Integration of AI, Computational Thinking, Pattern Recognition, and Abstraction in Programming Students: A Systematic Review

Abstract

This umbrella systematic review analyzes the convergence between computational thinking, computational abstraction, pattern recognition, and artificial intelligence in the training of programming students. A total of 50 studies published up to May 2025 – without restrictions on the year of publication – were examined, selected from databases such as Scopus, Web of Science, IEEE Xplore, among others, following the PRISMA protocol. The thematic analysis identified recurring patterns and significant gaps in this convergence. Findings reveal that, although there is abundant literature on each axis separately, no study simultaneously integrates all four components. Moreover, 64.7% of the reviews address only one of the concepts, highlighting a conceptual fragmentation. This review proposes an integrative framework for programming education that brings together these dimensions, outlining future research directions focused on the development of pedagogical frameworks, assessment instruments, and empirical validation in educational contexts.