THE EFFECT OF REINFORCEMENT MAGNITUDE UPON RESPONDING UNDER FIXED‐RATIO SCHEDULES¹

Abstract

Responding under fixed-ratio schedules was studied as a function of two durations of food presentation. Latency of the first response after food presentation (post-reinforcement pause) was consistently shorter when food was presented for the longer duration. Only one of the four pigeons studied showed a consistently higher response rate, exclusive of post-reinforcement pause, as a function of the longer access to food. When ratio size was reduced, pause durations decreased, and the differences related to the two durations of food presentations became progressively smaller.It has been demonstrated consistently that the latency of the first response after reinforcement (post-reinforcement pause) increases as the size of the fixed-ratio schedule of reinforcement (FR) is increased (Kaplan, 1956;Ferster and Skinner, 1957;Premack, Schaeffer, and Hundt, 1964; Thompson, 1964;Winograd, 1965;Felton and Lyon, 1966;Mintz, Mourer, and Weinberg, 1966;Mintz, Mourer, and Gofseyeff, 1967;Powell, 1968). The mean response rate, exclusive of the post-reinforcement pause, does not show a consistent relationship to the FR response requirement (Felton and Lyon, 1966;Powell, 1968).The relationship between reinforcement magnitude and responding under FR schedules has been studied by varying the ratio requirement while the duration of each food presentation remained constant. The relationship could be elaborated further by varying duration of food presentation while the FR requirement remained constant. While it has been found in several experiments that response rates were relatively insensitive to changes in reinforcement magnitude (Keesey and Kling, 1961;Catania, 1963;Neuringer, 1967), one study (Powell, 1968) ProcedureThe independent variable was the duration of reinforcement, i.e., the time that the pigeon was permitted access to a grain mixture of 60% kafir, 30% vetch, and 10% hemp. The three naive pigeons were initially trained to key peck in the presence of a white 605 1969, 12,[605][606][607][608] NUMBER 4 (JULY)

Steady‐state Performance on Fixed‐, Mixed‐, and Random‐ratio Schedules

“…For example, researchers analyzing the reinforcement of different classes of IRTs have proposed that short IRTs are more frequently reinforced on VR schedules than on VI schedules, and as a result response rates are faster on VR schedules (e.g., Anger, 1956;Morse, 1966;Skinner, 1938). Regarding FR schedules, Powell (1969) proposed that running response rates are essentially constant as ratio size varies, and Sidman and Stebbins (1954) found that running response rates were not affected when subjects' levels of deprivation were changed. More recently, Mazur and Hyslop (1982) presented additional evidence supporting the notion of a constant response tempo on FR schedules.…”

Operant Conditioning

“…One assumption is necessary: that the waiting fraction, a, in Equation 1 is smaller when the upcoming reinforcer is large than when it is small (Powell 1969 andPerone &Courtney 1992 showed this for fixedratio schedules; Howerton & Meltzer 1983, for fixed-interval). Given this assumption, the linear waiting analysis is even simpler than hyperbolic discounting.…”

Incentive Theory: Iv. Magnitude of Reward

“…My only hypothesis about this finding involves the experimenters' use of fixed-ratio schedules of reinforcement. Animals pause longer after a larger reward (Harzem & Harzem, 1981;Staddon, 1974; but see Powell, 1969) and response rate after the pause does not change reliably with quality (Lowe, Davey, & Harzem, 1974) or duration (Powell, 1969) of reward-perhaps because the ratio contingency has already moved response rate close to its ceiling. These peculiarities of fixed schedules, and in particular fixed-ratio schedules, somewhat weaken the force of Essock and Reese's findings as evidence against decreasing marginal utility for duration of a reward.…”