Processing of Emotional Words in Children and Adolescents with Autism Spectrum Disorders.
Autistic tweens' brains respond weakly to positive emotional words, and the gap tracks with social struggles—so check valence when teaching feeling words.
01Research in Context
What this study did
Yeh et al. (2026) watched tweens read emotional words while wearing EEG caps. The team compared autistic and neurotypical kids aged 11-14. They wanted to see if the brain's late positivity wave differed between groups.
Late positivity is a small EEG bump that shows deeper word processing. The task was simple: read happy, sad, or neutral words on a screen. No teaching or therapy happened; it was pure lab science.
What they found
Both groups read the words equally fast, but their brains acted differently. Autistic tweens showed a weaker late positivity only for positive words like 'joy' or 'smile'. The smaller the positivity, the bigger the child's social deficit score.
Negative and neutral words looked the same in both groups. The result points to a valence-specific glitch, not a global language problem.
How this fits with other research
Lartseva et al. (2014) saw a similar dip in late positivity, but in adults with ASD. The new study extends the finding down to middle-schoolers, showing the pattern starts early.
Fink et al. (2014) looks like a contradiction: after controlling for verbal IQ, autistic kids had no trouble recognizing facial emotions. The key difference is modality—faces versus words—and outcome level—behavior versus brain waves. Behavior can look fine while EEG still flags a difference.
Ganz et al. (2009) adds another puzzle piece: emotional words failed to grab attention in ASD adults. Together, the studies suggest emotional language processing is subtly off across ages, even when scores look normal.
Why it matters
When you teach feeling vocabulary, do not assume 'happy' is as clear to the learner as 'sad'. Probe comprehension of positive words separately and pair them with extra examples or visuals. If social skills are lagging, check whether the child truly grasps positive emotion terms before moving to complex social scripts.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →During feeling-word drills, present 'happy' and 'excited' twice as often as 'sad' or 'angry' and ask the learner to give a real-life example each time.
02At a glance
03Original abstract
Individuals with autism spectrum disorder (ASD) have difficulties in understanding emotional language, but little research has discussed the developmental course of the processing of emotional words in the clinical population. Previous studies have revealed distinct processing for emotion-label (e.g., happiness) and emotion-laden (e.g., birthday) words in typically developing (TD) children and adolescents. Extending these findings, the study used event-related potentials (ERPs) to explore the processing of these two types of emotional words in children and adolescents with ASD. The stimuli included two-character Chinese words with factors of word type (emotion-label versus emotion-laden) and valence (positive versus negative). The participants were 11 to 14-year-old children and adolescents with ASD (N = 23) and age-matched TD peers (N = 23). They categorized emotion valence for words while their brain responses were recorded. Both the TD and the ASD groups exhibited emotional processing for all emotional words across the N400 and late positivity component (LPC). The emotional processing was modulated by word type but varied with group and valence. A trend for group differences was observed in processing positive words at 500-600 ms. In particular, the emotion effects of positive emotion-label words were positively correlated with social dysfunction across all participants. These findings suggested that children and adolescents with ASD have a selective impairment in understanding emotional concepts from language. The ERP measurements may reflect atypical emotional word processing for individuals with higher autistic severity in positive valence contexts.
Journal of autism and developmental disorders, 2026 · doi:10.3389/fnhum.2017.00455