Beyond surface form: A pipeline for semantic analysis in Alzheimer's Disease detection from spontaneous speech

TL;DR

This paper introduces a novel pipeline that isolates semantic content in speech to improve Alzheimer's detection, demonstrating that models can identify semantic impairments even when surface features are altered or obscured.

Contribution

The study presents a new approach to analyze how semantic information influences language model performance in AD detection, including transformations that preserve semantics but alter surface form.

Findings

Models perform similarly with original and transformed texts, indicating reliance on semantic content.

Transformations that alter syntax and vocabulary significantly reduce surface similarity scores.

Semantic-based classifiers can detect AD even when surface linguistic features are obscured.

Abstract

Alzheimer's Disease (AD) is a progressive neurodegenerative condition that adversely affects cognitive abilities. Language-related changes can be automatically identified through the analysis of outputs from linguistic assessment tasks, such as picture description. Language models show promise as a basis for screening tools for AD, but their limited interpretability poses a challenge in distinguishing true linguistic markers of cognitive decline from surface-level textual patterns. To address this issue, we examine how surface form variation affects classification performance, with the goal of assessing the ability of language models to represent underlying semantic indicators. We introduce a novel approach where texts surface forms are transformed by altering syntax and vocabulary while preserving semantic content. The transformations significantly modify the structure and lexical content, as indicated by low BLEU and chrF scores, yet retain the underlying semantics, as reflected in high semantic similarity scores, isolating the effect of semantic information, and finding models perform similarly to if they were using the original text, with only small deviations in macro-F1. We also investigate whether language from picture descriptions retains enough detail to reconstruct the original image using generative models. We found that image-based transformations add substantial noise reducing classification accuracy. Our methodology provides a novel way of looking at what features influence model predictions, and allows the removal of possible spurious correlations. We find that just using semantic information, language model based classifiers can still detect AD. This work shows that difficult to detect semantic impairment can be identified, addressing an overlooked feature of linguistic deterioration, and opening new pathways for early detection systems.