Effects of Reinforcement Schedules on Preferences and Variability of Responses in Laboratory Rats

Autor: Emery E. Benson, William J. Griffiths
Rok vydání: 1967
Předmět:
Zdroj: Perceptual and Motor Skills. 24:105-106
ISSN: 1558-688X
0031-5125
DOI: 10.2466/pms.1967.24.1.105
Popis: Herrnstein (3) found pigeons, previously trained under continuous reinforcement (CRF) t o peck at a horizontally oriented rubber strip, subsequently increased the tendency to peck at one or the other end of the rubber strip when transferred to an intecmittent schedule of reinforcement (VI-3 or variable interval of 3 min.). Studies of children using intermittent schedules of reinforcement (1, 2, 4) suggest that training under variable-ratio (VR) and fixed-ratio (FR) reintorcement schedules in a singlemanipulandum situation led to increased variability when Ss were transferred to a multimanipulandum situation. Since the argument of species differences may cloud the theoretical implications, the present study was a replication with rats as Ss. Method.-Standard Scientific Protorype Rat Test Cages, containing single levers attached to ,Anger pellet dispensers and event counters, were used to train Ss on the reinforcement schedules. For the transfer phase a Gerbrands rat box was modified by replacing the metal side plate with five movable fiberboard panels; each one contained a metal food cup, a chain lever, a push-plunger, a spoked wheel, or a pump-handle lever. Movement of each device operated a microswitch and was connected via relays to an Anger food dispenser and an Esterline-Angus event recorder. Eight male albino Norway rats from che Farris Institute and eight male hooded rats from the Animal Supply Co. (Napa, Calif.), 60 days of age, were assigned randomly to three experimental groups and one control. The experimental groups were given massed practice on the single-lever apparatus under one of three reinforcement schedules: continuous (CRF), fixed-ratio-4 (FR-4), and variable-ratio-4 (VR-4). Ss were 24-hr. food deprived and made 2,000 presses; the first 500 were under continuous reinforcement and the remaining 1,500 presses were given under their respective reinforcement schedule. Controls received no training or experience in the sing!e-lever apparatus. Forty-elghc hours after training was completed Ss were transferred to the apparatus containing four novel rnanipulanda. Each S was 24-hr food deprived and was given two blocks of five 12-min. trials, separated by 5-min. intervals. All Ss received a food peller for manipulating any of the Four devices, regardless of their previous training under different schedules. The food-cup panel was moved successively to each of the five positions. The devices were randomly assigned the remaining four positions for the first >-trial block. These arrangements were replicated for the second 5-trial block. Re~ults.-The frequencies of responses were very similar for the albino Norway 'Supported in part by Grant M-4724 (C-I) from the National Institute of Mental Health and, in part, by Grant 687-R from the Graduate School, Universiry of Montana. Transcripts of raw data have been filed with the AD1 Auxiliary Publications Project, Photoduplication Service, Library of Congress, Washington, D. C. 20540. Remit $27.50 for photocopies or $7.75 for 35-mm. microfilm of Document 9248. (Reproduction copy is poor.) TEB at VA Hospital, Augusta, Ga; WJG deceased, October 26, 1963. We are indebted to H. C. Salzberg, VA Hospital, Augusta, Ga for help with statistical analyses.
Databáze: OpenAIRE
Externí odkaz: