The goal of this study is to compare the neural correlates of predictive brain mechanisms during high-level social and non-social flexibility tasks, between autistic and non-autistic males and females. The aim is to better understand if autistic individuals differ from non-autistic individuals in their cognitive flexibility according to the social and predictable nature of the stimuli, at a behavioral and neurofunctional level using the functional Magnetic Resonance Imaging (fMRI). As autistic females usually present a more subtle profile than autistic males, the investigators will also explore if autistic males and females present neurofunctional differences.

Background:

Autism spectrum conditions (ASC) may involve specificities in predictive brain mechanisms, which guide perception together with sensory input. These predictive mechanisms are at play in various aspects of life, such as social interactions, in which autistic individuals have difficulties. They are also important for cognitive flexibility, which involves changing tasks, strategies, or perspectives according to the situation. recent research highlighted that autistic individuals faced difficulties in flexibly adjusting their predictions to unpredictable social contexts during a socio-emotional flexibility task (social stimuli). However, they found no difference between autistic and non-autistic individuals during a cognitive flexibility task involving predictable changes of non-social stimuli. This study aims to more precisely investigate which conditions lead to flexibility difficulties in autism (i.e., the social context, the unpredictability, or both) and to investigate the brain networks involved in autistic and non-autistic participants during high-level cognitive flexibility tasks involving social and non-social stimuli. In addition, autistic females often presenting better socio-communicational abilities than autistic males and a different profile from males in past research involving cognitive flexibility, the investigators will explore if they present a different neurofunctional signature from males during these tasks.

Procedure:

The protocol begins with an imaging localization phase (5 minutes). Subsequently, functional MRI tasks (T2*) commence. The protocol consists of two cognitive tasks, performed within the MRI scanner, each preceded by a training phase. The acquisition of each participant's anatomical sequence (T1; 5 minutes) is conducted between the first two cognitive tasks to avoid compromising participant concentration. Finally, functional connectivity data are collected during a 'resting state' sequence (10 minutes). The study comprises a single visit.

Cognitive Tasks:

The protocol starts with an Emotional Shifting Task (EST), where participants evaluate the valence of emotion (positive vs. negative) on an image while it's truncated; then, the image appears not truncated, and the participant must reevaluate its valence. In some cases, the valence has changed (from positive to negative or from negative to positive), and the participant must consider scene cues and demonstrate flexibility to reassess their judgment. This task lasts approximately 11 minutes.

In a second group of tasks, the participant will perform adapted versions of the Task-Switching Task (TST). In the first task, a letter and a number appear side by side, surrounded by either a blue or yellow frame. When the stimulus appears in a blue frame, the participant must focus on the letter and press the left response button if it's a consonant and the right response button if it's a vowel. When the stimulus appears in a yellow frame, the participant must focus on the numbers and press the left response button when it's odd and the right response button when it's even. In the first part, the subject first practices only the letter task to assimilate the task. Then, similarly, the subject practices only the number task. Finally, the subject performs the alternating task, with 2 conditions of predictability: predictable (task change every three trials), and not predictable (random change). This task lasts approximately 10 minutes. The second task adds an emotional component to the TST task: a face (male or female) associated with an emotion (positive or negative) is surrounded by either a blue or yellow frame. When the stimulus appears in a blue frame, the subject must focus on the gender and press the right response button if it's a female and the left response button if it's a male. When the stimulus appears in a yellow frame, the subject must focus on the facial emotion and press the right response button if it's a positive facial emotion and the left response button if it's a negative facial emotion. In the first part, the subject first practices only the gender task to assimilate the task. Then, similarly, the subject practices only the emotional task. Finally, the subject performs the alternating task, with the 2 conditions of predictability. This task also lasts 10 minutes.

Psychometric Assessment :

The subject will also undergo a psychometric assessment including the Social Responsiveness Scale (SRS), Camouflaging Autistic Trait - questionnaire (CAT-Q), Weschler Adult Intelligence Scale 4th edition (WAIS-4), Behavior Rating Inventory of Executive Function-Adult Version (BRIEF-A), Empathy Quotient (EQ), Autism Spectrum Quotient (AQ), and Autism Diagnostic Interview -revised (ADI-R) with the parent (if possible).

Hypotheses:

Given the literature, the investigators hypothesize a reduction in activation of certain areas in predictable conditions compared to unpredictable conditions, indicating the presence of a prediction error. Specifically, in the socio-emotional flexibility task (EST), the investigators expect to observe this reduction, particularly in the precuneus, the temporal lobe, the fusiform gyrus, and the locus coeruleus. In the cognitive flexibility task (adapted TST), the investigators will more specifically focus on the locus coeruleus. The investigators also anticipate observing the involvement of a global functional network showing differences in activations between autistic and non-autistic individuals, as well as a modulation of this reduction in activation (prediction error) between the two groups. Finally, the investigators hypothesize an intermediate profile in autistic females, both at the behavioral and neurofunctional level, situated between those of autistic males and non-autistic females.

Data Analysis:

Behavioral data will be analyzed in terms of accuracy rates and reaction times for both tasks. fMRI data will undergo classical analysis (General Linear Model - GLM). This GLM analysis will allow a direct comparison of experimental conditions within the same group of participants and identify possible differences in activations between the groups.

Inclusion Criteria:

For non autistic individuals

• Affiliation with social security scheme.
• Registration in the national database of individuals willing to participate in research involving human subjects.
• Absence of contraindications for MRI.
• Signed informed consent before any study-related procedure.
• Visual acuity allowing normal perception of stimuli or corrected vision.
• Autism Quotient (AQ) (Baron-Cohen et al., 2001) < 32.

For autistic individuals:

• Affiliation with social security scheme.
• Absence of contraindications for MRI.
• Signed informed consent before any study-related procedure.
• Visual acuity allowing normal perception of stimuli or corrected vision.
• Diagnosis of autism established by a competent professional based on DSM/ICD criteria.

Exclusion Criteria:

• Any subject with a vascular stent implanted less than 6 weeks before the examination.
• Any subject with implanted biomedical material deemed "unsafe" or "not safe" in the list: http://www.mrisafety.com/TheList\_search.asp.
• Any acquisition procedure not meeting the conditions required for "conditional" use in subjects with implanted biomedical material classified as "conditional" in the list: http://www.mrisafety.com/TheList\_search.asp.
• Any subject with biomedical material such as a cardiac, neuronal, or sensory stimulator (cochlear implant due to the risk of demagnetization, heating, electrode, and artifacts) or a ventricular shunt valve without medical and paramedical supervision trained in MRI performance in these subjects.
• Intraocular or intracranial ferromagnetic foreign body near nerve structures (risk of displacement and complications such as ocular or cerebral lesions).
• Claustrophobia - Psychiatric or neurological conditions other than those specified.
• Existence of a severe general health condition incompatible with the study: cardiac, respiratory, hematological, renal, hepatic, or cancerous.
• Alcohol ingestion before the examination.
• Persons covered by articles L1121-5 to L1121-8 of the Public Health Code (pregnant or breastfeeding women, minors or adults under legal protection, individuals under administrative or judicial surveillance).
• Visual impairments.
• Intellectual disability.
• Inability to understand research instructions or provide informed consent.

Autistic with a diagnosis and non autistic adults.