Introduction Although evidence suggests physical activity (PA) may be associated with mental well-being at older ages, it is unclear whether some types of PA are more important than others. The purpose of this study is to investigate associations of monitored total PA under free-living conditions, self-reported leisure-time PA (LTPA), and walking for pleasure with mental well-being at age 60-64 years. Methods Data on 930 (47%) men and 1,046 (53%) women from the United Kingdom Medical Research Council (MRC) National Survey of Health and Development collected in 2006-2011 at age 60-64 were used in 2013-2014 to test the associations of PA (PA energy expenditure and time spent in different intensities of activity assessed using combined heart rate and acceleration monitors worn for 5 days, self-reported LTPA, and walking for pleasure) with the Warwick-Edinburgh Mental Well-being Scale (WEMWBS; range, 14-70). Results In linear regression models adjusted for gender, long-term limiting illness, smoking, employment, socioeconomic position, personality, and prior PA, those who walked for >1 hour/week had mean WEMWBS scores 1.47 (95% CI=0.60, 2.34) points higher than those who reported no walking. Those who participated in LTPA at least five times/month had WEMWBS scores 1.25 (95% CI=0.34, 2.16) points higher than those who did not engage in LTPA. There were no statistically significant associations between free-living PA and WEMWBS scores. Conclusions In adults aged 60-64 years, participation in self-selected activities such as LTPA and walking are positively related to mental well-being, whereas total levels of free-living PA are not.