Discovering Indicators of Dementia in Literary Texts


  • Neža Franca University of Ljubljana



Alzheimer’s dementia is one of the most common diseases troubling the aging population. Diagnosing it often includes methods strenuous for the individual [1] which calls for a more efficient and kinder diagnosis approach. Among several markers of cognitive decline due to AD are also different linguistic features that can be found years prior to disease onset. Discovering potential AD markers early on can contribute to disease delay or in the future even prevention. For this reason, exploring linguistic markers in literary texts is an idea worth exploring and can help develop a kinder diagnosis tool that could be used in the future also with non-writer populations.


In order to discover how linguistic markers of AD in literary texts change over time it is necessary to pick a prolific author suffering from dementia whose works span over the period of their entire life. These texts will then be scanned and converted to text. Using the texts, I will create a custom corpus that will enable me to search for specific linguistic markers of dementia within the writer’s works like the decreased size of the vocabulary, increase in repetitions, increase in the usage of filler words and indefinite nouns. Statistical analysis will be done using simple linear regression.

Expected Results

I expect results to show that individual linguistic markers that are typical of dementia patients will reveal themselves in the texts of the author I chose years prior to their official diagnosis thus confirming the results of Le et. al. which claim that linguistic markers of dementia can be found in literary texts years prior to disease onset.


Confirming that the linguistic markers of dementia are evident in text years prior to disease onset can lead to more research being focused on such cost effective and patient-friendly diagnostic tool development. More research in the field can also incentivize others to try finding linguistic cues in shorter texts belonging to non-writer populations and its application in clinical use.


[1] X. Le, I. Lancashire, G. Hirst, and R. Jokel, “Longitudinal detection of dementia through lexical and syntactic changes in writing: A case study of three British novelists,” Literary and Linguistic Computing, vol. 26, no. 4, pp. 435–461, May 2011. doi:10.1093/llc/fqr013