Ensinando a mensurar a dúvida com inteligência artificial: uma proposta educacional para analisar as causas da violência em Guayaquil sob a perspectiva da indeterminação e da incerteza
Year: 2025
Type: article
Study design: Clinical Trial
Abstract: Introduction: This article presents an innovative educational proposal for analyzing complex social problems, such as urban violence in Guayaquil, by teaching the measurement and visualization of doubt and uncertainty. A neutrosophic epistemic evaluation framework, implemented with large language models (LLMs), is proposed to overcome traditional binary and unicausal analytical approaches. Materials and methods: The Neutrosophic Epistemic Evaluation Protocol (NEEP) was developed, which decomposes causal claims into three independent dimensions: Truth (T), Indeterminacy (I), and Falsity (F). A quasi-experimental case study (pre-test/post-test) was conducted with 75 participants in Guayaquil, who interacted with analyses generated by the NEEP on five common causal propositions of violence. The change in perception was assessed using questionnaires that measured initial stances, understanding of complexity, and T, I, F assessments. Results: 77.4% of participants experienced a positive cognitive shift, moving from a simplified to a more nuanced and complex view of the problem. Quantitative analysis revealed significant means across the three dimensions (T=7.12, I=6.29, F=7.09) and low correlations between them (r < 0.5), empirically validating their theoretical independence. The sum of T+I+F significantly exceeded the neutral value, indicating a perception of epistemic conflict (hyper-truth). Discussion: The results demonstrate that the neutrosophic framework, facilitated by AI, acts as an effective pedagogical tool for structuring critical thinking and legitimizing uncertainty as an analyzable component. The system's ability to "refrain" from making judgments in the face of conflicting or insufficient evidence models epistemic humility, a crucial virtue for contemporary citizenship. Conclusions: Integrating neutrosophical logic with AI offers a powerful educational paradigm for the age of complexity, focused on developing competencies to map uncertainty, assess trust, and recognize the limits of knowledge, transforming doubt into an object of learning.
Source: Maestro y sociedad
Institutions: University of Guayaquil, University of New Mexico
Language: Spanish
Subfield: Mathematical Physics
Field: Mathematics
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W7220103652
