Language abnormalities in Alzheimer's disease indicate reduced informativeness

Abstract

Objective: This study aims to elucidate the cognitive underpinnings of language abnormalities in Alzheimer's Disease (AD) using a computational cross-linguistic approach and ultimately enhance the understanding and diagnostic accuracy of the disease.

Methods: Computational analyses were conducted on language samples of 156 English and 50 Persian speakers, comprising both AD patients and healthy controls, to extract language indicators of AD. Furthermore, we introduced a machine learning-based metric, Language Informativeness Index (LII), to quantify empty speech.

Results: Despite considerable disparities in surface structures between the two languages, we observed consistency across language indicators of AD in both English and Persian. Notably, indicators of AD in English resulted in a classification accuracy of 90% in classifying AD in Persian. The substantial degree of transferability suggests that the language abnormalities of AD do not tightly link to the surface structures specific to English. Subsequently, we posited that these abnormalities stem from impairments in a more universal aspect of language production: the ability to generate informative messages independent of the language spoken. Consistent with this hypothesis, we found significant correlations between language indicators of AD and empty speech in both English and Persian.

Interpretation: The findings of this study suggest that language impairments in AD arise from a deficit in a universal aspect of message formation rather than from the breakdown of language-specific morphosyntactic structures. Beyond enhancing our understanding of the psycholinguistic deficits of AD, our approach fosters the development of diagnostic tools across various languages, enhancing health equity and biocultural diversity.

Figure 1

The artistic rending of the Indo‐European language family. Persian (Farsi) belongs to a distant typological branch relative to English and originates from the Indo‐Iranian branch.

Figure 2

Syntactic parsing for determining POS tags and dependency relations in English (A) and Persian (B) of a similar sentence. The specific word order of each language results in varying distances between certain syntactic relations, such as subject–verb relationship (nsubj).

Figure 3

The correlation coefficients of language features with an AD diagnosis. (A) The graph sorts the features based on the strength of correlation in English, with Persian features superimposed for direct comparison. (B) The graph sorts the features based on the strength of correlation in the Persian cohort. Redundant features, such as adjectives as parts of speech (POS) and dependency relations, are shown only once to avoid repetition.

Figure 4

(A) The reduction in LII as a function of excluding various content units in the Cookie Theft Picture. As each information unit is sequentially eliminated from the target text, its similarity to the reference text declines. (B) Box plots of LII in pwAD and healthy individuals (HC) in English and Persian.

Figure 5

Radar charts depicting correlation coefficients between Alzheimer's Disease (AD) and low informativeness (emptiness) across various linguistic features in English (A) and Persian (B). Substantial overlap exists between the two variables with respect to linguistic features in both languages. Asterisks (*) indicate the significance levels of the correlations with AD: *P < 0.05, **P < 0.01, and ***P < 0.001. Daggers (†) denote the significance levels of the correlations with low informativeness: †P < 0.05, ††P < 0.01, and †††P < 0.001.