Association for Behavior Analysis International

Rather than refers merely to the exploration of different values of an independent variable, the parametric perspective alluded in the title of the present symposium refers to the integration of different experimental results on the basis of their quantitative and/or functional continuities. This theoretical strategy is based on the philosophy of temporally-defined schedules proposed by William Schoenfeld and his coworkers. Four experiments following such perspective will be described in the present symposium. Using a self-controlled eating behavior paradigm, vila and Ortega observed that responses that are incompatible with impulsive behavior can be trained as any other operant response. Flores, Mateos and Killen show that the typical decremental effect of delay of reinforcement depends on the value of the inter-reinforcement interval. Similarly, using temporally-defined schedules, Torres, Hernndez, and Silva observed that stimulus control increase or decreases as the length of the discriminative stimulus is shortened. Finally, Serrano and Blanco observed that inter-dimensional signals produced a higher percentage of correct responses than intra-dimensional signals under a limited-hold, two-choice conditional discrimination procedure.

Self-controlled eating behavior occurs when a subject does not eat food available until a pre-established criterion is reached. It was suggested that to emit a distracting activity while food is freely available could facilitate the occurrence of self-controlled eating behavior. This possibility was evaluated with twelve pigeons that were exposed to a 64 s time cycle (T) in which food was presented for 3 (SR1) and could be presented for other 3 s once the T cycle elapsed (SR2), according to the following contingency: trying to eat during SR1 interrupted it and cancelled SR2 presentation; otherwise, eating behavior during SR2 could occur. According to a factorial design the contribution of the following variables was explored: 1) Previous training or not training of key pecking. 2) An ABA or a BAB order of presentation of the following three conditions: A) SR1 presentation signaled by a 3 s change in the key color from red to green, B) SR1 presented without any signaling stimulus, A) SR1 presentation signaled again. Each condition was in effect for 20 sessions of 50 T cycles each. The previous key-pecking training resulted in a low or high level of the number of SR1 interruptions (A or B conditions) and a correspondingly high level of R>0 in the A conditions. Without previous key-pecking training there was no effect of signaling SR1 presentations, regardless of the ABA or BAB sequence. It was concluded that self-controlled eating behavior was facilitated by explicit training of a distracting activity that occurs concurrently with food availability.

In a variety of operant and classical procedures, the effects of delays between environmental stimuli are reduced when the overall time between reinforcers is longer. This modulation of delay effects by the inter-reinforcement interval was referred by Williams (1998) as the relative time effect. The present study was designed to explore the effects of different inter-reinforcement interval (IRI) duration on food accumulation. For four rats the IRI was 40 s, whereas for another four rats was 120 s. On successive phases (30 sessions each) the delay of reinforcement was lengthened (0, 2, and 8 s). For all rats the number of responses-obtained food pellets increased as the delay was lengthened. The number of responses was higher in the subjects with the long IRI (120 s) than the subjects with 40 s IRI. It was concluded that IRI duration controls the number of responses as relative time effect.

Six rats were exposed to a 60 s long temporally-defined schedule. Each T cycle was divided in two 30 s long sub-cycles: tD and t?. Reinforcement probability was 1.0 for the first response in tD and 0.0 for any response in t?. A visual stimulus (green light) was added to tD sub-cycle while no signal was added to t?. The length of the stimulus was varied between phases (30, 3, 0.3, 0 s) and the order of presentation of such values was inverse between triad of rats. At the end of the experiment each triad was exposed to a redetermination condition similar to the first phase in each case. Regardless of triad, response rate along t? was inversely related to the length of the stimulus in tD. A proportional relationship between the length of the stimulus and the post-reinforcement pause was also found. Results are discussed in terms of inhibitory effects of reinforcement and observing responses.

Triads of rats were exposed to a two-choice conditional discrimination procedure whose signals were visual, auditory or visual and auditory. Signals were differentially correlated with water and tapioca pellets for half of the rats, while pellets and water were randomly presented for remaining rats. Performance was higher under visual and auditory signals than under visual or auditory signals, in that order. The correlation between signals and reinforcers did not enhance performance but negatively affected the accuracy on water trials. Results are discussed in relation to previous findings under simple discrimination contingencies and the so called differential outcomes effect.