Predictive Coding in ASD and ADHD: Modulation of P300 by Sequence Learning
Poster No:
353
Submission Type:
Abstract Submission
Authors:
Yiyuan Huang1, Maria Gonzalez-Gadea2, Tristan Bekinschtein3, Agustin Ibanez4, Zenas Chao1
Institutions:
1The University of Tokyo, Tokyo, Japan, 2University of San Andres, Buenos Aires, Argentina, 3University of Cambridge, Cambridge, United Kingdom, 4Global Brain Health Institute, University of California, CA, United States
First Author:
Co-Author(s):
Introduction:
Neural signatures of prediction errors elicited by unexpected stimuli have been extensively studied by mismatch negativity (MMN) and P300 responses in electroencephalography (EEG). These responses can be modulated by sequential contexts, such as attention allocation and sequence learning, and atypical modulations were found in autism spectrum disorder (ASD) and attention deficit hyperactivity disorder (ADHD). For attention allocation, the P300 response was more pronounced for unexpected tones that were attended to compared to those that were not, and the difference was observed in both healthy controls (HC) and ASD, but not in ADHD. Therefore, P300 reveals a lack of neural distinction between attended and unattended novel stimuli in ADHD, and could be an effective biomarker. For sequence learning, MMN reduced when an unexpected tone became predictable as multi-tone sequence structure is learned, but this reduction was weaker in ASD than in HC. However, it remains unclear how P300 is modulated by sequence learning and how such modulation may vary among HC, ASD, and ADHD.
Methods:
In this study, we analyze EEG data from 13 HC (8 males; mean age: 11.6 ± 2.7 years old), 21 individuals with ASD (20 males; mean age: 10 ± 2 years old), and 13 individuals with ADHD (10 males; mean age: 12 ± 3 years old) during a local-global oddball paradigm. In this paradigm, two 5-tone sequences were used: xxxxy and xxxxx. The tone "y" served as a local deviant tone within the sequence which is more predictable at the global level when xxxxy was presented frequently (the highPred condition) and less predictable when xxxxy sequence was presented infrequently (the lowPred condition). To examine how P300 evoked by the local deviant tone "y" is modulated by the global sequence predictability, we compare event-related potentials (ERPs) between the two sequences (xxxxy – xxxxx) in both conditions (Figure panel A).
Results:
Using a clustered-wised statistical analysis on P300, we show that HC has a positive response in the central area in the lowPred condition but a negative response in the frontal areas in the highPred condition. In ASD and ADHD, the response is positive in the parietal-occipital area in the lowPred condition but is negative in the frontal area in the highPred condition. We further evaluate the difference in P300 between the lowPred and highPred conditions. For each individual, we first identify the peak P300 amplitude in each condition from all channels and time points where P300 is significantly different from zero, and then evaluate its reduction from lowPred to highPred conditions in the three groups (Figure panel B). All three groups show a significant reduction in P300 when the sequence is predictable, and this reduction is significantly less in ASD, compared to HC and ADHD (Figure panel C).
Conclusions:
Our study, which builds upon prior research, demonstrates that P300 can be used to evaluate distinct predictive coding elements across disorders. Specifically, previous work showed that ADHD is associated with impairments in attention allocation, consistent with atypical prediction precision in predictive coding theory. On the other hand, the current investigation shows that ASD is linked to challenges in sequence learning, suggesting atypical prediction updates.
Disorders of the Nervous System:
Neurodevelopmental/ Early Life (eg. ADHD, autism) 1
Novel Imaging Acquisition Methods:
EEG
Perception, Attention and Motor Behavior:
Perception: Auditory/ Vestibular 2
Keywords:
Attention Deficit Disorder
Autism
Electroencephaolography (EEG)
Other - Sequence learning
1|2Indicates the priority used for review
Provide references using author date format
Gonzalez-Gadea, Maria Luz, et al. (2015) "Predictive coding in autism spectrum disorder and attention deficit hyperactivity disorder." Journal of Neurophysiology 114.5 2625-2636
Goris, Judith, et al. (2018) "Sensory prediction errors are less modulated by global context in autism spectrum disorder." Biological Psychiatry: Cognitive Neuroscience and Neuroimaging 3.8 667-674.