No abstract

Water drinking behavior of albino rats was studied as a function of two schedules of food reinforcement and one schedule of shock avoidance. Considerable water drinking occurred concurrently with the food reinforced response but none with the shock avoidance response. This difference might be attributed either to the use of food instead of shock or to the presence of timing behavior in conjunction with the food schedules but not the shock schedules. Problem Falk (1961) has reported that rats on a variable interval schedule of food reinforcement for bar pressing show a very high rate of drinking when a water bottle is placed in the test chamber. He has termed this phenomenon "psychogenic polydipsia" since the excessive intake of water is dependent on psychological factors, i.e. the schedule of food reinforcement for the bar-press response. It should be emphasized that the food reinforcement is not contingent on the water drinking response. Segal (1963) has demonstrated this phenomenon using the schedule of differential reinforcement of low rates. In the present experiment an attempt was made to replicate this phenomenon, not only with the above schedules of positive reinforcement, but also with a schedule of negative reinforcement. Method Six naive male albino rats were used. They were all food deprived and fed at the end of each daily 50 min. session. They were maintained at approximately 80% of normal body weight. A Grason-Stadler test chamber, shock generator and scrambler were used. Food reinforcements consisted of 45 mg Noyes pellets. The spout of the water bottle protruded through the right hand side of the intelligence panel into the test chamber. A Lehigh Valley Electronics drinkometer was used to record the drinking response. Two Ss were reinforced on a variable interval schedule with a mean of 64 sec.; two Ss were reinforced on a differential reinforcement of low rate schedule in which the required pause between responses was 8 sec.; and two Ss were reinforced on a non-cued avoidance schedule in which both the time between uninterrupted shocks and the time between an avoidance response and the next shock was 16 sec. The shock was .5 rna and lasted for .2 sec. Results The behavior of the food-reinforced animals was essentially the same as that previously reported by Falk (1961) and Segal (1963). The median numbers of drinking responses per session during the last three sessions of exposure to these schedules (sessions 16-18) were 4287 and 1279 for the animals on the variable interval schedule and 6586 and 5036 for the animals on the low rate schedule. Neither of the animals on the avoidance schedule emitted any drinking responses during these sessions. Discussion In non-cued avoidance experiments with rats, two different types of avoidance behavior are sometimes observed. In some instances, a sort of timing behavior develops in which the bar-press responses are so effectively spaced that most of the shocks are avoided and the average interresponse time is approximately the same as the amount of time the ...

A brief summary of the first report (Falk, 1961a) on schedule-induced polydipsia will suffice to introduce this phenomenon. Fourteen normal rats were maintained at 70-80 percent of their free-feeding body weight by limiting their intake of food. The animals earned most of their daily food ration by bar-pressing on a variable-interval one-minute (VI 1 min) schedule for 45-mg Noyes food pellets during 3.17-hour daily experimental sessions. On this schedule, a barpress is reinforced by a food pellet at varying times from a previous reinforcement (from a few seconds to two minutes), the average interreinforcement time being one minute. Each animal performed in a sound-attenuated box, with the automatic control and recording equipment in an adjoining room. No external environmental cue informed the animal exactly when a bar-press would yield a pellet. At the end of each daily session, the animal was returned to its home cage and given any food supplement necessary to maintain body weight at the value selected.Water was always available. During the session licks from a calibrated water reservoir were recorded electronically. In the home cage, any water drunk from a calibrated animal drinking tube between sessions was measured.On this schedule a typical behavior pattern rapidly developed. After earning a food pellet, the rat would consume it and drink about 0.5 ml of water. During the 3.17-hour session, intake averaged 92.5 ml, or 3.43 times the preexperimental, 24-hour water intake level. Between sessions, although water was available, intake averaged less than 1 ml.Such a phenomenon is strange and unprecedented, for the animals are drinking approximately one-half their body weight in a few hours. Water deprivation, heat stress, or osmotic-loading techniques do not approach comparable stimulation of water intake. Under normal laboratory maintenance conditions, daily water intake levels remain rather constant, and even strong facilitating stimuli induce only moderate increases.The evidence will be examined to determine whether this polydipsic phenomenon can be explained either on physiological grounds or with respect to simple behavioral considerations. The variables of which the polydipsia is a function will be summarized as far as these are known at present. A theoretical explanation in terms of adjunctive behavior will be given, and methodological implications and cautions for thirst research attempted. PHYSIOLOGICAL CONSIDERATIONS Food Deprivation and Water IntakeThe sole, unusual physiological condition of existence for these animals is their maintenance at 80 percent of free-feeding weight by limiting food intake. How-* Annals New York Academy of Sciences ever, food deprivation decreases, rather than increases, water intake. I found that animals maintained at 80 percent body weight in their home cages drank less water than when allowed unlimited access to food (Falk, 1964, p. 98). Strominger (1946 reported similar results, as has Kutcher (1967). Recently, I allowed animals held at 80 percent body weight to consum...

Rats were trained to bar-press for Noyes pellets on an FI schedule which was increased serially through several values from 2 sec to as high as 300 sec. Concurrently, water was freely available. As FI length was increased, the degree of polydipsia increased linearly to a maximum value.