Cognição animal: identidade generalizada e simetria em macaco-prego (Cebus apella)

• Autor(a) principal: SANTOS, José Ricardo dos
• Data de Publicação: 2003
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Repositório Institucional da UFPA
• Texto Completo: http://www.repositorio.ufpa.br:8080/jspui/handle/2011/1914
• topic: CNPQ::CIENCIAS HUMANAS::PSICOLOGIA::PSICOLOGIA COMPARATIVA; Cebus apella; Psicologia do desenvolvimento; Comportamento animal; Macaco-prego

Complex behavioral repertories, such as generalized identity matching and equivalence classes, have been easily found in normal humans, children, and youngsters with learning deficits. However, it is not easy to find such a positive results with non-human subjects. Symmetry is one of the most difficultly found defining properties of equivalence in nonhumans. It might happen because symmetry involves sample-comparison function reversals as well as modification in the sequence and position of stimuli presentation. The negative results in obtaining generalized identity matching and equivalence class formation in non-humans subjects may be related to incoherence between the SCT (Stimulus Control Topography) planned by the experimenter and the SCT presented by the subjects. So it suggests the necessity of a more specific methodological development. The present study proposed to apply the training and testing experimental procedures to obtain generalized identity matching and to verify the possibility of emergence of symmetry after arbitrary matching to sample training, through sample stimulus control shaping procedure and in the absence of correlation between the function and the positions of the stimuli. One capuchin monkey (Cebus apella) served as subject. He was young-adult and naive. Two experiments were executed. In the Experiment I, we carried out simple discrimination reversals training, conditional discrimination training, with an identity matching to sample procedure, and generalized identity test in extinction. The results showed that the procedure used to train simple discrimination (and reversals) was efficient as well as the procedure to train identity matching. All generalized identity tests reached positive results. In the Experiment II, we carried out arbitrary matching to sample training, with a sample stimulus shaping procedure in 8 steps, and one BA symmetry test. This study aimed to verify if elements positively related in conditional discriminations (AB training, for example) might be recombined by the subject without additional training. The performance of the subject in the BA symmetry test reached 100% of correct choices, showing that it is possible to obtain the property of symmetry in arbitrary conditional discriminations with non-humans subjects. The data also suggest that additional research has to be carried out in order to contribute to specifying the necessary conditions to obtaining complex repertory such as equivalence class formation in non-human subjects.