Contextual conditional discriminations

Three experiments used a discriminated operant procedure to study conditional discrimination learning in rats. The first experiment showed that rats were capable of learning a biconditional discrimination in which two contexts served as conditional cues signalling the reinforcement contingencies associated with two discriminative stimuli. The discrimination was learned equally well when one discriminative stimulus signalled food, the other its absence, and when one stimulus signalled food, the other extinction plus mild footshock.

In Experiment 2 it was shown that prior training on such a conditional discrimination enhanced the subsequent context specificity of simple conditioning relative to control groups of animals for whom the prior training had not been conditional. Experiment 3 showed that a reversal of the significance of one pair of discriminative stimuli produced no spontaneous reversal in performance to a second, target, pair.

The pattern of results is best accounted for by an analysis of contextual conditional discrimination learning in terms of stimulus configurations and offers no support for the notion that rats may learn a general conditional rule or set.     

Second-order schedules of token reinforcement: effects of varying the schedule of food presentation

In the initial link of a complex schedule, one discriminative stimulus was presented and lever pressing produced tokens on fixed-ratio schedules. In the terminal link, signalled by a second discriminative stimulus, deposits of the tokens produced food. With two rats, the terminal link was presented after each sixth component schedule of token reinforcement was completed. With the other two rats, the terminal link was presented following the first component schedule completed after a fixed interval. During the terminal link, each token deposit initially produced food. The schedule of food presentation was subsequently increased such that an increasing number of token deposits in the terminal link was required for each food presentation. Rates of lever pressing in the initial link were inversely related to the schedule of food presentation in the terminal link. These results are similar to those of experiments that have varied schedules of food presentation in chained schedules. Rates and patterns of responding controlled throughout the initial link were more similar to those ordinarily controlled by second-order brief-stimulus schedules than to those controlled by comparable extended chained schedules.     

Effects on responding of mixed and multiple schedules of signalled and unsignalled response-dependent electric-shock delivery

Responding in two rats was maintained under mixed and multiple variable-interval 35-sec variable-interval 35-sec food delivery schedules. Similar rates and patterns of responding occurred in each component of the two schedules. Mixed and multiple variable-interval 65-sec variable-interval 65-sec schedules of response-dependent shock delivery were super-imposed on the mixed and multiple baseline food schedules, respectively. In one component, a 5-sec stimulus was presented on the average of once every 65 sec. Offset of the stimulus arranged that the next response would produce shock. In the other component, no stimulus was presented during the 5-sec period. The mixed schedule of signalled and unsignalled dependent shock delivery yielded similar degrees of response suppression in each component, but the multiple schedule of shock delivery revealed differential degrees of response suppression. Considerably more suppression occurred in the component not associated with the preshock stimulus, thus implicating the discriminative functions of the correlated stimulus.     

Preference and response suppression under different correlations between shock and a positive reinforcer in rats

A lever-press response by rats was reinforced by food in two successively presented types of trial signalled by different discriminative stimuli. When responding was punished by a shock in one type of trial, a groups for which the shock always preceded the reinforcer by .5 sec (positive correlation) showed less suppression in those trials than a group which received the shock and food separated in time (negative correlation). When, in a second experiment, rats were given a choice between food alone in one end of a shuttle box and either positively or negatively correlated shock and food in the other on a concurrent schedule, the group receiving the negatively correlated shock showed a greater preference for the food alone end. On the basis of a third experiment in which a tone was substituted for the shock in the choice situation, it was argued that the effect of correlation was not simply due to the stimulus properties of the shock. A final experiment demonstrated that when shock punishment is administered during extinction of the lever-press response, the rate of extinction is slower if the shock has been previously paired with the food reinforcer. Pairing a shock with food seems to attenuate the intrinsic aversiveness of the shock through Pavlovian counterconditioning.     

Instrumental learning as a result of omission training on wheel running

The present experiment sought to provide unequivocal evidence of instrumental learning under omission training. Hungry rats received free food reinforcement while spontaneously running in a wheel. For an omission group, running postponed or cancelled reinforcers in the presence of a discriminative stimulus (SD) requiring subjects to reduce responding to earn food. Background food presentations were then yoked to reinforcement delivered in the presence of the discriminative stimulus. For a control group, which received the same stimulus presentations, reinforcement delivery was yoked to the experimental group at all times. The procedure allowed both within- and between-subject comparisons between omission and response-independent schedules. The response-reinforcer delay under the omission contingency was adjusted so as to equate reinforcement frequency in the presence and absence of the SD. As the SD was not correlated differentially with reinforcement and the running response did not involve approach or withdrawal to the site of food delivery, the successful discrimination performance observed in this experiment cannot be accounted for by appeal to implicit classical conditioning. Instead, it is suggested that decreased running in the presence of the discriminative stimulus was based on the animals' veridical representation of the negative contingency between the response and reinforcement.     

Airflow as a discriminative stimulus

In one experiment, pigeons were taught to discriminate airflow by having availability of reinforcement signalled by its presence and extinction signalled by its absence. After they reached criterion, some were trained on a discrimination reversal. Others were trained on an intradimensional discrimination with a low airflow velocity associated with reinforcement and a higher airflow velocity associated with extinction. All discriminations were learned rapidly, indicating that airflow velocity can function as a discriminative stimulus. In the second and third experiments, naive pigeons were trained to discriminate the presence of a compound stimulus (one of three tonal intensities paired with one of three airflow velocities) from its absence. These pigeons were subsequently given a component stimulus test during extinction on four stimulus values; the two training values, the tone alone, and the airflow alone. High or moderate velocity airflow controlled more responding than any of the three tone intensities. However, low velocity airflow controlled more responding only when a low intensity tone was employed.     

Conditioned inhibition from incomplete reductions in the probability of reinforcement

Five experiments, all using appetitive, discrete trial operant conditioning, studied the properties of a stimulus that signalled a reduction in the probability of reward. Discriminations were trained in which reinforcement was available on 100% of trials signalled by a tone, but only on some occasions when the tone was presented simultaneously with a light. The properties of the light were assessed in summation tests with a clicker. The first two experiments established that if the tone-light compound signalled reinforcement on only 25%, 33% or 50% of trials the light acted as a discriminative inhibitor, suppressing responding maintained by the clicker. In these experiments reinforcement had been available on 86.7% of clicker trials during initial training. Experiments 3, 4 and 5 examined further the properties of the light after animals had been trained on a discrimination where reinforcement was available on 50% of trials signalled by the tone-light compound. The light was evaluated in a summation test with a clicker, which had signalled only a 25% or 15% probability of a reinforcement. In this instance, the light did not suppress the responding maintained by the clicker. Instead, the light acted as an excitatory discriminative stimulus to enhance that responding. This dependence of the properties of the light on the reinforcement history of the clicker poses problems for most contemporary models of inhibitory performance.     

Rate of response as a visual social stimulus

In Exp. 1, a high rate of responding (chain pulling) of a stimulus monkey was established as a visual positive discriminative stimulus for the operant behavior (bar pressing) of an observer monkey. The terminal performance of the observer under conditions in which a high rate of response of the stimulus monkey alternated in a variable temporal arrangement with a zero rate of response of the stimulus monkey (negative discriminative stimulus) was essentially the same as when nonbehavioral stimuli are correlated with the availability of reinforcement. By manipulating the schedule of reinforcement to change the rate of responding of the stimulus subject without changing its rate of reinforcement, Exp. 2 showed that the effective behavioral stimulus for the observer was the rate of chain pulling by the stimulus subject. A novel intermediate rate of responding by the stimulus monkey resulted in an intermediate rate (generalization) on the part of the observer during an extinction test. These experiments demonstrated that the rate of responding of one organism can function as a discriminative stimulus to control the rate of responding of another organism; and that the rate of responding is similar to other physical stimuli in terms of discrimination and generalization.     

Reinforcement contingencies maintaining collateral responding under a DRL schedule

Two-key conjunctive schedules were studied with one key (food key) under a differential-reinforcement-of-low-rate 20-sec schedule, while the consequences of responding on another key (collateral key) were varied. When food depended not only upon a food-key interresponse time in excess of 20 sec, but also upon the occurrence of one or more collateral-key responses during the food-key interresponse time, the rate of collateral-key responding was low and food-key interresponse times rarely exceeded 20 sec. When collateral-key responses could produce a discriminative stimulus correlated with the availability of food under the DRL schedule, the discriminative stimulus functioned as a conditioned reinforcer to maintain higher rates of collateral-key responding, and the spacing of food-key responses increased. If the occurrence of the discriminative stimulus was independent of collateral-key responses, the rate of collateral-key responding was again low, but the spacing of food-key responses was still controlled by the discriminative stimulus. Both the conditioned reinforcer and the explicit reinforcement contingency could maintain collateral-key responding, but the adventitious correlation between collateral-key responses and the delivery of food could not maintain very much collateral-key responding. The pattern of responding on the food-key was determined to a much greater extent by the correlation between the discriminative stimulus and the delivery of food than by the pattern of responding on the collateral key.     

The relative aversiveness of signalled vs unsignalled avoidance

Subjects avoided shock by pressing on one lever under an unsignalled condition, but by pressing a separate lever they changed the condition to signalled avoidance for 1-min periods. Signalled avoidance periods were identified by a correlated stimulus. All eight subjects responded to change the unsignalled schedule to a signalled one. Once contact with signalled avoidance was made, subjects continued responding to remain in that condition. Other tests showed that changeover responding was greater when the correlated stimulus was presented without the signal than when the signal was presented without the correlated stimulus. An analysis based upon shock and shock-free periods is presented.     

Effects of associatively significant posttrial stimuli on learning and memory

Pigeons were trained on a delayed conditional discrimination in which the choice between two simultaneously presented stimuli depended on how the trial started. Choice of one of the stimuli was reinforced if the trial had been initiated by presentation of a food sample and choice of the other was reinforced if no sample had been presented. Subsequently, test trials were administered on which an associatively significant stimulus was presented during the retention interval. This manipulation was intended to modulate the short-term retention of information about the food sample. It was found that performance on food sample test trials was enhanced by presentation of an excitor for food, disrupted by presentation of an inhibitor for food and unaffected by presentation of an associatively neutral stimulus. The impact of these posttrial stimuli was also assessed on the ability of the food sample to serve as a reinforcer. This was done by recording the development of responding to a keylight that signalled the food sample on these test trials. Compared to the associatively neutral stimulus, both the excitor and the inhibitor interfered with the development of keypecking. These results are discussed with regard to the issue of how posttrial events modulate associative learning.     

Further Investigation of Responding Elicited by BC and C After A+, AB-, ABC+ Training

Two experiments with pigeons explored conditioned keypeck responding to new visual targets after visual compound discrimination training. In the first experiment, pigeons were trained with an A+, AB-, ABC+, AD-, ADC+ task, in which stimulus A signalled food, stimulus compounds AB and AD signalled no food, and stimulus compounds ABC and ADC signalled food. According to both an elemental model (Rescorla & Wagner, 1972) and a configural model (Pearce, 1987) of Pavlovian conditioning, test compounds BC and DC should elicit less responding than should C alone. However, the birds responded more to BC and DC than to C. In the second experiment, another set of pigeons was trained with an A+, AB-, ABC+, AD-, ADE+ task, in which stimulus A signalled food, stimulus compounds AB and AD signalled no food, and stimulus compounds ABC and ADE signalled food. Contrary to the prediction of the Rescorla-Wagner model, keypeck responding was not less on BC and DE trials than on C and E trials in testing. However, B and D attenuated responding to E and C, respectively, when presented in compounds BE and DC. The Pearce model was able to accommodate these results.     

Analysis of the pavlovian properties of signals for punishment

Rats learned to suppress lever pressing in the presence of a compound stimulus that signalled either response-independent shock (conditioned suppression) or response-dependent shock (punishment). Suppression to one element of this compound was blocked if the other element had previously signalled the same contingency as that holding in the presence of the compound. Two experiments, however, failed to find significant evidence of blocking if the pretrained element had signalled conditioned suppression and the compound signalled a punishment contingency. On the other hand, if one stimulus signalled punishment of one response it was able to block the acquisition of control by the second stimulus when the compound signalled punishment of a different response. Finally, although animals showed no suppression to one element of a compound signalling conditioned suppression if the other had previously signalled punishment, this apparent blocking effect concealed that what they learned was that the added element signalled the cancellation of the instrumental contingency. These results provide little support for the view that stimuli signalling punishment contingencies suppress responding in whole or in part by virtue of their Pavlovian relation to shock.     

A note on behavioural contrast and frustration

The suggestion has been made that positive contrast, the increase in response rate in one successively-presented stimulus following a change in conditions that decreases responding in the other, may not depend on differential stimulus control. Evidence is reviewed suggesting that contrast effects on most multiple interval schedules do depend upon discrimination, but that effects similar to contrast that result from the omission of reinforcement on fixed-interval schedules (the frustration/omission effect) reflect other factors. The link between contrast and the omission effect is the discrimination mechanism which allows animals to respond only at times or places correlated with food delivery. Contrast is a direct result of this mechanism; but the omission effect depends on the difference between the inhibitory discriminative after-effects acquired by food on fixed-interval schedules, and those acquired by a “neutral” stimulus presented in its place.     

Discrimination and differentiation of response number in stimulus directed pecking of pigeons.

In Experiment 1, autoshaping trials terminated with food only if pigeons emitted more than a target number of responses during a trial in one condition and fewer than a target number in another. The median number of responses per trial shifted in accordance wtih the requirements. The responding of yoked-control birds that received response-independent reinforcers did not vary with the response requirements. In Experiment 2, the number of responses in autoshaping trial became the discriminative stimulus for reinforcement in the second component of a chained schedule. In one condition, responding was reinforced only if the number of responses in the first component was above a target value; in the other condition, responding was reinforced only if the number was below the target value. The distribution of the first-component response numbers did not shift systematically between discrimination conditions, but response rates in the second component indicated that the number of responses in the autoshaping trial was a discriminable property behavior.     

Control of responding during stimuli that precede transitions in reinforcement frequency

Pigeons' responses were reinforced on a variant of a mixed variable-interval extinction schedule of reinforcement in which the transition to the higher reinforcement rate was signaled by a trace stimulus projected on the response key prior to the onset of the component correlated with food delivery. In the first of two experiments, the duration of the trace stimulus preceding the component correlated with food delivery was varied from 1.5 to 50.0 s and in the second experiment, the reinforcement frequency in the same component was varied from 10 to 60 reinforcers per hour. Pigeons pecked at the trace stimulus preceding the onset of the component correlated with food delivery even though responding was not reinforced in its presence and only one of the changes in reinforcement rate (i.e., from extinction to reinforcement) was signaled. The rate of pecking during the trace stimulus was a function of its duration but not of the reinforcement frequency in the following component. Higher rates generally occurred at the shorter trace-stimulus durations. Component responding following the offset of the trace stimulus was under discriminative control of the trace stimulus whether or not responding occurred in the presence of the trace stimulus.     

Discriminative responding on associated mixed and multiple schedules as a function of food and ICS reinforcement

Performance on associated mixed and multiple variable-interval-extinction schedules was studied as a function of food versus intracranial stimulation (ICS) reinforcement. For the mixed schedule, differential responding was greater for an ICS reinforcement group than for a food reinforcement group, demonstrating that conditions affecting resistance to extinction help to determine the control exerted by a mixed schedule. Performance on the multiple schedule demonstrated greater differential responding for an ICS group than for a food reinforcement group during the early training sessions, indicating that the control exerted by mixed schedules interacts with that exerted by the exteroceptive discriminative stimuli. The results suggest that the influence of the associated mixed schedule on discriminative responding would be greater, the greater the difficulty of the stimulus discrimination.     

Differential autoshaping to common and distinctive elements of positive and negative discriminative stimuli

The learning by hungry pigeons of a discrimination between two successively presented compound visual stimuli was investigated using a two-key autoshaping procedure. Common and distinctive stimulus elements were simultaneously presented on separate keys and either followed by food delivery, S+, or not, S−. The subjects acquired both between-trial and within-trial discriminations. On S+ trials, pigeons pecked the distinctive stimulus more than the common stimulus; before responding ceased on S− trials, they pecked the common stimulus more than the distinctive one. Mastery of the within-display discrimination during S+ trials preceded mastery of the between-trials discrimination. These findings extend the Jenkins-Sainsbury analysis of discriminations based upon a single distinguishing feature to discriminations in which common and distinctive elements are associated with both the positive and negative discriminative stimuli. The similarity of these findings to other effects found in autoshaping—approach to signals that forecast reinforcement and withdrawal from signals that forecast nonreinforcement—is also discussed.     

The control of appetitive instrumental responding does not depend on classical conditioning to the discriminative stimulus

In Experiment I, rats were exposed to a classical relationship between a clicker-light compound and response-independent food. Conditioning to the light was blocked if the clicker had previously served as a classical signal for food, but not if it had been established as a discriminative stimulus for food-reinforced lever pressing. In Experiment II, a tone-light compound served as a discriminative stimulus for lever pressing. Control by the light was blocked if the tone was independently trained as a discriminative stimulus, but not if it was trained as a classical signal for response-independent food. These results suggest that discriminative stimuli do not come to control appetitive instrumental responding by virtue of their implicit classical relationship to the instrumental reinforcer.     

Time allocation in concurrent schedules: the effect of signalled reinforcement

The responses of five pigeons were reinforced on concurrent variable-interval variable-interval reinforcement schedules in which changeover key responses changed the stimulus and reinforcement schedules associated with the food key. While the reinforcement availability in one component remained unchanged throughout the experiment, the reinforcement availability in the other component was, during several conditions, signalled by the onset of an additional discriminative stimulus. During unsignalled conditions, both the relative frequency of responding and the relative time spent in each component approximated the obtained relative reinforcement frequency in each component. The effect of signalling reinforcer availability in one component was to (1) reduce responding in the signalled component to near-zero levels, and (2) increase the relative time in the unsignalled component, without a corresponding increase in the obtained relative reinforcement frequency. The magnitude of the increase in relative time in the unsignalled component decreased as the overall frequency of reinforcement increased. This deviation in the matching relation between relative time and the obtained relative reinforcement frequency was eliminated if the overall reinforcement frequency was increased before the signal was introduced and then, without removing the signal, gradually reduced.