Printer Friendly

Appetitive backward conditioning in pigeons.

The nature, reliability, and even the very existence of Pavlovian backward conditioning comprise one of the oldest debates in modern learning theory. From a theoretic point of view, it frequently has been argued that backward conditioning either results in no learning (the basis for its use as a control condition for forward conditioning), or yields inhibitory associations owing to the negative forward relation between a backward conditioned stimulus (CS) and future unconditioned stimulus (US) presentations. Nonetheless, Spetch, Wilkie and Pinel (1981), in a critical review of the empirical literature concluded that there is ample evidence for backward conditioning, particularly when aversive unconditioned stimuli and a small number of conditioning trials are employed. Data from two recent studies are consistent with this generalization (Ayres, Haddad, & Albert, 1987; Matzel, Held, & Miller, 1988).

Somewhat different findings were reported by Hearst (1989) who found evidence for backward conditioning in pigeons when an appetitive US was used. Furthermore, strength of backward conditioning was not influenced by amount of backward training (50-350 trials). Additional procedural variables distinguish this study from its predecessors, including the use of pigeon subjects; an autoshaping preparation; a relatively large number of backward conditioning trials (minimum = 50 trials); relatively closely spaced trials; and a discrimination procedure during backward training and subsequent test phases. Because Hearst's procedure represents a substantial departure from those of other studies showing excitatory backward conditioning, we conducted the present study to replicate Hearst's training procedure, and to test the generality of his findings by employing a within-subject design, and three additional tests for backward conditioning.

In Hearst's study (Experiment 4), pigeons were exposed to two backward CSs that differed in their backward relation with food presentation (the US). The 6-s positive backward CS was immediately preceded by a 3-s US presentation. The 6-s negative backward CS was immediately preceded by a 3-s illumination of the hopper light without the US. A different group of birds was exposed to a control procedure in which the two CSs were uncorrelated with the two hopper conditions. Evidence for backward conditioning was sought in a resistance-to-reinforcement transfer test. For different groups of birds, the positive or the negative backward cue was paired with grain presentation in a forward arrangement, whereas the other cue was paired with the hopper light alone. For birds in the consistent group, the positive backward cue was paired with food during the transfer test; for birds in the inconsistent group, the negative backward cue was paired with food in testing. On three measures of performance, birds in the consistent group generally showed more learning than birds in the inconsistent group or in a random control group. These findings were attributed to the differential backward conditioning histories across groups.

In the present study, birds were trained under the backward discrimination training procedure used by Hearst, but tests for backward conditioning were conducted using a within-subject rather than a between-group design. In light of recent evidence of differential sensitivity to conditioning among behavioral assays (e.g., Brown, Hemmes, Cabeza de Vaca, & Pagano, 1993; Hearst, 1987; Hemmes, Brown, & Cabeza de Vaca, 1990; Matzel et al., 1988; Tait & Saladin, 1986), several additional postacquisition tests were employed, including higher-order conditioning, summation testing, and resistance-to-reinforcement.

After keypecking was acquired to both backward cues under the forward, resistance-to-reinforcement procedure, responding was eliminated by return to the backward training procedure. In accord with the observations of Hearst (1987) and others that behaviorally silent associations may be revealed during extinction (EXT), an EXT test was implemented in the present study. This test also served as an extension of the work by Lindblom and Jenkins (1981), Rescorla (1989), and others, who found that CS-US associations acquired prior to response elimination by means of random or negative CS-US arrangements, were revealed during a subsequent EXT phase. Of interest in the present study is the generality of this effect to response elimination by a backward CS-US arrangement.

Because little-to-no keypecking was expected to occur under the backward or higher-order conditioning procedures, pigeons' proximity to the CS was also recorded as a potentially more sensitive index of learning (Wasserman, Franklin, & Hearst, 1974). Under all phases, differential keypecking or proximity to positive and negative cues was taken as evidence for backward associative learning.

Experiment 1



Six male White Carneaux pigeons with previous experience under trace autoshaping procedures with a red keylight CS were maintained at 75% of their free feeding weights. Two of the birds had also been exposed to delay conditioning with the red keylight. Birds were randomly assigned to two squads, with the restriction that one bird with prior delay conditioning be assigned to each squad. Birds in each squad were randomly assigned to one of three pigeon chambers. All birds in a squad were run simultaneously.


Conditioning was conducted in three standard three-key pigeon chambers (BRS/LVE chamber model SEC-002, 36.8 cm high x 50.8 cm wide x 34.3 cm deep). Three walls of each chamber were white and the surface of the front wall was anodized aluminum. The center key (diameter: 2.5 cm) and hopper aperture (15 cm x 15 cm) were centered on the front wall of the chamber, 24 cm and 12 cm, respectively, from the floor. Side keys were dark and not used.

Pecks to the center key were recorded, as was duration of time spent by the pigeon in proximity to the center key. Proximity was detected by means of an infrared emitter/detector device mounted 2.0 cm below the lower rim of the center key. Output from a light-emitting diode reflective scanner (Squint-Skan, S12 Series, Skan-A-Matic) was amplified and fed to circuitry that delivered an output when reflected light reached a threshold level. The device was calibrated prior to each session so that proximity was recorded only when the pigeon was directly in front of the key at a distance no greater than 4.5 cm. Keypeck activity was not always associated with proximity detection because pecks could be delivered from a position lateral to the key, beyond the range of the detector.

Visual stimuli were presented by means of an in-line projector (IEE, BRS/LVE pattern IC-900-696) mounted behind the center key. White noise and a ventilator fan provided sound masking in the chamber. Experimental events were controlled by an IBM AT microcomputer and CONMAN (Spyder Systems) software with .1 s data acquisition resolution.


General procedure. Under all conditions, sessions commenced with an intertrial interval (ITI) during which only the houselight was illuminated. Fifty-two trials were presented on each session, and the houselight was continuously illuminated. Conditioned stimuli, consisting of colors and/or symbols projected on the center key, were 6 s in duration. The US was a 3-s presentation of the illuminated, grain-filled hopper. Intertrial interval duration was variable: Mean duration was 45 s and the range was 15-75 s. Table 1 indicates the sequence of training and test phases which was presented to all birds.
Table 1

Experiment 1: Training and Test Phases

Condition Sessions

Backward Conditioning BC1 7
Higher-Order Test HO 2
Forward Conditioning FC 10
Backward Conditioning BC2 2
Summation Test SUM 3
Backward Conditioning BC3 2
Resistance-to-Reinforcement Test RR 7
Backward Conditioning BC4 7
Extinction Test EXT 4

Pretraining. Pigeons were magazine trained when necessary and exposed to two sessions of 52, unsignaled US presentations, followed by EXT sessions for the previously trained red keylight. Extinction was run until no pecks were recorded during an entire session, for a minimum of three sessions. After EXT to the red keylight, birds were again magazine trained (if necessary) and exposed to one additional session of unsignaled US presentations.

Backward conditioning. On backward discrimination training two types of trials, 26 of each, were presented during each session. On positive backward trials, the US was presented for 3 s, that is, the hopper was illuminated and raised. Simultaneously with US offset, the positive backward CS was presented. For three randomly chosen birds, the positive backward CS was a circle on an otherwise dark key. For the other three birds, an "X" was used. During negative backward trials, the hopper was illuminated, but not raised, for 3 s, followed by onset of the other, negative backward CS. Trial type alternated randomly with a restriction that no more than three of the same type of trial occur in succession.

Higher-order conditioning. During two sessions of higher-order conditioning, 6-s presentations of a blue or a yellow cue were immediately followed by onset of the positive backward CS for 6 s. The other higher-order cue was paired with the negative backward CS. For four birds, the yellow light was paired with the positive backward cue, and for two birds the blue light was paired with the positive backward cue. Neither food nor the hopper light was presented during this phase. Order of trial presentation was arranged as in the previous phase. Prior to the first higher-order conditioning session, 10 backward conditioning trials, five of each type, were presented.

Forward conditioning and summation testing. During 10 forward conditioning sessions, both higher-order cues were paired in a forward fashion with presentations of the illuminated, grain-filled hopper. In each session, there were 26 6-s presentations of the yellow cue followed by 3-s of access to the hopper, and 26 identical trials with the blue stimulus.

The original backward conditioning procedure was reinstated for two sessions followed by summation testing during the next three sessions. On summation trials one backward CS (the circle or the "X") and one forward CS (blue or yellow light) were superimposed on the center key. Thirteen 6-s trials of each possible combination of backward and forward cues were presented in a quasi-random order in each session. A given stimulus combination was presented on no more than two consecutive trials. Food was not presented during these sessions. Prior to the first session of summation testing, 10 backward conditioning trials, 5 of each type, were presented, followed by 10 forward conditioning trials, 5 of each type.

Resistance to reinforcement. The original backward conditioning procedure was reinstated for two sessions, followed by seven resistance-to-reinforcement test sessions. On test sessions, the positive and negative backward cues were paired with food in a forward fashion.

Response elimination (backward conditioning) and extinction. All birds were returned to the original backward procedure for seven sessions, followed by four sessions under EXT for the backward cues. Extinction sessions were identical to backward conditioning sessions, except that the hopper and hopper light were not presented.

Results and Discussion


Keypecking did not emerge under the backward or higher-order conditioning procedures, and there was no evidence for backward conditioning under the summation test (i.e., rate of responding did not differ for stimulus compounds that included the positive versus the negative backward CS). Evidence for backward conditioning was obtained on the resistance-to-reinforcement tests as shown in Figure 1. Four of the six birds acquired pecking to both cues, and each required a greater number of trials to reach a keypeck acquisition criterion for the negative versus the positive cue. Data in this figure were based on a criterion of three consecutive trials with a peck, and data are shown only for the four of six birds meeting this criterion. One of the remaining birds met the criterion for the positive cue only, and the other for the negative cue only. A t test based on the data presented in the figure yielded a significant difference in acquisition speed to the two cues, t(3) = 5.42, p |is less than~ .05. On the less stringent acquisition criterion used by Hearst, trials prior to the first peck, no significant difference was found. There were no reliable differences in rates of maintained pecking on this or any other phase.


There were no reliable differences in duration of proximity to the positive and negative backward cues during the higher-order, summation, or resistance-to-reinforcement tests. However, evidence for backward learning on the proximity measure was obtained during the EXT condition. Figure 2 represents proximity duration as proportion of time in proximity to each of the backward CSs during the last four phases of this experiment. On EXT, proportion of time proximal to the positive backward cue was higher than that for the negative backward cue. Although this difference was not reliable with a t test, difference scores across the four sessions were in the same direction for all birds (p |is less than~ .05, sign test). This finding extends to backward conditioning the enhancement of discrimination performance during a posttraining extinction test (for a review, see Hearst, 1987). A possible basis for the present performance-enhancing effects of extinction may be seen by examining data from the backward conditioning phases in Figure 2. During backward conditioning, proximity duration was higher for the negative versus the positive backward CS. This effect (reliable only in Phase BC3, t(5) = 2.91, p |is less than~ .05) probably represents hopper-related behavior attending eating on positive backward trials--an effect that would be absent during extinction.

Extinction Recovery

Recovery of responding during the extinction phase was not obtained for either dependent measure. This finding contrasts with those of Lindblom and Jenkins (1981) and Rescorla (1989) who showed recovery during extinction when autopecking was eliminated by means of noncontingent or negatively contingent CS-US arrangements. Apparently the mechanism controlling response elimination under a backward conditioning procedure differs from that evoked by random or explicitly unpaired procedures.

Experiment 2

Because evidence for backward conditioning was not robust in Experiment 1, a second experiment was conducted. In addition to replication of the backward conditioning effects of Experiment 1, a second objective was to provide a more exact replication of the testing procedure employed by Hearst (1989). The third objective was to replicate the somewhat surprising failure to find recovery from response suppression by backward conditioning during extinction.



Six pigeons were maintained at 75% of their free feeding weights. Four had been previously trained under a delay autoshaping procedure employing red, blue, and yellow keylights as CSs. Two other birds were experimentally naive. Birds were randomly assigned to two squads, with the restriction that one naive bird be assigned to each squad. All birds in a squad were run simultaneously.


Training and testing were conducted using the apparatus described for Experiment 1. The naive birds were trained to eat from the hopper in a conditioning chamber that differed from the ones employed in the experimental phases in that the surfaces of the side and rear walls were uncoated stainless steel.


Pretraining. After the naive pigeons were magazine trained, all birds were exposed to two sessions of US-alone presentations in the experimental chambers, followed by EXT sessions for the red, blue, and yellow keylights to which the experienced birds had been previously exposed. Each cue was presented 18 times per session, and order of presentation was random. A minimum of three sessions was run for each bird, and EXT was discontinued after no pecks were recorded for an entire session.

Training and testing. Following this pretraining phase, all birds were exposed to the following sequence of phases: backward conditioning, resistance-to-reinforcement, backward conditioning, EXT. These procedures were identical to those described for the first experiment. A backward conditioning effect was not observed during the EXT phase. Because the effect in Experiment I was observed only after exposure to two previous test phases conducted under conditions of extinction, the foregoing sequence of phases was repeated two additional times as shown in Table 2.
Table 2

Experiment 2: Training and Test Phases

Condition Sessions

Backward Conditioning BC 7
Resistance-to-Reinforcement Test RR 26
Backward Conditioning BC 14
Extinction Test EXT 4
Backward Conditioning BC 4
Resistance-to-Reinforcement Test RR 4
Backward Conditioning BC 4
Extinction Test EXT 4
Backward Conditioning BC 4
Resistance-to-Reinforcement Test RR 5
Backward Conditioning BC 4
Extinction Test EXT 4
Backward Conditioning BC 6
Differential Resistance-to-Reinforcement Test 7

After the third cycle, six additional sessions of backward training were followed by a differential resistance-to-reinforcement test employing the between-group logic described by Hearst. Three birds, one from Squad 1 and two from Squad 2, were assigned to the consistent group. The other birds comprised the inconsistent group. Assignment of birds to conditioning chambers was balanced; that is, in each group, one bird was run in each of the three chambers. For the birds in the consistent group (n = 3), the positive backward cue was converted to a positive forward cue, and the negative backward cue was converted to a negative forward cue. For birds in the inconsistent group the relation between backward and forward training was reversed: The previous positive backward cue was converted to a negative forward cue, and the backward negative cue was converted to a positive forward cue. The forward conditioning procedure was identical to that of backward conditioning except that the temporal sequence of keylight and hopper stimuli was reversed.

Results and Discussion


There were no reliable consistent differences in rate of keypecking to the positive and negative cues within the backward conditioning, resistance-to-reinforcement, or EXT phases. However, as in Experiment 1, speed of acquisition differed across positive and negative cues during the first resistance-to-reinforcement phase, Figure 3 shows number of trials preceding three consecutive trials with a peck for all six birds. Acquisition speed was significantly faster for the positive versus the negative backward cue, t(5) = 2.70, p |is less than~ .05.


Figure 4 shows group mean proportion of time proximal to the backward CSs across the repeating 4-phase sequence. A backward conditioning effect is apparent during the resistance-to-reinforcement test phases, as proximity was higher to the positive versus negative backward cue. Although this effect was not statistically significant when difference scores were evaluated across all sessions, F(1,5) = 5.04, the mean discrimination ratio of proximity duration (positive/positive+negative) was significantly greater than 0.5, t(5) = 2.69, p |is less than~ .05.

In an apparent reversal of our finding for Experiment 1, evidence for backward conditioning was not obtained on the EXT phases. Figure 4 indicates that proximity duration was not higher for the positive versus the negative cue. In fact, the reverse appears to be true for the first EXT phase. An ANOVA taken across all repetitions of this condition showed no significant effect for cue, F(1,5) = 4.24, but a reliable cue x phase interaction, F(2,10) = 4.88, p |is less than~ .05. Separate tests for each EXT phase showed that proximity duration was significantly greater for the negative versus the positive cue in the first phase, F(1,5) = 7.25, p |is less than~ .05, with no significant differences for the other two phases. However, inspection of Figure 4 suggests the emergence of control by the positive cue as a function of repeated extinction testing (recall that the backward conditioning effect reported for Experiment 1 was observed only after two previous exposures to extinction).

On the backward conditioning phases, proximity duration was higher to the negative versus the positive CS, as in Experiment 1. Based on all sessions of backward conditioning, there was a significant effect for cue (positive versus negative; F(1,5) = 7.62, p |is less than~ .05).

Response Recovery Test

In this experiment, birds were exposed to three repetitions of the following sequence of phases: response acquisition (RR), response elimination (BC), and extinction (EXT). As in Experiment 1, EXT did not occasion recovery from the response-suppressing effects of backward conditioning for either measure of performance. Therefore, backward conditioning, unlike random or explicitly unpaired procedures, apparently interferes with a previously learned forward association.

Between-Subject Test

The between-subject manipulation, which replicated the testing procedure employed by Hearst, presented each CS in a positive or negative forward relation to the US. The left panel of Figure 5 shows group mean proximity duration (expressed as proportion of time proximal to each cue) for each session of the differential resistance-to-reinforcement phase. The filled symbols show that proximity duration for the positive CS was higher for the consistent versus the inconsistent group. This significant effect, F(1,4) = 11.73, p |is less than~ .05, can be attributed to the differential backward conditioning histories of these groups. For the consistent group, the positive CS previously served as the positive backward CS; for the inconsistent group, the positive CS previously served as the negative backward CS. The unfilled symbols show performance during the negative CS. On this measure, birds in the consistent group showed less proximity than birds in the inconsistent group, a result also predicted by their prior backward training. However, this difference was not reliable. A group x cue x session ANOVA yielded a reliable cue x session interaction, F(6,24) = 30.26, p |is less than~ .01, reflecting the growth of a difference in proximity on positive and negative trials. A significant group x cue interaction, F(1,4) = 37.45, p |is less than~ .01, reflected a difference in proximity between groups on positive trials, but not on negative trials.

Further analysis showed that both groups discriminated between the positive and negative cues. The difference in proximity between positive and negative trials was reliable for the consistent group on all sessions, t(2) = 23.79, p |is less than~ .01. For the inconsistent group, the difference was significant only later in testing, on Sessions 6 and 7, t(2) = 6.67, p |is less than~ .05.

The right-hand panel of Figure 5 shows group mean keypeck rates to each cue during the Hearst replication. Keypeck rate was higher on positive than negative trials for both groups combined, F(1,4) = 15.47, p |is less than~ .05, a difference that developed across sessions (cue x session interaction, F(6,24) = 9.10, p |is less than~ .01). However, there were no reliable group differences on this measure.

Failure to obtain evidence for backward conditioning on the keypeck measure under this procedure is consistent with the other resistance-to-reinforcement tests of this study. In those tests, evidence for backward conditioning was obtained for acquisition of keypecking, but not for maintained keypeck rate. Proximity duration, in contrast, yielded more persistent evidence for backward conditioning. This finding joins others in indicating the importance of obtaining more than one response measure during Pavlovian conditioning procedures (Boakes, 1977; Brown et al., 1993; Buzsaki, 1982; Holland, 1977; Zener, 1937).

General Discussion

In summary, the present findings corroborate Hearst's (1989) demonstration of appetitive backward conditioning as assayed by resistance-to-reinforcement tests. Pigeons acquired keypecking more rapidly to the positive versus the negative backward cue. This finding extends Hearst's results to a within-subject test, and to a resistance-to-reinforcement procedure under which all test stimuli are treated identically. In Experiment 2, overall levels of proximity were higher to the positive versus the negative backward cue on both within-subject and between-group resistance-to-reinforcement tests.

Unlike Hearst's procedure, the present design did not include a control condition under which the two CSs were uncorrelated with hopper condition (filled or empty) during training. Therefore, it could be argued that the positive backward cue acquired excitatory control, or that the negative backward cue became a conditioned inhibitor. In either case, however, the direction of the differences between positive and negative backward cues is not predicted by differences in the prevailing forward CS-US relations (cf. Brown, Hemmes, Coleman, Hassin, & Goldhammer, 1982; Fantino, 1977; Gibbon, Berryman, & Thompson, 1974; Jenkins, Barnes, & Barrera, 1981). The present findings thus imply control by the backward relations between CS and US events. It is possible that backward and forward conditioning effects coexist in a single procedure, and that effects of these procedures summate. This summation could underlie the finding that inhibitory properties of a CS are greatest when the CS is presented in an unpaired fashion, as opposed to presentation in a backward relation with the US (cf. Kaplan, 1984; LoLordo & Fairless, 1985).

No evidence for backward conditioning was obtained on higher-order or summation tests, and the significant effects present during EXT were not replicated between experiments. Therefore, the confirming evidence for backward conditioning under the resistance-to-reinforcement tests must be evaluated in light of the failure to show backward conditioning under other conventional tests for associative strength.

On a procedural level, backward conditioning effects in this study varied with the presence or absence of reinforcement on testing. All but the resistance-to-reinforcement tests were conducted in extinction. The importance of this variable may be understood in terms of a two-dimensional model of conditioning recently put forth by Matzel et al. (1988). They suggested that excitatory associative strength accrues to a CS owing to temporal contiguity between the CS and a US, even under backward or simultaneous training procedures. However, conditioning also imparts sequential learning which, in combination with associative strength, governs response tendency. Behavioral manifestation of the association is present only when the CS also predicts the imminent occurrence of the US, as in forward conditioning. Evidence for excitatory conditioning to a simultaneous or backward CS can be obtained on transfer tests in which a forward relation is arranged. As applied to the resistance-to-reinforcement transfer test in Hearst's and the present studies, this analysis suggests that the differential associative strength acquired by the positive and negative backward cues was expressed when these cues were placed in a forward CS-US arrangement. On the summation and EXT tests, no such forward arrangement was present. Only for the higher-order test does Matzel et al.'s theory fail to predict the obtained results. As in their sensory preconditioning tests, the higher-order |CS.sub.1~ assumes a predictive role for the presumed excitatory positive backward cue. Therefore, approach and possibly pecking should occur to the |CS.sub.1~ paired with the positive backward cue. However, the failure to observe any higher-order conditioning in the present study may be attributable to procedural factors, such as the use of a discrimination procedure during testing, and the absence of reinforced trials to prevent extinction of first-order conditioning.

Inconsistent results on postacquisition tests for backward conditioning have also been reported by Tait and Saladin (1986), who conditioned rabbits using an aversive US (paraorbital shock). Animals were trained to lick a water spout in one test chamber and then given backward conditioning in another chamber. Two tests for associative strength were then conducted: a lick suppression test and a resistance-to-reinforcement test. Unlike the present findings and those of Hearst, the backward CS retarded acquisition under the resistance-to-reinforcement test. In contrast, evidence for excitatory backward conditioning was obtained on the lick suppression test. Tait and Saladin concluded that excitatory and inhibitory associative effects coexisted for the backward CS. Like Matzel et al., they proposed a two-dimensional model to account for their findings. On their version, a CS acquires two independent functions--a motivational function and a cue function--which may govern differing performance tendencies under different tests for associative strength. This model (which is similar to those previously described by Konorski, 1967; Overmier, Bull, & Trapold, 1971; and Rescorla, 1978) makes no a priori predictions regarding performance under the differing tests for associative strength used in the present study; however it is compatible with the results obtained.

Apart from these theoretical issues, recent demonstrations of backward conditioning under a variety of conditions, including an operant paradigm (Zentall, Sherburne, & Steirn, 1992), indicate that an alternate empirical approach may be in order. Rather than focus on the status of appetitive backward conditioning as a phenomenon, research might profitably address the question of controlling variables. For example, a feature common to Hearst's study and the present one was the use of a discrimination training procedure, a departure from more conventional between-subject designs. The aversive US literature suggests that manifestation of backward conditioning may depend upon the nature of the behavioral assay employed. Both training and testing parameters are candidates for investigation in future research employing appetitive or aversive backward reinforcement procedures.


AYRES, J. J. B., HADDAD, C., & ALBERT, M. (1987). One-trial excitatory backward conditioning as assessed by conditioned suppression of licking in rats: Concurrent observations of lick suppression and defensive behaviors. Animal Learning & Behavior, 15, 212-217.

BOAKES, R. A. (1977). Performance on learning to associate a stimulus with positive reinforcement. In H. Davis & H. M. B. Hurwitz (Eds.), Operant-Pavlovian interactions (pp. 67-97). Hillsdale, NJ: Lawrence Erlbaum.

BROWN, B. L., HEMMES, N. S., CABEZA DE VACA, S., & PAGANO, C. (1993). Sign and goal tracking during delay and trace autoshaping in pigeons. Animal Learning & Behavior, 21, 360-368.

BROWN, B. L., HEMMES, N. S., COLEMAN, D. A., Jr., HASSIN, A., & GOLDHAMMER, E. (1982). Specification of the stimulus-reinforcer relation in multiple schedules: Delay and probability of reinforcement. Animal Learning & Behavior, 10, 365-376.

BUZSAKI, G. (1982). The "Where is it?" reflex: Autoshaping the orienting response. Journal of the Experimental Analysis of Behavior, 37, 461-484.

FANTINO, E. (1977). Conditioned reinforcement: Choice and information. In W. K. Honig & J. E. R. Staddon (Eds.), Handbook of operant behavior (pp. 313-339). Englewood Cliffs, NJ: Prentice-Hall.

GIBBON, J., BERRYMAN, R., & THOMPSON, R. L. (1974). Contingency spaces and measures in classical conditioning. Journal of the Experimental Analysis of Behavior, 21, 585-605.

HEARST, E. (1987). Extinction reveals stimulus control: Latent learning of feature-negative discriminations in pigeons. Journal of Experimental Psychology: Animal Behavior Processes, 13, 52-64.

HEARST, E, (1989). Backward associations: Differential learning about stimuli that follow the presence versus the absence of food in pigeons. Animal Learning & Behavior, 17, 280-290.

HEMMES, N. S., BROWN, B. L., & CABEZA DE VACA, S. (1990). Effects of trial duration on overall and momentary rates of maintained autoshaped keypecking: Choice and single stimulus trials. Animal Learning & Behavior, 18, 171-178.

HOLLAND, P. C. (1977). Conditioned stimulus as a determinant of the form of the Pavlovian conditioned response. Journal of Experimental Psychology: Animal Behavior Processes, 3, 77-104.

JENKINS, H. M., BARNES, R. A., & BARRERA, F. J. (1981). Why autoshaping depends upon trial spacing. In C. M. Locurto, H. S. Terrace, & J. Gibbon (Eds.), Autoshaping and conditioning theory (pp. 255-284). New York: Academic Press.

KAPLAN, P. S. (1984). Importance of relative temporal parameters in trace autoshaping: From excitation to inhibition. Journal of Experimental Psychology: Animal Behavior Processes, 10, 113-126.

KONORSKI, J. (1967). Integrative activity of the brain. Chicago: University of Chicago Press.

LINDBLOM, L. L., & JENKINS, H. M. (1981). Responses eliminated by noncontingent or negatively contingent reinforcement recover in extinction. Journal of Experimental Psychology: Animal Behavior Processes, 7, 175-190.

LOLORDO, V. M., & FAIRLESS, J. L. (1985). Pavlovian conditioned inhibition: The literature since 1969. In R. R. Miller & N. E. Spear (Eds.), Information processing in animals: Conditioned inhibition (pp. 1-49). Hillsdale, NJ: Lawrence Erlbaum.

MATZEL, L. D., HELD, F. P., & MILLER, R. R. (1988). Information and expression of simultaneous and backward associations: Implications for contiguity theory. Learning and Motivation, 19, 317-344.

OVERMIER, J. B., BULL, J. A., III, & TRAPOLD, M. A. (1971). Discriminative cue properties of different fears and their role in response selection in dogs. Journal of Comparative and Physiological Psychology, 76, 478-482.

RESCORLA, R. A. (1978). Some implications of a cognitive perspective on Pavlovian conditioning. In S. H. Hulse, H. Fowler, & W. K. Honig (Eds.), Cognitive processes in animal behavior (pp. 15-50). Hillsdale, NJ: Lawrence Erlbaum.

RESCORLA, R. A. (1989). Redundant treatments of neutral and excitatory stimuli in autoshaping. Journal of Experimental Psychology: Animal Behavior Processes, 15, 212-223.

SPETCH, M. L., WILKIE, D. M., & PINEL, J. P. J. (1981). Backward conditioning: A reevaluation of the empirical evidence. Psychological Bulletin, 89, 163-175.

TAIT, R. W., & SALADIN, M. E. (1986). Concurrent development of excitatory and inhibitory associations during backward conditioning. Animal Learning & Behavior, 14, 133-137.

WASSERMAN, E. A., FRANKLIN, S. R., & HEARST, E. (1974). Pavlovian appetitive contingencies and approach versus withdrawal to conditioned stimuli in pigeons. Journal of Comparative and Physiological Psychology, 86, 616-627.

ZENER, K. (1937). The significance of behavior accompanying conditioned salivary secretion for theories of the conditioned response. American Journal of Psychology, 50, 384-403.

ZENTALL, T. R., SHERBURNE, L. M., & STEIRN, J. N. (1992). Development of excitatory backward associations during the establishment of forward associations in a delayed conditional discrimination by pigeons. Animal Learning & Behavior, 20, 199-206.
COPYRIGHT 1994 The Psychological Record
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1994 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Hemmes, Nancy S.; Brown, Bruce L.; De Vaca, Soledad Cabeza
Publication:The Psychological Record
Date:Mar 22, 1994
Previous Article:Coping with cold-pressor pain: effects of mood and covert imaginal modeling.
Next Article:Multidimensional evaluation of monetary incentive strategies for weight control.

Related Articles
Emergent relations in the formation of stimulus classes by pigeons.
Value transfer in discriminative conditioning with pigeons.
The effect of water deprivation on shock-escape impairment after exposure to inescapable shock.
Variable-interval schedule control following response acquisition with delayed reinforcement.
Choice and the initial delay to a reinforcer.
Bring them home; Champion-bred birds are stolen before competition.

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters