Psicologia do testemunho e uma nova técnica de entrevista investigativa: a versão brasileira da Self-Administered Interview

摘要

Witnesses play a key role in police and legal areas. However, nearly four decades of research on the Eyewitness Memory shows that because of the fallibility of human memory, a testimony can not be fully reliable. The best way to obtain reliable reports is gathering information properly as soon as after the crime. A few years ago was developed in the UK, the Self-Administered Interview (SAI), a written investigative interview protocol, to be used as soon as the police arrives at the scene of the crime. This method, inspired by the Cognitive Interview (Fisher & Geiselman, 1992) has shown consistent results to obtain accounts with high amount and accuracy of information, while protecting memory against suggestions effects. This thesis did the translation, adaptation and testing of the SAI Brazilian version. Our aim was also to propose the spoken modality of the SAI and investigate the effects of reviewing the own account, provided by SAI, before answering a questionnaire about a crime. To achieve these goals, the thesis consisted of one theoretical study and three empirical studies. The Theoretical Study review and discusses the theoretical assumptions of one of the most effective techniques of SAI: mental reinstatement of context. The Empirical Study 1 focused on the process of translation, adaptation and the first test of the SAI Brazilian version. As a whole, it was concluded that the process of translation and adaptation was successful, since it was found that the Brazilians have provided quantitatively and qualitatively information quite similar to those who used the original SAI (in English). The Empirical Study 2 aimed to compare two formats of providing an initial account via SAI: written vs. spoken. Due to the inherent SAI limitation for require reading and writing, it was investigated whether the spoken format of the protocol would cause differences in terms of quantity and accuracy of recalled information. The results showed that, in general, both SAI modalities obtained roughly equivalent performance, suggesting that in cases where the witness is unable or has any difficulties with writing, the spoken SAI can be applied without losses on the amount and on the accuracy of information collected. Finally, the third empirical study aimed to investigate the effects of reviewing a statement (written on the day of crime through SAI) before responding to a questionnaire about the crime. The results indicated that the opportunity to review the own statement did not generate superior performance in the questionnaire responses compared to those who did not review their statements. However, regardless of the variable 'review', the study confirmed SAI beneficial effects, since participants who completed it obtained better performance, a week later, than the control-group. Together, these studies bring new contributions to the Eyewitness Memory and Investigative Interviewing areas, especially in the Brazilian context. Given the necessity to put in place effective and viable techniques that enhance investigations and reduce damages, the SAI emerges as a possible alternative to move in this direction.