The effect of intertrial interval on avoidance learning in fish

Abstract

The effect of intertrial interval on the acquisition of a shuttlebox avoidance response in fish, X iphophorus he fieri, was studied over five days of 20 trials each with intertrial intervals of .5, 1,2,5, 10, and 20 minutes. The results sup­ ported the hypothesis that learning is facilitated by longer intertrial intervals, but the a-shaped function generally ob­ tained with rats was not evident within the limits of the time intervals employed in this experiment. Several theoretical interpretations of avoidance learning suggest that the instrumental avoidance re­ sponse is reinforced by the reduction of conditioned fear, which is assumed to occur during the time interval between the response and the onset of the CS on the following trial. The length of this intertrial interval (ITI) should, therefore, have a direct relation­ ship to the speed of acquisition of the avoidance re­ sponse. Data available on the effect of ITI on acquisition of an avoidance response vary with the type of response required. Murphy & Miller (1956), Levine & England (1960), and Brush (1962) have all reported better per­ formance by rats at longer ITIs in the shutUebox. The ITIs varied from 20 sec. in the Levine and England study to 20 min. in the Brush experiment. Pearl (1963), as well as Pearl & Fitzgerald (1966), have shown, on the other hand, that the learning of a discriminated lever press avoidance response by rats is facilitated by shorter ITIs. The purpose of the present experiment was two-fold: first, to study the effect of ITI on the acquisition of a shuttlebox avoidance response in fish, and, second, to compare the effect of the ITI on fish with the data available on the rat. Subjects The Ss for this study were 79 experimentally naive Mexican Swordtails, Xiphophorus hellai. They were obtained locally and housed for several weeks prior to the experiment in 20 gallon community tanks. Apparatus The aquatic shuttlebox used in this study was built from an 18 in. Betta splendens display tank enclosed in a plywood housing. The tank was divided into two compartments of equal size by a gray Plexiglas door­ way. The CS was the onset of a 25 w light at the end of the compartment occupied by S at the beginning of each trial. The US was an electric shock delivered through a pair of electrodes set in the ends of the tank. The shock was developed from a 110 v ac source, passed through a variable resistor, and pulsed by a