Medial Prefrontal Cortex Brain Magnetic Stimulation in Autism

The Reading the Mind in the Eyes Test (RMET; Baron-Cohen et al., 2001) will be used to assess recognition of emotional and mental states from facial cues. First, participants will complete the Basic Emotion Recognition Task, identifying six basic emotions-happiness, sadness, anger, fear, disgust, and surprise-from whole-face photographs using a two-choice format. They will then complete the standard RMET, which contains 36 photographs of the eye region depicting complex mental states. For each image, participants select the most appropriate of four descriptors. To examine possible cultural differences, the Taiwanese RMET (TW-RMET) is administered alongside the original UK version. The TW-RMET includes 43 culturally adapted items and has been validated in Taiwanese populations (Li et al., 2022). Higher scores indicates a stronger ability to accurately interpret emotional and mental cues from facial expressions.

The SRS (Constantino, 2002) is a 65-item rating scale assessing the severity of ASD symptoms in natural settings. It provides clinical description of an individual's social impairments, covering social awareness, social information processing, capacity for reciprocal social communication, social avoidance, and autistic mannerisms. Higher scores on the SRS total scores reflect greater impairment and autistic symptoms with scores≧60 typically considered within above the clinical threshold. The Mandarin Chinese version of the SRS has demonstrated strong psychometric properties, with satisfactory four-factor structure and high internal consistency (Cronbach's alpha, .94-.95), i.e., social communication, autistic mannerism, social awareness, and social emotion (Gau et al., 2013). In this project, we use the SRS to measure social communication deficits and stereotyped behaviors.

The Autism-Spectrum Quotient (AQ; Baron-Cohen et al., 2001) is a 50-item self- or parent-report measure of autistic traits across social skills, attention switching, attention to detail, communication, and imagination. Items use a four-point response scale, with higher scores indicating more autistic traits. The AQ has good internal consistency (α=.82), test-retest reliability (.70), and screening validity, with a clinical threshold of 26 (Woodbury-Smith et al., 2005). The Mandarin Chinese AQ uses a 1-4 Likert scale and has been validated in Taiwanese adults (Lau et al., 2013). It shows good total-score internal consistency (α=.836), moderate-to-good test-retest reliability (ICC=.647), and somewhat different factor structure from the original version.

The RAADS-R (Ritvo et al., 2011) is a self-report instrument designed to assist in the identification of adult autistics older than 18 years, particularly those who may have escaped earlier diagnosis due to subclinical symptom presentation. The RAADS-R is an updated version of the original Ritvo Autism Asperger Diagnostic Scale (RAADS) (Ritvo et al., 2008). Questions on the initial RAADS assess developmental pathology in three symptom domains: language, social relatedness, and sensory-motor (of the DSM-IV-TR). The revised 80-item RAADS-R includes the addition of a fourth domain-circumscribed interests, as well as two new items and multiple wording clarifications based on clinical feedback and factor analytic findings. Subsequent factor analysis of the RAADS-R supported a four-factor structure, comprising: social-relatedness, circumscribed interests, sensory-motor behaviors, and social anxiety. Scores above 65 strongly suggest the presence of autistic traits.

The EQ (Baron-Cohen & Wheelwright, 2004) is a widely used self-report measure designed to assess cognitive and affective components of empathy. Empirical evidence has demonstrated that approximately 81% of autistic individuals score below 30 on the EQ, whereas only 12% of neurotypical controls fall within this range, indicating its clinical utility in distinguishing empathic traits in ASD populations. The EQ demonstrates excellent psychometric properties, with an internal consistency of α = .92 and a test-retest reliability of r = .97. The Mandarin Chinese version of the EQ, developed and validated by Huang H.-Y. and Gau S.S.-F. (unpublished), has shown satisfactory reliability and validity and will be utilized in the current study to assess empathic functioning in autistic participants. Higher scores represent higher empathy levels, while lower indicate social and communication obstacles.

The AASP (Catana E. Brown, 2002) is a 60-item self-rating scale designed to examine sensory-related features, including preference, emotional reactions, or behavior changes prompted by perceptions. It measures six modalities: taste/smell, movement, visual, touch, activity, and auditory processing. It adopts a 5-point Linkert rating scale with '1' for never (0%), '2' for seldom (25%), '3' for sometimes (50%), '4' for frequently (75%), and '5' for always (100%). Aligned with Dunn's 4-dimension model of sensory functioning (Dunn, 1997; Dunn & Brown, 1997), the scoring system includes Low Registration, Sensation Seeking, Sensory Sensitivity, and Sensation Avoiding subscales, where higher scores indicate greater sensory reactivity. The Mandarin version of the AASP has been validated in Taiwanese populations, showing acceptable internal consistency (Cronbach's α = 0.71-0.80) and strong test-retest reliability (intraclass correlation coefficients = 0.80-0.86) (Tseng MH, 2009).

The 10-item short SPQ, SPQ-10 (Greenberg et al., 2018) is a brief, self-report instrument designed to assess sensory hypersensitivity across five sensory modalities: vision, hearing, touch, taste, and smell. The scale includes two items per modality, and each item is rated on a 4-point Likert scale ranging from 0 (strongly disagree) to 3 (strongly agree). The total score, obtained by summing responses to all items, provides an overall index of sensory hypersensitivity, with higher scores indicating greater sensitivity. The SPQ-10 has demonstrated strong convergent validity, with high negative correlations reported with both the full-length SPQ (r = -0.828, p < 0.01) and the 35-item short version (r = -0.920, p < 0.01) (Greenberg et al., 2018), supporting its construct validity and efficiency as a screening tool. The Mandarin Chinese has been validated by our research team (paper in press).

The SEIS is a 33-item self-report inventory designed to assess various components of emotional intelligence. Factor analytic studies have supported a four-factor structure for the SEIS encompassing the following domains: (Ciarrochi et al., 2001; Petrides & Furnham, 2000; Saklofske et al., 2003), including perception of emotions, managing emotions in the self, social skills or managing others' emotions, and utilizing emotions. The SEIS was translated into Mandarin Chinese in a previous project conducted by our team, and its psychometric properties have been validated in a Taiwanese sample (Chien et al., 2023).

This 20-item self-report scale measures social anxiety experienced in social situations based on DSM-IV criteria of social anxiety disorder. Respondents rate each item on a 5-point Likert scale, ranging from 0 (not at all characteristic of me) to 4 (extremely characteristic of me), yielding a total score between 0 and 80, with higher scores indicating greater levels of social anxiety. The Chinese version of the SIAS (Yang, 2003), has demonstrated adequate psychometric properties, including good internal consistency and construct validity. It has been widely used in both clinical and community samples in Chinese-speaking populations. In this project, the SIAS will be employed as a measure of social anxiety severity to evaluate potential changes before and after intervention.

The BDI (Beck et al., 1961) is a 21-item self-report instrument measuring the severity of depression in adolescents and adults. The BDI-II (Beck et al., 1996) was revised in 1996 to be more consistent with the DSM-IV criteria for major depressive disorder. Each item is rated on a 4-point Likert scale ranging from 0 (not at all) to 3 (severe), producing a total score from 0 to 63, with higher scores reflecting greater depressive symptomatology. A Mandarin Chinese version of the BDI was translated and validated by Zheng et al. (1988). This version demonstrated good internal consistency (Cronbach's α = 0.85) and acceptable concurrent validity with clinical measures (r = 0.566), supporting its applicability in Chinese-speaking populations.

The BAI (Beck et al., 1988) is a 21-item self-report questionnaire designed to assess the severity of anxiety symptoms experienced over the past month. The scale encompasses multiple dimensions of anxiety, including autonomic, neurophysiological, panic-related, and subjective components. Respondents rate each item on a 4-point Likert scale, ranging from 0 (not at all) to 3 (severely), yielding a total score ranging from 0 to 63, with higher scores indicating greater anxiety severity. The Mandarin Chinese version of the BAI was adapted by Luo et al. (2004) and further validated in clinical populations with anxiety disorders. Its psychometric properties have been well established, demonstrating high internal consistency (Cronbach's α = 0.95) and good construct validity (Che et al., 2006).

The WHOQOL-100 was originally developed by the WHOQOL research group as a comprehensive instrument to evaluate quality of life across diverse cultural settings. A shorter version, the WHOQOL-BREF, was later developed to enhance feasibility in clinical and research contexts (The WHOQOL Group, 1998). The WHOQOL-BREF comprises 28 items, covering four major domains: physical health (7 items), psychological health (6 items), social relationships (3 items), and environment (8 items). In addition, it includes two general items assessing overall quality of life and general satisfaction with health A Taiwanese version of the WHOQOL-BREF has been developed and culturally adapted. Each item is rated on a five-point Likert scale ranging from 1 (very dissatisfied/poor) to 5 (very satisfied/good), with higher scores indicating better perceived quality of life.

Images will be acquired using a Siemens MAGNETOM Trio 3.0 Tesla. High resolution T1 structural image will serve for frameless stereotaxy navigation of TMS experiments. For the resting-state functional imaging, subjects will be instructed to remain awake with their eyes open. Forty-seven contiguous horizontal slices parallel to the intercommissural plane will be acquired by using a gradient echo T2*-weighted sequence (TR 2500 msec, TE 30 msec, flip angle 90°). The Statistical Parametric Mapping 12 (SPM12, Wellcome Centre for Human Neuroimaging, University College London, UK) and the SPMbased CONN (Whitfield-Gabrieli & Nieto-Castanon, 2012) will be used to preprocess and analyze the fMRI data. The preprocessing consists of realignment and unwrapping of the functional data, slice-timing correction, outlier identification, direct segmentation and normalization of functional and anatomical data, functional smoothing, and denoising by linear regression and temporal band-pass filtering.

Resting-state EEG will be recorded for 3 minutes under eyes-open (EO) and eyes-closed (EC) conditions, with condition order counterbalanced. During EO, participants fixate on a central white cross while remaining still and relaxed; during EC, they close their eyes and relax without engaging in specific mental activity. To quantify aperiodic activity, the power spectrum is separated into periodic oscillations and the 1/f-like aperiodic component using the Fitting Oscillations and One-Over-F (FOOOF) algorithm (Donoghue et al., 2020). The aperiodic fit is estimated from 35-45 Hz to reduce contamination from oscillatory peaks and residual line noise. The slope of the linear fit in log-log space is extracted as an index of excitation/inhibition balance. Slopes are calculated for each electrode and averaged by region: frontal (Fz, F3, F4, F7, F8) and posterior (Pz, P3, P4, P7, P8, O1, O2), for group comparisons and correlation analyses.