Processing False Beliefs in Preschool Children and Adults: Developing a Set of Custom Tasks to Test the Theory of Mind in Neuroimaging and Behavioral Research

The Neural Bases of Language Processing During Social and Non-Social Contexts: A fNIRS Study of Autistic and Neurotypical Preschool-Aged Children

Background

Little is known about how the brains of autistic children process language during real-world "social contexts," despite the fact that challenges with language, communication, and social interaction are core features of Autism Spectrum Disorder (ASD).

Methods

We investigated the neural bases of language processing during social and non-social contexts in a sample of N=20 autistic and N=20 neurotypical (NT) preschool-aged children, 3 to 6 years old. Functional near-infrared spectroscopy (fNIRS) was used to measure children's brain response to "live language" spoken by a live experimenter during an in-person social context (i.e., book reading), and "recorded language" played via an audio recording during a non-social context (i.e., screen time). We examined within-group and between-group differences in the strength and localization of brain response to live language and recorded language, as well as correlations between children's brain response and language skills measured by the Preschool Language Scales.

Results

In the NT group, brain response to live language was greater than brain response to recorded language in the right temporal parietal junction (TPJ). In the ASD group, the strength of brain response did not differ between conditions. The ASD group showed greater brain response to recorded language than the NT group in the right inferior and middle frontal gyrus (IMFG). Across groups, children's language skills were negatively associated with brain response to recorded language in the right IMFG, suggesting that processing recorded language required more cognitive effort for children with lower language skills. Children's language skills were also positively associated with the difference in brain response between conditions in the right TPJ, demonstrating that children who showed a greater difference in brain response to live language versus recorded language had higher language skills.

Limitations

Findings should be considered preliminary until they are replicated in a larger sample.

Conclusions

Findings suggest that the brains of NT children, but not autistic children, process language differently during social and non-social contexts. Individual differences in how the brain processes language during social and non-social contexts may help to explain why language skills are so variable across children with and without autism.

Theory of Mind and Parental Mental-State Talk in Children with CIs

Previous studies have suggested that parents may support the development of theory of mind (ToM) in their child by talking about mental states (mental state talk; MST). However, MST has not been sufficiently explored in deaf children with cochlear implants (CIs). This study investigated ToM and availability of parental MST in deaf children with CIs (n = 39, Mage = 62.92, SD = 15.23) in comparison with their peers with typical hearing (TH; n = 52, Mage = 52.48, SD = 1.07). MST was measured during shared storybook reading. Parents' narratives were coded for cognitive, emotional, literal, and non-mental references. ToM was measured with a parental questionnaire. Children with CIs had lower ToM scores than their peers with TH, and their parents used more literal references during shared storybook reading. There were no significant differences in the frequencies of cognitive and emotional references between groups. Parental emotional references contributed positively to children's ToM scores when controlling for the child's age and receptive grammar only in the CI group. These results indicated some distinctive features in parents of deaf children with CIs' MST and highlighted the role of MST in the development of ToM abilities in this group.

Relational vs representational social cognitive processing: a coordinate-based meta-analysis of neuroimaging data

The neurocognitive bases of social cognition have been framed in terms of representing others' actions through the mirror system and their mental states via the mentalizing network. Alongside representing another person's actions or mental states, however, social cognitive processing is also shaped by their (mis)match with one's own corresponding states. Here, we addressed the distinction between representing others' states through the action observation or mentalizing networks (i.e. representational processing) and detecting the extent to which such states align with one's own ones (i.e. relational processing, mediated by social conflict). We took a meta-analytic approach to unveil the neural bases of both relational and representational processing by focusing on previously reported brain activations from functional magnetic resonance imaging studies using false-belief and action observation tasks. Our findings suggest that relational processing for belief and action states involves, respectively, the left and right temporo-parietal junction, likely contributing to self-other differentiation. Moreover, distinct sectors of the posterior fronto-medial cortex support social conflict processing for belief and action, possibly through the inhibition of conflictual representations. These data might pave the way for further studies addressing social conflict as an important component of normal and pathological processing, and inform the design of rehabilitative treatments for social deficits.

False Belief Understanding in Deaf Children With Cochlear Implants

Theory of mind (ToM) is crucial for social interactions. Previous research has indicated that deaf and hard-of-hearing children born into hearing families (DoH) are at risk of delayed ToM development. However, it is unclear whether this is the case for DoH children who receive cochlear implants (CIs) before and around the second year of life. The present study aimed to investigate false belief understanding (FBU) in DoH children with CIs. The relationships between false belief task (FBT) performance, sentence comprehension, age at implantation, duration of CI use, and Speech Recognition Threshold were explored. A total of 94 children with typical levels of hearing (TH) and 45 DoH children (age range: 3-8), who received their first CI between 6 and 27 months of age, were tested on the FBT and a sentence comprehension test. Results showed that 4- and 5-year-old children with CIs performed significantly worse than their peers with TH on the FBT; 6- to 8-year-old children with CIs performed similarly to age-matched children with TH. Age at implantation and duration of CI use were correlated with sentence comprehension but not with the FBT. The results indicated that FBU was delayed until the age of 6 years in most of children with CIs.