logo_8850.jpg

NeuralXAI: Unravel the Code of Brain’s Trust in AI – Neural Interface for Real-Time Explanation Tuning

4th International Conference on Global Entrepreneurship Summit 2025 & 2nd International Congress on Artificial Intelligence and Machine Learning

August 19-20, 2025 | Toronto, Canada

Rizwana Naz Asif

Ulster University London, UK

Abstract :

With the rise of Artificial Intelligence (AI) systems to play an advancing role in creating high-stakes decisions, human and reliable explanations of the action have become critical. Although Explainable AI (XAI) provides an interpretable result, the majority of the existing approach­es produce a static and one-size-fits-all outcome that does not take into consideration the emotional and cognitive condition of a user. The study proposes NeuroXAI a new framework that connects online neural feedback with the explanation generation, producing dynamic, personalized and emotionally sensitive AI explanation. NeuroXAI is a non-invasive system of continuously inferring user trust, confusion or anxiety during AI interaction by combining non-invasive EEG-based brain-computer interfaces and a multimodal affect recognition system (facial expressions, speech tone, eye tracking). Through a neuro-symbolic reasoning engine, an explanation is then altered in an adaptive fashion using visual, analogical, and counterfactual strategies. There is also a trust calibration component in the system to avoid the effects of over- or under-confidence especially critical in an area like mental health diag­nostics. Rudimentary prototypes indicate higher levels of user trust, lowered cognitive load, and better correspondence with the judgment of experts. NeuroXAI alters the definition of explainability toward being co-evolutionary by creating a double loop of feedback between human affect and the AI introspection into an explanation process in which the current state of explaining what is being done and why is regarding a state of not justification but under­standing and empathetic communication. This piece of work postulates a new domain of hu­man-AI interactions that have to be not only transparent but also cognitively and emotionally aware.

Biography :

Rizwana Naz Asif, is an outstanding educator and researcher who has a Ph.D. in Computer Science, focusing on utilizing AI in healthcare applications. Having completed 2 MSc degrees in Robotics at Middlesex University and Management at BPP University she is a technically innovative leader. She has an MSCS Bioinformatics degree and an MCS in Hajvery University, with a sound base in life sciences. She has written extensively in leading jour­nals and has more than 20 years of teaching experience in Pakistan and the UK and is a devoted advocate of education, health, and AI in the future, through the use of technology.