The molarity of molecular theory and the molecularity of molar theory.

Dinsmoor (2001) rejects shock-frequency reduction as a reinforcer for avoidance behavior, and considers this to be an invalidation of so-called molar avoidance theory. This is a narrow view of operant avoidance theory, for which shock-frequency reduction is by no means the only reinforcer.     

Explaining avoidance: two factors are still better than one.

Two-factor theory remains a viable account of avoidance behavior. By emphasizing the interplay of respondent and operant contingencies, two-factor theory encourages the analysis of stimuli that mediate molar consequences and incorporates control by local events as well as events that are temporally remote, improbable, or cumulative.     

Still No Evidence For Temporally Extended Shock-frequency Reduction As A Reinforcer.

There is no consensus and very little overlap in the criticisms of my target article. Because the primary consequences of avoidance behavior are by definition alterations in the distribution of shocks in time, any theory about the reinforcement of such behavior necessarily must begin with that dimension. However, the safety‐signal version of two‐process theory calls on positively and negatively correlated stimuli, including the responses themselves serving as stimuli, to transmit the effects of those alterations to the relevant behavior. Meanwhile, the Herrnstein—Hineline single‐process theory hypothesizes an additional source of reinforcement: a direct effect of reduction in the density of shock over some extended period of time. I can find no data that selectively support that hypothesis.     

Beyond the molar-molecular distinction: we need multiscaled analyses.

Dinsmoor's (2001) adherence to molecular analyses may require him to assert that molar and molecular principles are mutually exclusive, but to instead analyze the phenomena of avoidance as inherently multiscaled is to follow a well-established practice in the natural sciences. Besides the issue of scale, two-factor theory, which Dinsmoor advocates, has little to say about some important and longstanding results in experiments that qualify as avoidance.     

The puzzle of responding maintained by response-contingent shock.

Four squirrel monkeys were first exposed to a sequence of procedures that reliably generate responding maintained by brief response-contingent electric shocks arranged according to a fixed-interval schedule. After responding had become stable on the fixed-interval schedule, additional contingencies were added in tandem, whereby after completion of the interval, the spacing of responses affected shock delivery. In one procedure, responses had to be spaced more widely than their previous median value if shock were to be delivered. In the other procedure, responses had to be spaced more closely to produce shock. On the first of these procedures, decreased but stable responses rates would indicate that shock functioned as a positive reinforcer; on the second, increased response rates would indicate the positively reinforcing function. Instead, response rates accelerated on the procedure that targeted more widely spaced responses for shock delivery, and decelerated or ceased on the procedure that arranged for shocks to be produced by more closely spaced responses. Consistent with other recent findings, these results question the interpretation of performances maintained by response-contingent shock as engendered by positive reinforcement and are consistent with aversive-control interpretations. The details of that aversive control are not entirely clear, however, and these same procedures would be informative if applied to shock-maintained behavior that is generated in other ways.     

Schedules of food postponement: II. Maintenance of behavior by food postponement and effects of the schedule parameter

In Experiment I, food-deprived, feeder-trained squirrel monkeys pressed a lever to postpone brief electric shocks (Response-Shock=Shock-Shock interval=30 seconds). Forty-one three-hour sessions of shock postponement were followed by 120 sessions of concurrent shock and food postponement. The shock schedule was unchanged and the food schedule was Response-food interval–20 seconds, Food-food interval 10 seconds. After concurrent shock and food postponement, the shock schedule was discontinued and 40 sessions of food postponement ensued, followed by 53 sessions of extinction. After extinction, food postponement was resumed for 11 sessions. Stable responding with low food rates was maintained under food-postponement after the concurrent schedule. Responding decreased to low levels under extinction and recovered immediately to previous levels when the food-postponement schedule was re-instated. In Experiment II, a parameter of the food-postponement schedule was studied sequentially. Using the same subjects, the Response-food–Food-food interval was manipulated from four seconds to 80 seconds with several orders of presentation. Relations of response rates and food rates to the parameter were similar to those seen under shock postponement. Exposure to very short postponement times (four seconds), resulting in very high food rates, decreased but did not abolish subsequent responding at longer postponement times. Results are discussed from the point of view that reinforcing functions of stimuli consequent on responding depend on a prior history of scheduled contact with those stimuli.