Select control and reject control in conditional discriminations in humans and pigeons

Author(s)

Edson Massayuki Huziwara

• Document type: Doctoral Thesis
• Press: São Paulo.
• Institution: Universidade de São Paulo (USP). Instituto de Psicologia (IP/SBD)
• Defense date: 2010-05-07

Examining board members

Gerson Aparecido Yukio Tomanari; Paula Debert; Jair Lopes Junior; Miriam Garcia Mijares; Nicolau Kuckartz Pergher

• Advisor: Gerson Aparecido Yukio Tomanari; Deisy das Graças de Souza

Abstract

The study of responding controlled by selection or rejection relations is within the scope of research on the necessary and sufficient conditions for the formation of equivalent stimuli classes. This project sought to investigate aspects related to this subject by performing experiments with humans and pigeons. The first experiment evaluated whether the record of eye movement - in terms of stimuli observation time and tracking patterns exhibited throughout conditional discrimination training - could provide ancillary measures on the acquisition process of control by selection rejection during the conditional discrimination training on human participants. For this purpose it was important to prepare an experimental situation that predictably generate differences in the discriminative learning outcomes. Thus, it would be possible to verify if the eye movement could provide additional clues about the establishment of responding controlled by relations of selection or rejection. Previous studies have shown that using the keyboard or mouse, in MTS tasks, produces different results in class formation testing. The objective was, therefore, to check if different response topographies would generate different results in conditional discrimination acquisition and class formation tests and, furthermore, if it would generate different tracking patterns, in terms of stimuli observation time. College students were subjected to conditional discriminations training involving six sets of stimuli (training EF, DE, CD, BC and AB) while using a device that recorded eye movement. Results suggest that different response topographies originate different patterns of stimulus tracking, however there was no evidence that such patterns referred to the establishment of relations controlled by selection or rejection. The second experiment sought verifying the existence of transitivity relations from conditional training involving temporal stimuli. Subjects were 12 pigeons (Columba livia) experimentally naive and kept at 80% of their ad lib weight. The equipment used consisted of a box containing three standard experimental response keys horizontally arranged. A training attempt occurred as follows: house light and the central key were lit at the same time. If, for example, the programmed interval was 4 s, a white light in the central key remained on for that period. After 4 s, the light from the central key was turned off and colors were presented on the side keys. A peck on either of the side keys ended the attempt. Correct answers were followed by a short period of access to the feeder and an ITI of 20 s. Incorrect answers were followed by the ITI and the repetition of the same trial (correction procedure). The procedure consisted of three phases of training: in Training A attempts, the subjects learned to choose the red colored key when the 1 s model was presented, and to choose the green colored key when the 4 s model was presented. In Training B attempts, they learned to choose the blue colored key when the 4 s model was presented and to choose the yellow colored key when the 16 s model was presented. During Training C, subjects learned to choose the blue colored key when green color was presented and to choose yellow colored key when red color model was presented. Regarding A and B trainings, blue and green stimuli were related to the same temporal stimulus 4 s, while red and yellow ones did not share any common temporal stimulus during earlier training. Results suggest that relations between blue and green were acquired more quickly, a fact that seems to demonstrate formation of transitivity relations in earlier stages of training (AU)