Multi-Level Narrative Evaluation Outperforms Lexical Features for Mental Health

TL;DR

This study introduces a three-level narrative evaluation framework that significantly improves mental health prediction from therapeutic texts, emphasizing macro-structural organization over lexical features.

Contribution

The paper presents a hierarchical framework combining lexical, semantic, and macro-level narrative evaluation, demonstrating the importance of narrative structure in mental health assessment.

Findings

Macro-level evaluation outperforms lexical and embedding features in predicting mental health.

Formal narrative structures carry significant predictive signals for psychological states.

Semantic embeddings provide minimal independent value but enhance multi-level classification.

Abstract

How people narrate their experiences offers a window into how the mind organizes them. Computational approaches to therapeutic writing have evolved from lexical counting to neural methods, yet remain fragmented: dictionary tools miss discourse structure, while embeddings conflate local coherence with global organization. No existing framework maps these techniques onto the hierarchical processes through which narratives are constructed. Here we introduce a three-level framework - micro-level lexical features, meso-level semantic embeddings, and macro-level LLM narrative evaluation - and show, across 830 Chinese therapeutic texts spanning depression, anxiety, and trauma, that macro-level evaluation substantially outperforms lexical and embedding features for mental health prediction. This challenges the field's emphasis on word-counting: formal structural features (Labov's story grammar, RST coherence, propositional composition) demonstrate that narrative organization per se carries predictive signal, while clinically-grounded narrative dimensions capture how psychological states are expressed through discourse. Semantic embeddings add minimal independent value but yield incremental gains in multi-level classification. By grounding computational levels in discourse processing theory, this framework identifies macro-structural organization as the primary locus of clinical signal and generates testable hypotheses for intervention design and longitudinal research.