BCI Training for Social Cognition and Error Monitoring in ASD (LEARNAUT)

Improving Social Cognition and Error Monitoring in Autism Spectrum Disorder Through a Brain Computer Interface Based on Reinforcement Learning

Social cognition seems to develop atypically in autism, particularly in processes such as faces perception, joint attention and social information processing. In this sense, and using an Emotional Paradigm of Facial Expressions (EFP) with a Brain-Computer Interface (BCI) based on Electroencephalography (EEG), the investigators intend to evaluate its effectiveness as a medical device in Autism Spectrum Disorder (ASD), namely: improving 1) social skills and 2) reducing generalized anxiety, 3) improve error monitoring, and consequently verify 4) an increase in motivation. To this end, the investigators will test a gamified intervention (EFP), using personalized feedback in real time. In this gamified interface, there is an artificial agent that learns rules through Reinforcement Learning using the evoked potentials from the participant as they observe the agent's right or wrong actions. The hypothesis is that this approach allows, during the gamified task (EFP), not only the agent/interface to learn, but also the participant through operant conditioning and implicit scrutiny of errors, which makes it particularly interesting for disorders in which error monitoring processes are compromised, as in ASD.

Study Overview

• Status: Recruiting
• Conditions: Autism Spectrum Disorder (ASD)
• Intervention / Treatment: Device: Emotional Facial Expression Paradigm (EFP); Device: Speller task

• Study Type: Interventional
• Enrollment (Estimated): 28
• Phase: Not Applicable

Contacts and Locations

• Study Contact: Name: Miguel Castelo-Branco; Phone Number: +351239488510; Email: mcbranco@fmed.uc.pt

Study Locations

• Portugal
  • Coimbra, Portugal, 3000 548
    • Recruiting
    • ICNAS
    • Contact: Miguel Castelo Branco, MD PhD; Phone Number: +351936360050; Email: mcbranco@fmed.uc.pt
    • Contact: Joana Amaral, MsC; Email: joana.aguiar.amaral@gmail.com

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Child; Adult
• Accepts Healthy Volunteers: No

Description

Inclusion Criteria:

• Participant is able and willing to give written informed consent/assent
• Previous diagnosis of Autism Spectrum Disorder by a qualified clinician according to gold-standard instruments
• Aged between 16 and 55 years old
• Normal or corrected-to-normal vision

Exclusion Criteria:

• Global Intelligence Quotient <70
• Dermatological diseases of the scalp
• Concurrent neurofeedback therapy
• Concomitant medication without a stable dosage for at least 4 weeks
• History of seizures and/or epilepsy

Study Plan

How is the study designed?

Design Details

• Primary Purpose: Treatment
• Allocation: Randomized
• Interventional Model: Parallel Assignment
• Masking: Single

Number of Arms

2

Arms and Interventions

Participant Group / Arm	Intervention / Treatment
Experimental: Experimental Arm; Experimental Arm The experimental group (n=17) will undergo three sessions of BCI intervention up to a maximum of six sessions. The intervention will be applied once a week, during three weeks, minimum, up to six weeks, maximum. The final number of sessions to be completed was defined after piloting. It will consist on a gamified task to train social cognition and to explore its viability for error-monitoring using EEG based non-volitional neurofeedback. The intervention comprises two modules, a visual paradigm called 'Emotional Facial Expression Paradigm' (EFP), where the subject is asked to observe the movements of a facial expression, and an 'Intelligent Agent' that learns through Reinforcement Learning from participant's judgments to the actions of the Intelligent Agent.	Device: Emotional Facial Expression Paradigm (EFP); The experimental intervention is a social cognition gamified training, through a software that implements communication between a participant and a computer, using a detection and classification algorithm based on EEG signal. The training consists of a visual stimulation model, specifically an emotional facial paradigm, where the subject is asked to observe the movements of a facial expression (referred to as cue image) expressing either happiness or sadness and to adapt the visual gaze according the cue image.
Other: Control Arm; The control group (n=17) will undergo a minimum of three and up to six sessions of BCI control intervention. The control intervention will be applied once per week. It will consist on a gamified attention task of executive function training, through a P300 BCI non-invasive interface based on EEG. The task consists of a speller, in which the participant is asked to write a word, by sequentially focusing the attention on the necessary letters.	Device: Speller task; The control intervention is a executive function gamified training, through a software that implements communication between a participant and a computer, using a detection and classification algorithm based on the P300 EEG signal. The task consists of a speller, in which the participant is asked to write a word, by sequentially focusing the attention on the necessary letters.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Change from baseline to post-intervention in a computerized attention task (Subway Game)	Responses consist on the time spent to complete 3 mazes, using computer mouse navigation. Responses are measured in seconds and the best outcomes correspond to the fastest responses.	From baseline to the end of treatment (maximum of 8 weeks)
Change from baseline to post-intervention in the Anxiety Sub-Scale from "The Hospital Anxiety and Depression Scale"	Participants scores ranges from 0 (best outcome) to 21 (worst outcome).	From baseline to the end of treatment (maximum of 8 weeks).

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Change in error-monitoring in Emotional Facial Expression Paradigm (EFP) across intervention	Improvement of the amplitude/discrimination of waveform morphology of Event-Related Potentials (ERPs).	At each intervention session, from first to last (intervention lasts up to 8 weeks and starts within 2 weeks after baseline)
Change from baseline to post-intervention in the Interest/Enjoyment subscale of "Intrinsic Motivation Inventory (IMI)"	Higher scores indicate greater interest/enjoyment when practicing the task (better outcome). Scores are calculated by averaging the item scores of the subscale.	From baseline to the end of treatment (maximum of 8 weeks)
Change from baseline to post-intervention in a computerized attention task (Subway Game - eye gaze)	Responses consist on the time spent to complete 3 mazes, using eye gaze. Responses are measured in seconds and the best outcomes correspond to the fastest responses.	From baseline to the end of treatment (maximum of 8 weeks).

Other Outcome Measures

Outcome Measure	Measure Description	Time Frame
Debriefing Questionnaire	The questionnaire will cover the participant's strategies applied on the intervention; their general experience of the intervention process; and safety assessment (adverse effects).	Immediately after each intervention session
Change in spelling accuracy in a Speller Task across intervention	Results consist in the spelling accuracy, i.e. correctly identified letters overall.	At each intervention session, from first to last (intervention lasts up to 8 weeks)
Change from baseline to post-intervention in the "Autism Treatment Evaluation Checklist (ATEC)"	The ATEC is a checklist to assess treatment effectiveness in ASD comprising 4 subtests: I. Speech/Language Communication (14 items, min.0-max.28); II. Sociability (20 items, min.0-max.40); III. Sensory/Cognitive Awareness (18 items, min.0-max.36); and IV. Health/Phys./Behavior (25 items, min.0-max.75). Total score (sum) ranges from min.0-max.179. The lower the score, the fewer the problems.	From baseline to the end of treatment (maximum of 8 weeks)
Change from baseline to post-intervention in total score of "Toronto Alexithymia Scale-20 (TAS-20)"	Participants scores ranges from 20 (best outcome) to 100 (worst outcome).	From baseline to the end of treatment (maximum of 8 weeks)
Change from baseline to post-intervention and across intervention in total score of "Profile of Mood State" short version (POMS)	Higher scores indicate a greater degree of mood disturbance.	From baseline to the end of treatment (maximum of 8 weeks) and across the intervention.
Change from baseline to post-intervention in the total score of "Anxiety Scale for Autism-Adults (ASA-A)"	Participants scores ranges from 0 (best outcome) to 60 (worst outcome). A total score of ≥ 28 on the ASA-A may indicate the presence of significant levels of anxiety.	From baseline to the end of treatment (maximum of 8 weeks).

Collaborators and Investigators

• Sponsor: University of Coimbra

Study record dates

Study Major Dates

• Study Start (Actual): March 1, 2025
• Primary Completion (Estimated): September 1, 2026
• Study Completion (Estimated): December 1, 2026

Study Registration Dates

• First Submitted: February 3, 2025
• First Submitted That Met QC Criteria: February 10, 2025
• First Posted (Actual): February 17, 2025

Study Record Updates

• Last Update Posted (Actual): August 1, 2025
• Last Update Submitted That Met QC Criteria: July 29, 2025
• Last Verified: July 1, 2025

More Information

Terms related to this study

• Additional Relevant MeSH Terms: Mental Disorders; Neurodevelopmental Disorders; Child Development Disorders, Pervasive; Autism Spectrum Disorder
• Other Study ID Numbers: CIB02ASD; 2022.12232.BDANA (Other Grant/Funding Number: Fundação para a Ciência e Tecnologia (FCT))

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: UNDECIDED

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No
• product manufactured in and exported from the U.S.: No