Language use in depressed and non-depressed mothers and their adolescent offspring

Laura A. Cariola, Lisa Sheeber, Nicholas B. Allen, Maneesh Bilalpur, Timothy Bird, Saurabh Hinduja, Louis–Philippe Morency, Jeffrey F. CohnJournal of Affective Disorders 2024 · DOI

TL;DR

This study reveals that the way mothers and their adolescent children communicate during problem-solving interactions can reveal differences associated with maternal depression. Specifically, linguistic patterns like pronoun use, negation, and temporal word usage differ between dyads with depressed mothers and those without, offering a novel way to understand the intergenerational impact of depression.

What Problem Does This Paper Address?

Maternal depression is a significant risk factor for offspring depression, yet the precise mechanisms through which this risk is transmitted are not fully understood. While previous research has explored linguistic markers of depression in written text, there's a notable gap in understanding how these linguistic patterns manifest in spoken, dyadic interactions between depressed mothers and their adolescent children. This study aims to bridge this gap by analyzing the language used during these interactions to see if it can differentiate between depressed and non-depressed mothers and their offspring.

Methods

The researchers employed a computer-assisted linguistic analysis tool called Linguistic Inquiry and Word Count (LIWC). This program analyzes text (in this case, transcribed speech) by categorizing words into various semantic and syntactic categories, such as pronoun use, emotion words, and temporal markers. They used LIWC to analyze transcripts of 151 low-income mother-adolescent dyads engaged in a 15-minute, lab-based problem-solving interaction task. Statistical analyses, including one-way multivariate analyses of variance (MANOVAs) and logistic regression, were used to identify significant differences in linguistic features between groups and to predict maternal depression status.

Datasets

The study utilized data from 151 low-income mother-adolescent dyads. Participants were recruited through the Oregon Health Plan and online advertisements. Mothers completed the Patient Health Questionnaire-9 (PHQ-9) to assess depressive symptoms. The data consists of transcribed audio recordings of mother-adolescent problem-solving interactions. No new benchmark datasets were introduced.

Key Insights and Ideas

  • Language as a Window into Depression: The study strongly suggests that spoken language during interactions is a sensitive indicator of maternal depression and its potential impact on offspring.
  • Beyond Written Text: This research extends previous findings by demonstrating the relevance of linguistic analysis in spoken, dyadic contexts, moving beyond social media or written tasks.
  • Intergenerational Linguistic Patterns: The findings hint at how maternal depression might influence offspring's communication styles, potentially through social learning or direct interactional effects.
  • The Nuance of Negation: While direct negative emotion words didn't show significant group differences, the increased use of negation words by depressed mothers was a key predictor, suggesting a more implicit way of expressing negative schemas.
  • Temporal Focus Shifts: Depressed mothers used more present-tense words and fewer past-tense words, which might reflect differences in emotion regulation strategies.

Key Results

  • Mothers' Speech: Depressed mothers used significantly more first-person singular pronouns, second-person pronouns, negation words, and present-tense words, while using fewer past-tense words compared to non-depressed mothers.
  • Offspring Speech: Higher-risk adolescents (offspring of depressed mothers) used more second-person pronouns, anger words, sadness words, and present-tense words, but fewer future-tense words than lower-risk adolescents.
  • Predictive Power: Linguistic features were able to classify dyads with depressed mothers with a reasonable degree of accuracy (around 69%). Specifically, negation words and reduced past-tense word use in mothers, and increased sadness words and reduced future-tense word use in adolescents, were significant predictors of maternal depression status.
  • Gender Differences: Daughters used more anger words than sons, and both daughters and sons of depressed mothers used more present-tense words and fewer future-tense words.

Limitations and Future Work

The authors acknowledge several limitations:

  • Causality: The study cannot definitively establish whether mothers' language behavior caused offspring's communication patterns or vice versa.
  • Generalizability: The sample was relatively homogeneous in terms of race and income, limiting the generalizability of findings to more diverse populations.
  • Context Specificity: The study focused on lab-based interactions; findings might differ in more naturalistic settings.
  • Intensity of Discussions: While home-based discussions showed some intensity differences, lab-based intensity ratings were not available, making it difficult to fully account for the impact of discussion intensity.

Future research directions include:

  • Replicating findings in more naturalistic settings and diverse samples.
  • Using time-series analysis to explore dynamic interactive processes in mother-offspring communication.
  • Investigating how maternal language influences offspring's communication more broadly, beyond dyadic interactions.
  • Exploring the mechanisms (e.g., emotion socialization, spillover effects) through which parental psychopathology influences parent-child interactions.

Summary updated

Abstract

The findings extend the existing literature on linguistic features of depression by indicating that mothers' depression is associated with linguistic behaviour during mother-adolescent interaction.
BibTeX
@article{cariola2024language,
  author = {Laura A. Cariola and Lisa Sheeber and Nicholas B. Allen and Maneesh Bilalpur and Timothy Bird and Saurabh Hinduja and Louis–Philippe Morency and Jeffrey F. Cohn},
  title = {{Language use in depressed and non-depressed mothers and their adolescent offspring}},
  journal = {Journal of Affective Disorders},
  year = {2024},
  doi = {10.1016/j.jad.2024.08.131},
}
Figure