Processing of pragmatic communication in ASD: a video-based brain imaging study

Aija Kotila, Aapo Hyvärinen, Leena Mäkinen, Eeva Leinonen, Tuula Hurtig, Hanna Ebeling, Vesa Korhonen, Vesa Kiviniemi, Soile LoukusaScientific 2020 · DOI

TL;DR

This study used fMRI to investigate how young adults with Autism Spectrum Disorder (ASD) process complex social communication compared to neurotypical controls. They found that while overall brain activity was similar, individuals with ASD showed differences in social brain network activity when processing situations with simultaneous verbal and non-verbal communication cues, suggesting a challenge in integrating multimodal information.

What Problem Does This Paper Address?

Social communication and pragmatic understanding are core areas of difficulty for individuals with Autism Spectrum Disorder (ASD). While these challenges are well-documented behaviorally, the underlying neural mechanisms are less understood, particularly in adults. Existing research often focuses on simpler stimuli like pictures or sentences, which may not fully capture the complexity of real-life social interactions. This study aims to bridge this gap by examining brain activity in response to naturalistic, pragmatically challenging video clips, specifically investigating how individuals with ASD process situations involving multiple, concurrent communication streams (verbal and non-verbal).

Methods

The researchers employed functional magnetic resonance imaging (fMRI) to measure brain activity in young adults with ASD and a neurotypical (NT) control group. They used a technique called Independent Component Analysis (ICA) to identify and analyze specific brain networks associated with social and pragmatic processing. Three key independent components (ICs) – IC4, IC15, and IC27 – were selected based on their known involvement in social cognition and pragmatic understanding. The fMRI data, collected while participants watched video clips depicting complex social scenarios, were analyzed by comparing the timecourses of activity within these selected brain networks between the ASD and NT groups. Statistical analysis, including permutation tests and FDR correction, was used to identify significant differences in brain activity. A qualitative analysis of the video content was also performed to understand the specific pragmatic challenges presented in clips where group differences were observed.

Datasets

The study utilized a dataset of 7 concatenated video clips from a Finnish TV soap opera called "Ruusun aika." These clips were specifically chosen for their naturalistic social communication content and the pragmatic inferences they required from viewers. The clips were pre-assessed by university students to identify their pragmatic features. Transcriptions of video clips 2 and 4, where significant group differences were found, are provided in the supplementary materials.

Key Insights and Ideas

  • Multimodal Integration is Key: The most significant finding is that differences in brain activity between ASD and NT groups emerged specifically when participants were exposed to situations with concurrent verbal and non-verbal communication. This highlights a potential difficulty in integrating information from multiple sensory and communicative channels simultaneously.
  • Social Brain Networks Under Scrutiny: The study successfully used ICA to isolate and analyze the activity of specific social brain networks (salience network and components of the default mode network) in response to complex social stimuli. This approach allows for a more nuanced understanding of how these networks function in individuals with ASD.
  • Adult ASD Processing is Complex: While many studies focus on children, this research demonstrates that pragmatic communication challenges can persist into adulthood and are reflected in distinct neural processing patterns, even in individuals with mild ASD.
  • Naturalistic Stimuli Reveal Differences: The use of realistic video clips, rather than simplified stimuli, was crucial in eliciting observable differences in brain activity, suggesting that more ecologically valid tasks are necessary to uncover subtle processing differences in ASD.

Key Results

  • Overall, the timecourses of the selected independent brain network components were largely similar between the ASD and NT groups across most of the video clips.
  • Statistically significant differences in brain activity were found in specific timepoints within video clips 2 and 4.
  • These significant differences were linked to episodes within these clips that involved two overlapping communication events: one verbal and one non-verbal (gesture/gaze-based), requiring attention shifting between them.
  • Video clips that did not feature such parallel communication events (e.g., one-to-one or one-to-many verbal communication on a single topic) did not elicit significant group differences.

Limitations and Future Work

  • Diagnosis Confirmation: The study acknowledges that participants diagnosed with ASD as children did not have their diagnosis re-confirmed in adulthood. It suggests that future studies should confirm ASD diagnoses in adulthood, as some individuals may no longer meet diagnostic criteria.
  • Sample Size: The study involved a relatively small sample size (19 ASD, 19 NT), and the authors recommend larger studies for greater statistical power.
  • Temporal Resolution: The use of BOLD signal in fMRI has inherent temporal limitations. Future research could benefit from faster fMRI techniques for more temporally accurate analysis of hemodynamic responses.
  • Naturalistic Settings: While the video clips were naturalistic, the study was conducted in a controlled lab setting. The authors suggest that more demanding, real-world scenarios, potentially using virtual reality (VR), could further illuminate pragmatic processing differences.
  • Intervention Design: The findings suggest that pragmatic challenges in ASD may be more pronounced in situations with high processing load due to overlapping cues. This understanding can inform the design of more targeted intervention programs for individuals with ASD.

Summary updated

Abstract

Social and pragmatic difficulties in autism spectrum disorder (ASD) are widely recognized, although their underlying neural level processing is not well understood. The aim of this study was to examine the activity of the brain network components linked to social and pragmatic understanding in order to reveal whether complex socio-pragmatic events evoke differences in brain activity between the ASD and control groups. Nineteen young adults (mean age 23.6 years) with ASD and 19 controls (mean age 22.7 years) were recruited for the study. The stimulus data consisted of video clips showing complex social events that demanded processing of pragmatic communication. In the analysis, the functional magnetic resonance imaging signal responses of the selected brain network components linked to social and pragmatic information processing were compared. Although the processing of the young adults with ASD was similar to that of the control group during the majority of the social scenes, differences between the groups were found in the activity of the social brain network components when the participants were observing situations with concurrent verbal and non-verbal communication events. The results suggest that the ASD group had challenges in processing concurrent multimodal cues in complex pragmatic communication situations.
BibTeX
@inproceedings{kotila2020processing,
  author = {Aija Kotila and Aapo Hyvärinen and Leena Mäkinen and Eeva Leinonen and Tuula Hurtig and Hanna Ebeling and Vesa Korhonen and Vesa Kiviniemi and Soile Loukusa},
  title = {{Processing of pragmatic communication in ASD: a video-based brain imaging study}},
  booktitle = {Scientific},
  year = {2020},
  doi = {10.1038/s41598-020-78874-2},
}
Figure