In this study memory change over 6 years was assessed using a large scale sample (16, 638 elderly individuals born <1948) from Health and Retirement Study. Growth curve models were analysed with reference to memory recall of a 10 word list and levels of social integration (i.e with family, volunteering, marital status).

Results indicated:

Socialisation demonstrated as a predictor of slower memory decline (p<.01). 

In individuals with vascular disease, socialisation observed to be protective buffer ( (p< 0.05)

Memory amongst least socialised deteriorated at twice the rate of other participants, with association greatest amongst those with <12 years of education (p<0.07)

The study indicates that socialisation and levels of education may be protective factors for memory decline. The study also suggests that socialisation may reduce risk for memory decline in individuals with vascular diseases. The study did not use a robust means of measuring memory capability, such as MMSE. Results for socialisation as a predictor of slower memory decline not statistically significant. However, socialisation in individuals with vascular disease as a protective factor was observed to be statistically significant. More research required as to the mechanisms of how socialisation reduces risk of memory decline in vascular disease. Findings for education as a protective factor were observed to be statistically significant. More research required into to what level of education is most protective against memory decline.

Abstract can be viewed here

Ertel, K. A., Glymour, M. M., & Berkman, L. F. (2008). Effects of social integration on preserving memory function in a nationally representative US elderly population. American journal of public health, 98(7), 1215–1220.