Dynamics of Brain Lateralization during Chinese Natural Speech Processing under the Influence of Sex Hormones: a 7T fMRI study

Ruohan Zhang, Shujie Geng, Xiaoqing Zheng, Wanwan Guo, Chun‐Yi Zac Lo, Jiaying Zhang, Chang Xiao, Xinran Wu, Jie Zhang, Miao Cao, Jianfeng FengbioRxiv 2023 · DOI

Abstract

Though language is considered unique to humans with left dominant lateralization in the brain, the dynamic nature of the interplay between hemispheres during language processing remains largely unknown. Here, we investigated whole-brain functional dynamic lateralization patterns during Chinese language processing and potential sex disparities using functional MRI data of 20 subjects listening to narrative stories in a 7T MRI scanner. Our findings revealed two distinct dynamic lateralization states, with regions of the language system consistently showing the left lateralization but reversed lateralization for other regions. These two states, characterized by higher-level functioning regions exhibiting left- or right-lateralization, corresponded to the processing of rational and emotional contents, respectively. We observed pronounced inclinations towards the former state in males and the latter state in females, especially during the processing of rational contents. Finally, genetic analyses revealed that the sex differences in lateralization states were potentially influenced by sex hormones.

Introduction

For decades, the textbook definition of language processing has been "left-brained." While generally true for basic linguistic tasks, this view is overly static. It fails to account for how the two hemispheres coordinate in real-time when we listen to complex, natural speech—such as a narrative story—rather than isolated words in a laboratory setting. Furthermore, while researchers have long noted sex differences in language tasks, the underlying neural dynamics and biological drivers of these differences have remained elusive. This study moves beyond "static maps" to understand the fluid, shifting nature of brain lateralization and how biological sex influences these transitions.

Methodology

The study utilized ultra-high-field (7 Tesla) imaging on 20 healthy, highly educated young adults to achieve superior spatial resolution while they listened to 162 minutes of Chinese narratives. To capture the brain in action, researchers employed several sophisticated techniques:

  • Dynamic Laterality Index (DLI): Rather than averaging brain activity over the whole scan, a "sliding window" approach was used to calculate how lateralization changed every few seconds.
  • Louvain Community Detection: This clustering algorithm identified recurring "states" of lateralization across the entire brain.
  • Natural Language Processing (NLP): To understand the triggers for state shifts, the team used NLP and manual sentiment analysis to categorize stories into "rational" (e.g., news, science) versus "emotional" (e.g., fairy tales, novels) content.
  • Gene Set Enrichment Analysis (GSEA): Findings were mapped onto the Allen Human Brain Atlas (AHBA) to determine if the brain regions involved were enriched with specific genetic markers.

To ensure the robustness of the findings, the researchers validated their results using resting-state fMRI data from 991 participants in the Human Connectome Project (HCP).

The "Stable" vs. the "Fluid" Architecture

The study revealed that the brain's language architecture consists of both stable and flexible components. The core language system, responsible for basic word and sentence processing, remains a "rock"—it stays left-lateralized regardless of the cognitive state. However, higher-level regions, such as the prefrontal cortex, act as "shifters," flipping their lateralization depending on the context.

The brain occupies two distinct configurations:

  • State 1 (Rational): Associated with rational content, showing stronger left-lateralization in higher-level regions.
  • State 2 (Emotional): Associated with emotional content, showing a shift toward right-lateralization in those same higher-level regions.

Key Results and Sex Disparities

The research uncovered significant differences in how males and females navigate these neural states:

  1. State Preference: Males showed a pronounced inclination toward State 1 (Rational), while females consistently preferred State 2 (Emotional). State 2 occurred more frequently overall (51.8% of the time) and had a longer "dwell time" compared to State 1.
  2. Content Sensitivity: These sex differences were most visible during the processing of rational content. When listening to rational stories, males stayed in the rational state longer, whereas females were more likely to transition into the emotional state.
  3. The Hormone Link: Genetic analysis provided a biological basis for these observations. State 1-related regions showed high expression of genes linked to male sex hormones, while State 2-related regions were linked to female sex hormones (and traits like menarche).
  4. Inherent Architecture: The states identified during speech processing were highly correlated (r=0.79) with states found in resting-state HCP data. This suggests that these dynamic configurations are fundamental properties of the human brain, which speech processing simply recruits or heightens.

Discussion and Limitations

These findings suggest that brain lateralization is a highly flexible process influenced by both the nature of the information being processed and the biological sex of the individual. The correlation with sex hormone-related genes suggests a molecular underpinning for these functional differences.

However, the study has limitations. The primary 7T study involved a small sample size of 20 participants, all of whom were young, highly educated Chinese speakers. Future work is needed to generalize these findings across different ages, education levels, and languages. Furthermore, while the genetic analysis shows a correlation with sex hormone markers, it does not prove causality. Longitudinal studies tracking hormone changes—such as those occurring during puberty or the menstrual cycle—would be a logical next step to confirm these biological drivers.

Summary updated

Abstract

Abstract Though language is considered unique to humans with left dominant lateralization in the brain, the dynamic nature of the interplay between hemispheres during language processing remains largely unknown. Here, we investigated whole-brain functional dynamic lateralization patterns during Chinese language processing and potential sex disparities using functional MRI data of 20 subjects listening to narrative stories in a 7T MRI scanner. Our findings revealed two distinct dynamic lateralization states, with regions of the language system consistently showing the left lateralization but reversed lateralization for other regions. These two states, characterized by higher-level functioning regions exhibiting left- or right-lateralization, corresponded to the processing of rational and emotional contents, respectively. We observed pronounced inclinations towards the former state in males and the latter state in females, especially during the processing of rational contents. Finally, genetic analyses revealed that the sex differences in lateralization states were potentially influenced by sex hormones.
BibTeX
@misc{zhang2023dynamics,
  author = {Ruohan Zhang and Shujie Geng and Xiaoqing Zheng and Wanwan Guo and Chun‐Yi Zac Lo and Jiaying Zhang and Chang Xiao and Xinran Wu and Jie Zhang and Miao Cao and Jianfeng Feng},
  title = {{Dynamics of Brain Lateralization during Chinese Natural Speech Processing under the Influence of Sex Hormones: a 7T fMRI study}},
  year = {2023},
  doi = {10.1101/2023.11.29.569155},
  howpublished = {bioRxiv},
}
Figure