Speakers
Description
Introduction. Artificial intelligence systems increasingly rely on large datasets, yet challenges remain in achieving human-like perception and pattern recognition. Gestalt theory, which explains how humans organize visual information into meaningful wholes, offers a promising framework for improving AI training processes.
Aim. This study analyzes theoretical aspects of Gestalt principles and explores their application in enhancing AI training and perceptual performance.
Methods. A literature review was conducted using the Elsevier Scopus database. Relevant studies were selected based on keywords: Gestalt theory, perception, artificial intelligence, machine learning, and pattern recognition.
Results. Key Gestalt principles—such as proximity, similarity, closure, and continuity—can support more efficient feature extraction and pattern grouping in AI models. Integrating these principles into training processes improves object recognition, reduces data noise, and enhances interpretability. Additionally, AI systems trained with perceptual structuring show improved generalization and robustness across varied datasets.
Conclusions. Applying Gestalt theory in AI training provides a structured approach to perception, improving model performance and interpretability. Future research should focus on operationalizing these principles within scalable machine learning architectures.
Keywords: Gestalt theory, artificial intelligence, perception, machine learning, pattern recognition
