O RECONHECIMENTO FOTOGRÁFICO COMO MEIO DE PROVA NO PROCESSO PENAL: ANÁLISE CRÍTICA, DESAFIOS NORMATIVOS E GARANTIAS PROCESSUAIS

Abstract

The dissertation conducts an interdisciplinary investigation into the reliability and risks associated with the use of photographic identification within the Brazilian criminal justice system. It adopts an approach that integrates normative, psychological, and empirical analyses to critically assess the legal implications of this evidentiary method, with emphasis on Article 226 of the Code of Criminal Procedure, CNJ Resolution No. 484/2022, and contributions from the psychology of testimony. The objective is to understand how the procedure can be improved to reduce judicial errors and ensure greater legal certainty, based on the hypothesis that photographic identification, when carried out without technical and scientific criteria, is insufficient to support criminal convictions, particularly due to its vulnerability to cognitive biases and discriminatory practices. The research began with an analysis of current legislation, which outlines steps such as the prior description of the suspect, the inclusion of similar individuals (fillers), and the formalization of the procedure. However, it was observed that these requirements are frequently neglected by investigative agents and tolerated by the judiciary, revealing a practice that trivializes personal identification and allows its use as sole evidence, even when conducted outside legal boundaries. This disconnection between legal norms and practical application undermines both the validity of the evidence and the legitimacy of the criminal process, weakening the guarantees of due process and the presumption of innocence. The study employed a qualitative and documentary methodology, combining normative analysis with empirical case studies from the Innocence Project Brazil and institutional reports, such as the Executive Summary of the CNJ Working Group on Suspect Identification. Contributions from the psychology of testimony were essential in demonstrating that identification depends on memory mechanisms subject to distortion. Stress, external suggestion, time elapsed, and poorly conducted interviews are factors that directly affect identification accuracy, as evidenced by studies on false memories and the “cross-race effect.” Informal practices such as show-ups and unstandardized photo arrays were analyzed and identified as highly suggestive and incompatible with a reliable criminal procedure. Jurisprudence was also examined, notably HC No. 598.886/SC, which revealed that 74% of the cases analyzed by the Superior Court of Justice involved photographic identifications with procedural flaws. Data from CONDEGE show that 81% of mistaken identifications involved Black individuals, highlighting the influence of structural racism and the urgent need for effective application of the Protocol for Judging with a Racial Perspective. The dissertation concludes that isolated photographic identification constitutes fragile evidence and must always be corroborated by other robust elements. It recommends the national standardization of procedures, the elimination of show-ups, and the controlled use of technologies such as facial recognition, provided they are subject to independent auditing. Overcoming current shortcomings requires coordination between legislation, science, and institutional commitment. Only through critical review and the replacement of informal practices with technically validated methods will it be possible to ensure that criminal procedure operates within the framework of the Democratic Rule of Law, centered on human dignity and the prevention of judicial errors.