MOLAR AND MOLECULAR CONTINGENCIES AND EFFECTS OF PUNISHMENT IN A HUMAN SELF-CONTROL PARADIGM

An attempt was made to establish a human self-control paradigm where impulsive response binges produced a greater reinforcement density at the molecular, or short term, level but where self-control responses produced a greater reinforcer density at the molar, or long term, level. Humans earned points by pressing two buttons. Pressing one of the buttons, the impulsive choice, produced 2 points over 1 s. Pressing the other button, the self-controlled choice, produced 10 points over 1 s after a delay of 19 s. In Experiment 1, subjects made 30 choices in a delay condition and 30 choices in the no delay condition. In the delay condition reinforcement of an impulsive response was followed by a 19-s postreinforcer delay that made trial durations equal for both impulsive and self-control choices. In the no delay condition there were no postreinforcer delays following impulsive choices. However, there was a wait period after every 10 trials equal to 19 s times the number of impulsive choices on the previous 10 trials. Thus, in the no delay condition at molecular level, impulsive responding produced a reinforcement density that was greater than the reinforcement density produced by self-controlled responding. However, because of the wait periods, at the molar level self-control responding produced the greatest density of reinforcement. In the delay condition self-controlled responding produced the greatest density of reinforcement at both the molar and molecular levels. Subjects exposed to the no delay condition first responded impulsively during the no delay condition but in a self-controlled fashion during the delay condition. Subjects who received the delay condition first were self-controlled during the delay condition and maintained self-controlled responding during the no delay condition. In Experiment 2 the no delay paradigm was used for 60 trials with punishment and no punishment conditions. In the punishment condition, which occurred either during the first or final 30 trials, following an impulsive response, ''NO!'' would appear on the screen concurrent with a 0.5-s, 92-db tone. Subjects who experienced the no punishment condition first tended to be impulsive during the no punishment phase but became self-controlled during the punishment condition. Subjects exposed to punishment in the first 30 trials were self-controlled during punishment and maintained self-controlled responding during the no punishment condition. These results suggest that once preference for self-control is developed, self-control responding may be maintained even when impulsive responding produces a greater density of reinforcement at the molecular level.