What Can We Learn from the Past about the Future of Gerontology: Using Natural Language Processing to Examine the Field of Gerontology

Abstract

Objectives: We thematically classified all titles of eight top psychological and social gerontology journals over a period of six decades, between 1961 and February 2020. This was done in order to provide a broad overview of the main topics that interest the scientific community over time and place. Method: We used natural language processing in order to analyze the data. In order to capture the diverse thematic clusters covered by the journals, a cluster analysis, based on "topic detection"was conducted. Results: A total of 15,566 titles were classified into 38 thematic clusters. These clusters were then compared over time and geographic location. The majority of titles fell into a relatively small number of thematic clusters and a large number of thematic clusters were hardly addressed. The most frequently addressed thematic clusters were (a) Cognitive functioning, (b) Long-term care and formal care, (c) Emotional and personality functioning, (d) health, and (e) Family and informal care. The least frequently addressed thematic clusters were (a) Volunteering, (b) Sleep, (c) Addictions, (d) Suicide, and (e) Nutrition. There was limited variability over time and place with regard to the most frequently addressed themes. Discussion: Despite our focus on journals that specifically address psychological and social aspects of gerontology, the biomedicalization of the field is evident. The somewhat limited variability of themes over time and place is disconcerting as it potentially attests to slow progress and limited attention to contextual/societal variations.