We investigate the effect of retirement on memory using the Survey on Health, Ageing and Retirement in Europe (SHARE). The availability of a panel data-set allows individual heterogeneity to be controlled for when estimating the effect of transitions into retirement on a commonly employed memory measure, word recall. We control for endogeneity of the retirement decision applying an instrumental variable technique to our fixed-effects transformation. Our main finding is that, conditional on the average non-linear memory age path of the typical individual, time spent in retirement has a positive effect on word recall.