants with normal cognition. Subgroup analyses revealed the moderating effect of age ( p = 0.027), particularly studies on participants with a mean age between 55 and 68 years, as opposed to 69–85 [ 11 ].…”