International Journal of Science, Engineering and TechnologyAuthors: Dr. D. Muthulakshmi, S. Abinaya, B. Kaviya
Abstract: Simulation of human thought processes using artificial intelligence is seen as a revolutionary approach toward the design and development of cognitive computing paradigms, moving beyond pattern recognition and matching architecture. This paper discusses a thorough analytical review of thought simulation methodologies used in contemporary artificial intelligence systems, highlighting the transformation from large language models to cognitive artificial intelligence architecture. In this regard, through a thorough literature survey of 30 peer-reviewed articles published between 2021 and 2026, we identified three major thought simulation methodologies: personalized cognitive simulation using large-scale behavioral datasets, emulative cognitive simulation using world models for situated reasoning, and developmental cognitive architecture using functional equivalence to human cognitive growth. We propose a unified methodological framework for cognitive simulation architecture (CSA). The analysis process will be accompanied by five representative figures on the comparison of simulation fidelity, components of the architecture, performance measures, and comparative evaluation with reference models. A comparative analysis table will be included, which will compare existing approaches on seven criteria: cognitive grounding, scalability, transparency of reasoning, consistency of interventions, and computational efficiency. Our analysis suggests that the development of thought simulation capabilities has moved beyond the simple mimicry of surface-level behavioral processes and has reached the development of architectures with internal coherence, traceability, and the emergence of collective intelligence. However, many challenges remain to be overcome in the development of functional equivalence with human cognition and the ethics of the process.
DOI: https://doi.org/10.5281/zenodo.19049937
