Association for Behavior Analysis International

The Association for Behavior Analysis International® (ABAI) is a nonprofit membership organization with the mission to contribute to the well-being of society by developing, enhancing, and supporting the growth and vitality of the science of behavior analysis through research, education, and practice.


41st Annual Convention; San Antonio, TX; 2015

Event Details

Previous Page


Symposium #451
Response-Reinforcer Relations, Stimulus-Reinforcer Relations and Resistance to Change
Tuesday, May 26, 2015
9:00 AM–10:50 AM
006C (CC)
Area: EAB; Domain: Basic Research
Chair: Carlos Cancado (Universidade de Brasilia, Brazil)
Discussant: John A. Nevin (University of New Hampshire)

Experiments will be presented in which the effects of manipulating the response-reinforcer relation and the stimulus-reinforcer relation on resistance to change were assessed. Canado, Abreu-Rodrigues, Al and Doughty studied the resistance to change of rats lever pressing under a multiple schedule arranging equal rates, but different percentages of response-dependent events between components. Resistance to change was inversely related to the degree of response-reinforcer dependency. Galuska and Havens assessed, with rats, how constant and variable-sized reinforcers under random-ratio schedules affected response patterns, resistance to extinction and preference. Resistance to extinction was greater under variable-sized reinforcers, but preference for the constant-sized reinforcer occurred for all rats. Hall and Lattal studied the resistance of pigeons key pecking when timeouts from variable-interval schedules of food were response dependent and independent. Resistance to extinction was not systematically differential when timeouts were response-dependent, but was greater in the component uncorrelated with response-independent timeouts. Kuroda, Cook and Lattal studied the relation between response rates and resistance to extinction. With pigeons, resistance to extinction in a yoked-interval schedule was either greater or equal to that in a variable-ratio schedule. With rats, an inverse relation between rates of lever pressing under variable-interval schedules and resistance to extinction was obtained.


Response-reinforcer dependency and resistance to change

Carlos Cancado (Universidade de Brasilia, Brazil), Josele Abreu-Rodrigues (Universidade de Brasilia), Raquel Alo (Universidade de Brasília, Brazil), Adam H. Doughty (College of Charleston)

The effects of different degrees of response-reinforcer dependency on resistance to change were studied with four rats. Lever pressing produced water after variable interreinforcer intervals on a two-component multiple schedule. The percentage of response-dependent water was manipulated, while rate of water was equated, between components. In the first (100-100%) condition, 100% of the water in each component was response dependent. In the second (100-10%) condition, the percentage of dependent water was 100%, in one component, and 10% in the other. Replications of the first and second conditions (reversing stimuli between schedule-components) were conducted. For three rats, a replication of the second condition was then conducted. Baseline response rates were similar between components in the 100-100% conditions, and lower in the component with 10%-dependent water in the remaining conditions. Resistance to extinction (and satiation, for one rat) was not systematically differential in the 100-100% conditions. In the other conditions, resistance to extinction was generally greater in the component with 10%-dependent water. Thus, resistance to change varied inversely with the degree of response-reinforcer dependency when the rate of water was equated between multiple-schedule components.

Analyzing Factors Related to Gambling in Rats: Response Persistence and Preference under Random-Ratio Schedules Differing in Reinforcer Magnitude Variability
CHAD M. GALUSKA (College of Charleston), Crane A Havens (College of Charleston)
Abstract: The current study determined the effects of either a constant-sized (e.g., 2 pellets) or a variable-sized mean equivalent (e.g., 0-12 pellets) reinforcer on random-ratio responding in rats. Dependent measures of interest included within-session patterns of responding, resistance to extinction, and preference under a concurrent schedule. Under a multiple schedule in which the two components alternated every 5 min, the component associated with the variable-sized reinforcer engendered greater responding later in the session. Responding in this component also was considerably more resistant to extinction. Despite this, under the concurrent schedule, the constant-reinforcer alternative was preferred in all rats. The results are discussed within the context of habituation, the partial reinforcement effect, behavioral momentum, and implications to gambling. Parametric manipulations of reinforcer magnitude variability, the random-ratio parameter, and the role of cues signaling losses are underway.

Resistance to Extinction of Timeout Punished Responding

EZRA HALL (West Virginia University), James E. Cook (West Virginia University CED), Kennon Andy Lattal (West Virginia University)

A series of experiments were conducted to study the resistance to extinction of pigeons key pecking when 20-s timeouts were delivered response dependently and response independently. A strictly-alternating two-component multiple schedule was used where each component was 10 min long and occurred twice per session. Key pecking was maintained by a variable-interval 45-s schedule of food. Baseline sessions were followed by one of two conjoint timeout conditions in the first and third components of a session: a variable-ratio (VR) 2 or a variable-time (VT) 5-s schedule. Each condition lasted for 20 sessions. A single extended extinction session was then conducted with components strictly alternating 16 times. In some extinction conditions, timeouts continued to be arranged according to the VR-2 and VT 5-s schedule, and in some extinction conditions, timeouts were not delivered. Relative resistance to extinction was not consistently greater in the unpunished components when compared to the punished components when response-dependent timeouts were and were not continued during extinction. Relative resistance to extinction was consistently greater in the no timeout components when compared to the VT 5-s timeout components that were continued through the extinction tests. The results add to an understanding of punished and unpunished behavior.

Response Rate and Resistance to Extinction
Toshikazu Kuroda (Aichi Bunkyo University), James E. Cook (West Virginia University CED), KENNON ANDY LATTAL (West Virginia University)
Abstract: In two experiments, the role of response rates in resistance to extinction was examined. Behavior in transition from a variable-ratio (VR) schedule and its yoked-interval (YI) counterpart to extinction was compared in Experiment 1. Following maintenance of key pecking of pigeons under a multiple VR YI schedule of reinforcement, responding was extinguished in both components during a single 160-min extinction session. Responding decreased systematically across the extinction session. In both components, the response patterns comprised periods of steady responding alternating with increasing long periods without responses. Extinction bursts occurred in five of the nine pigeons but was not systematically related to the schedule immediately preceding extinction. Resistance of responding to extinction in the formerly YI component was either greater or equal to that in the formerly VR component. Experiment 2 further investigated the problem stated in the title by training 20 rats to respond on VI schedules. After such training responding by each rat was extinguished. Resistance to extinction was significantly greater for the five rats with the lowest response rates during VI compared to such resistance for the five rats with the highest rates. These results together suggest that lower response rates are more resistant to extinction than are higher response rates, when reinforcement rate is equal. Response rate thus must be considered a source of control during resistance to change tests.



Back to Top
Modifed by Eddie Soh