Atypical perception in autism spectrum disorder: Combining computational models, functional neuroimaging and MR spectroscopy to understand aberrant perceptual mechanisms

• Applicants: Professorin Dr. Gabriele Ende; Professor Dominik M. Endres, Ph.D.; Professor Dr. Andreas Jansen; Professorin Dr. Inge Kamp-Becker
• Subject Area: Clinical Psychiatry, Psychotherapy, Child and Adolescent Psychiatry ; General, Cognitive and Mathematical Psychology ; Human Cognitive and Systems Neuroscience
• Term: from 2018 to 2023
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 417284407

• Final Report Year: 2024

Final Report Abstract

Autism spectrum disorder (ASD) is a severe mental developmental disorder characterized by weaknesses in social-communicative skills and the presence of restricted interests and repetitive behaviors. In recent years, the further development of the theory of the "Bayesian brain" has laid the foundations for explaining all core symptoms of ASD in a coherent model and describing their correlates at the neural and synaptic level. The proposed project aimed to investigate the computational and neural mechanisms underlying altered perception in patients with ASD. The project consisted of three work packages. In the first work package, we developed an imaging task that uses mathematical modeling methods to describe facial identity learning processes (an ability that is often limited in patients with ASD) and functional magnetic resonance imaging (fMRI) to depict the underlying neural activation patterns. In a sample of healthy subjects, we were able to show that brain activity in the superior temporal sulcus (STS) was associated with contextual familiarity, while activity in the fusiform face area (FFA) covaried with the prediction error parameter that updates facial familiarity. These results can now be used to explore aberrant neural and computational mechanisms in patients with ASD. In the second work package, we investigated the dimensional structure ("social communication deficits" and "repetitive behaviors") of the DSM-5 conceptualization of ASD in a large sample of patients with ASD (N>1600), as this 2D characterization has been shown in clinical practice to lead to (probably) unnecessary variability in diagnosis. Our results suggest that the communication deficit can be better described by an extended subspace with two dimensions ("basic social quality", "interaction quality"). In the third project, we searched for 'computational markers' of ASD. In particular, we tested whether the relationship between the assumed sensory noise and the certainty with which predictions are made regarding future environmental states (e.g. presented sensory stimuli) is not correctly adjusted. For this purpose, a hierarchical internal model based on the Hierarchical Gaussian Filter was used. In a relatively small sample (ASD N=22, non-ASD N=11), tendencies but no reliable differences in volatility and learning rate of the HGFs fitted to the individual behavioral data could be found in a Wisconsin Card Sorting Test.

Publications

• Modeling aberrant volatility estimates in Autism Spectrum Disorder. Proceedings of the Annual Meeting of the Cognitive Science Society, 44.
  Niehaus, H., Stroth, S., Kamp-Becker, I. & Endres, D.
  
• Subdimensions of social‐communication behavior in autism—A replication study. JCPP Advances, 2(2).
  Stroth, Sanna; Niehaus, Hauke; Wolff, Nicole; Poustka, Luise; Roessner, Veit; Kamp‐Becker, Inge & Endres, Dominik
  
• Modeling face recognition in the predictive coding framework: A combined computational modeling and functional imaging study. Cortex, 168, 203-225.
  Zaragoza-Jimenez, Nestor; Niehaus, Hauke; Thome, Ina; Vogelbacher, Christoph; Ende, Gabriele; Kamp-Becker, Inge; Endres, Dominik & Jansen, Andreas