Social imaginaries on mental illness: a computational approach based on text mining

Resumen

This article presents research within an area of study between psychology and social communication.
The study approaches how the term “mental illness” is used in academic communication within a
dataset compared with other two online databases: one of newspaper articles and another of film
abstracts. More than 5000 abstracts extracted from those databases have been analyzed using
computational techniques with R programming: network relations, longitudinal analysis, correlations
calculus and sentiment analysis. We have been able to describe the social imaginaries present in
communication about mental illness in those datasets. Our findings revealed a significant gap between
the scientific standards and common view on mental health, somehow related with the stigma linked
to mental illness. Two social imaginaries of mental illness have been identified mining those three
datasets: the academical and the popular social imaginary of mental illness. Only by understanding
how complexity is simplified in social communication we would be able to manage better, not just the
suffering of living with a mental illness, but also the stigma surrounding mental illness.