Transitive inference in pigeons: Simplified procedures and a test of value transfer theory

Minimal procedures for the demonstration of transitive inference (TI) in animals have involved the training of four simultaneous discriminations: for example, A+B?, B+C?, C+D?, and D+E?, followed by the demonstration of a preference for B over D on test trials. In Experiment 1, we found that TI in pigeons can be found with successive training involving A+B?, B+C?, A+C?, C+D?, D+E?, C+E?, and A+E?. In Experiment 2, we found that demonstration of TI did not require inclusion of experience with the nonadjacent stimulus pairs (A+C?, C+E?, A+E?). Experiment 3 provided a test of value transfer theory (VTT; Fersen, Wynne, Delius, & Staddon, 1991). When pigeons were trained with stimulus pairs that did not permit the transitive ordering of stimuli, but did permit the differential transfer of value (e.g., A+B?, C?E+, C+D?, & A+E?), preference for B over D was still found. Analyses of the relation between direct experiences with reinforced and nonreinforced responding and stimulus preferences on test trials failed to support a reinforcement-history account of TI.     

Plus/minus grading and motivation: an empirical study of student choice and performance

This paper considers whether student motivation might be impacted by the replacement of a straight (A, B, C, D, F) grading system with a plus/minus system (A, A?, B+ … D?, F). The data that are examined are from several undergraduate economics classes at a mid‐sized midwestern university in the United States. The data includes student characteristics, student performance and students’ choices of either a plus/minus or a straight grading system. In this admittedly small‐scale study students, who chose plus/minus grading, were not significantly more motivated than students who opted for straight grading.     

Representations of single and compound stimuli in negative and positive patterning

In four experiments, rats were trained on different patterning discriminations before being tested with compounds composed of novel combinations of the trained stimuli. In Experiment 1, rats were trained on a negative-patterning schedule (A+ B+ AB-) intermixed with reinforced presentations of a second compound (CD+). On a subsequent test, the rats responded more to two novel compounds, AC and BD, than to A and B, but less than to CD. In Experiment 2, rats were trained on two concurrent negative-patterning discriminations (A+ B+ AB-, C+ D+ CD-). On test, they responded more to AC and BD than to AB and CD, but less than to the single stimuli. In Experiment 3, rats were trained on two concurrent positive-patterning discriminations (A-B- AB+, C- D- CD+). On test, their response rates to AC and BD were not different from the response rates to the trained compounds (AB and CD). Finally, in Experiment 4, rats were trained on a positive- and negative-patterning discrimination concurrently. Once again, on test, response rates to AC and BD were not different from responding on reinforced trials of the trained discriminations (A+, B+, and CD+). We discuss the implications of these findings for elemental and configural models of stimulus representation.     

An elemental model of retrospective revaluation without within-compound associations

When retrospective revaluation phenomena (e.g., unovershadowing: AB+, then A?, then test B) were discovered, simple elemental models were at a disadvantage because they could not explain such phenomena. Extensions of these models and novel models appealed to within-compound associations to accommodate these new data. Here, we present an elemental, neural network model of conditioning that explains retrospective revaluation apart from within-compound associations. In the model, previously paired stimuli (say, A and B, after AB+) come to activate similar ensembles of neurons, so that revaluation of one stimulus (A?) has the opposite effect on the other stimulus (B) through changes (decreases) in the strength of the inhibitory connections between neurons activated by B. The ventral striatum is discussed as a possible home for the structure and function of the present model.     

Acquisition and transfer of control by an ambiguous cue

In three experiments, the specificity of action of occasion setters was examined, using a discretetrial leverpress procedure. Rats’ acquisition of anXA+, A?, XB?, B+ discrimination, in which a single feature cue (X) signaled the reinforcement of responding in the presence of one target cue (A) and the nonreinforcement of responding during another target (B), was no more difficult than acquisition ofXA+, A?, X?, B+ orX+, A?, XB?, B+ discriminations, in whichX signaled only one target-response-reinforcement contingency. In transfer tests,X did not modulate responding controlled by other cues that were untrained or consistently reinforced, either when the elements of the compounds were presented serially or when they were presented simultaneously in training and testing. However, after serial, but not after simultaneous, compound training ,X facilitated responding controlled by a cue that was trained and then extinguished. Implications for the nature of occasion setting and configurai learning are discussed.     

Summation: Assessment of a configural theory

In five experiments, rats were given Pavlovian pairings of auditory and visual stimuli with delivery of food pellets. Experiment 1 found greater responding to an AB compound after training with the individual A and B stimuli, compared with responding both to the A and B elements and to a separately trained CD compound. Experiment 2 found this enhanced responding to depend on the associative strengths of A and B. In Experiment 3, responding was greater to a CD compound than to the other compounds after an AB-, AD+, BC+ training procedure. In Experiment 4, responding to an AB compound was greater than that to the elements after A was reinforced on a 100% schedule and B on a 50% schedule. In Experiment 5, responding to an AC compound was greater than that to either A or C after an AB+, CD+, A-training procedure. A configural theory, such as that proposed by Pearce (1987), anticipates summation in none of these procedures, unless the conditioned context is assumed to have a salience greater than zero. In order to predict summation in Experiments 3, 4, and 5, a context salience greater than that of the elements must be assumed. However, such an assumption also anticipates that extinction of a 100% stimulus should eliminate responding to a 50% stimulus. The results of Experiment 3 contradicted that prediction. These results conform better to the expectations of elemental models of conditioning.     

Value transfer in a simultaneous discrimination by pigeons: The value of the S+ is not specific to the simultaneous discrimination context

During simultaneous discrimination training, there is evidence that some of the value of the S⁺ transfers to the S^?. When the value of the S+ is altered outside the context of the simultaneous discrimination, two very different predictions are made concerning its effect on its S^?, depending on whether one views the S⁺ as an occasion setter or as a stimulus capable of transferring value. In four experiments, pigeons were trained with two similar simultaneous discriminations, A⁺B^? and C⁺D^?, and two single-stimulus trial types, A and C, (in which A always had greater nominal value than C). According to value transfer theory, on test trials, B should always be preferred over D, because B and D should be affected by the net values of A and C, respectively. According to an occasion setting account, however, D should be preferred over B because the presence of D signals a higher probability of reinforcement for responding to C than when C is alone, and/or the presence of B signals a lower probability of reinforcement for responding to A than when A is alone. In all four experiments, the pigeons preferred B over D, a result consistent with value transfer theory. Thus, an S^? can acquire value from an S⁺ even when that value is conditioned in a “context” different from that of the simultaneous discrimination.     

Transfer testing after serial feature-ambiguous discrimination in Pavlovian keypeck conditioning

Four experiments were performed to explore modulatory transfer after serial feature-ambiguous (F-A) discrimination (X→A+, A?, X→B?, B+) in Pavlovian keypeck conditioning (autoshaping) with pigeons. Diffuse features were used in the first three experiments, and no modulatory transfer was found: (1) There was no modulatory transfer between two concurrently trained F-A tasks; (2) modulatory transfer to conditioned and then extinguished stimuli was not observed; and (3) responding to the targets (A and B) after their contingencies were reversed was not affected by presentation of their original feature stimulus (X). Transfer between two serial F-A tasks was obtained in the last experiment, in which keylights were used as features, but other evidence indicated that this was due to stimulus generalization between the features. Together, the results strongly suggest that specific target-food occasion setting or configural learning is the underlying mechanism of F-A discrimination.     

Summation and overexpectation with qualitatively different outcomes

In five experiments, a Pavlovian appetitive conditioning preparation was used with rats to explore the interaction of associations with different, but equivalently valued, outcomes. Experiment 1 demonstrated summation in magazine responding when two stimuli previously associated with different outcomes were combined. Experiments 2A and 2B found that pairing that stimulus compound with one of the outcomes led to a decrease in performance (overexpectation) for both of the stimulus elements. Experiments 3A and 3B confirmed that result but demonstrated the continued presence of the original stimulus-outcome associations. The results are consonant with a view in which replacing one outcome with another leads to an associative structure containing both stimulus-outcome associations and an outcome-independent depressive process.     

Linking the components of a university program to the qualification profile of graduates: The case of a sustainability‐oriented environmental science curriculum

A university Environmental Sciences curriculum is described against the background of requirements for environmental problem solving for sustainability and then analyzed using data from regular surveys of graduates (N = 373). Three types of multiple regression models examine links between qualifications and curriculum components in order to derive some conclusions about which qualifications are enhanced by which curriculum components. The underlying rationale of these models is that assessments of the importance of a certain component (and the time graduates think should be allocated to it) should increase with the workplace demand for the qualifications it provides. A comprehensive set of 19 qualifications was used, subdivided into three areas of environmental problem‐solving skills (A, basic scientific and technological skills; B, transformation‐oriented skills; and C, sociopolitical skills) and two areas of transferable skills (D, individual key skills and E, social and communication skills). Relationships identified by different regression models are discussed in terms of mutual consistency and with regard to the design, content, and learning goals of the curriculum components. Many plausible relationships were identified. Using such regression models is a promising indirect method that may be generally applicable for the explorative qualification‐oriented evaluation of university curricula and their fundamental components based on graduates' judgments. © 2009 Wiley Periodicals, Inc. J Res Sci Teach 46: 537–569, 2009     

Summation and configuration in patterning schedules with the rat and rabbit

Discrimination between a tone + light compound and its components in positive and negative patterning schedules was examined. In the positive schedule, reinforced compound presentations (C+) were intermixed with unreinforced component presentations (T?, L?). In the negative schedule, the compound was unreinforced (C?) and the components were reinforced (T+, L+). In Experiment 1, appetitive conditioning of rats’ anticipatory magazine responses was used, and in Experiment 2, aversive conditioning of the rabbit’s nictitating membrane response was used. Both experiments revealed that the positive patterning schedule consistently produced rapid acquisition of appropriate discriminative responding. The results of the negative patterning schedule were more complex. Specifically, the results of Experiment 1 demonstrated that naive rats initially showed rapid acquisition of the negative patterning discrimination. However, schedule reversals revealed that experience with the positive patterning schedule virtually abolished subsequent acquisition of discriminative responding under the negative patterning schedule. The results of Experiment 2 revealed that naive rabbits showed very slow acquisition of discriminative responding under the negative patterning schedule. The results are discussed in relation to the unique-stimulus hypothesis, a contextual encoding hypothesis, and a configural hypothesis.     

Summation in autoshaping

Pigeon subjects received Pavlovian conditioning with stimulus elements and were then tested with compounds of those elements. Experiments 1–3 used localized keylight elements and found no evidence for greater responding to the compound than to the elements. Experiments 4A–4D found evidence for greater second-order conditioning by a compound of two elements than by the elements themselves when the elements consisted of two diffuse stimuli or one diffuse stimulus and one localized keylight. No greater second-order conditioning resulted from a compound of two localized keylight elements, suggesting the possibility of perceptual interactions that reduce identification of the elements in the compound. Experiment 6 found evidence of summation when that interaction was reduced by sequential presentation. However, one attempt to capture this interaction in terms of configural conditioning (Pearce, 1987) failed to receive confirmation. These results suggest that the localized stimuli conventionally employed in autoshaping experiments may show such substantial perceptual interaction as to recommend against their routine use for studying conditioning in compounds.     

Compound-component and conditional discrimination of colors and odors by honeybees: Further tests of a continuity model

In experiments previously reported, individual honeybees were trained in a variety of problems to discriminate color-odor compounds. The results could be modeled accurately on the assumption that the associative strength of each component of a compound stimulus changes independently with reinforcement or nonreinforcement of the compound (independence rule) and that the associative strength of a compound is equal to the sum of the strengths of its components (summation rule). In the present experiments, which were designed to challenge the model, honeybees were trained in compound-component problems (choosing between compounds and their separate components) and in conditional problems (choosing between colors on the basis of a common odor or between odors on the basis of a common color). The new data, together with all of the previous data, could be modeled accurately on the further assumption that interaction among the components of a compound generates a new, compound-unique component that gains and loses associative strength in the same way as other components and contributes in the same way to the strength of the compound; the independence and summation rules continue to apply.     

A surprising extension of the preconditioned stimulus beyond the cotermination of the US and the added element does not alleviate blocking to the added element

Rats were subjected to the Kamin two-stage blocking procedure. First, a stimulus, A, was conditioned and then reinforced in compound with a target stimulus, B. During compound training, an attempt was made to alleviate blocking of conditioning to B by presenting a “surprise” stimulus for 5 sec following the reinforced AB compound. The surprise stimulus consisted of the continued presentation of the previously reinforced element, A. During A training, A had never extended beyond the moment of reinforcement; thus, its extension beyond reinforcement during AB training was hypothesized to be “surprising” and, therefore, expected to alleviate blocking. Blocking in this condition was, however, just as strong as in a standard blocking condition. The results do not favor a surprise interpretation of unblocking but do seem to be consistent with other theoretical views (e.g., Rescorla & Wagner, 1972).     

Response strategies after overtraining in the jumping stand

Rats were trained to criterion (CT) and overtrained (OT) on a horizontal against vertical stripes discrimination in the jumping stand. A change in response strategy, from inspecting both stimuli to inspecting just the S?, was observed during overtraining. Replacing the S? with a novel stimulus disrupted performance in the OT Ss, but replacing the S+ did not. Performance in the CT Ss was disrupted, but not very severely, both when the S+ was replaced and when the S? was replaced. These results suggest that OT Ss in the jumping stand come to rely especially upon the S?. The overtraining effects found here are compared with those found in other types of apparatus.     

Contextual modulation of attention in human category learning

In a category-learning experiment, we assessed whether participants were able to selectively attend to different components of a compound stimulus in two distinct contexts. The participants were presented with stimulus compounds for which they had to learn categorical labels. Each compound comprised one feature from each of two dimensions, and on different trials the compound was presented in two contexts, X and Y. In Context X, Dimension A was relevant to the solution of the categorization task and Dimension B was irrelevant, whereas in Context Y, Dimension A was irrelevant and Dimension B was relevant. The results of transfer tests to novel stimuli suggested that people learned to attend selectively to Dimension A in Context X and Dimension B in Context Y. These findings contribute to the growing body of evidence that people can learn to selectively attend to particular dimensions of stimuli dependent on the context in which the stimuli are presented. Furthermore, the findings demonstrate that context-dependent changes in attention transfer to other categorization tasks involving novel stimuli.     

Value transfer in concurrent-schedule discriminations by pigeons

When pigeons are trained on a discrete-trial simultaneous discrimination, some of the value associated with the positive stimulus appears to transfer to the negative stimulus (Zentall & Sherburne, 1994). Pigeons preferred a negative stimulus that had been discriminated from an always-positive stimulus (S+) over a negative stimulus that had been discriminated from a sometimes-positive stimulus (S±). A very different finding (suggestive of transitivity of preference or contrast) was reported by Belke (1992). On concurrent probe tests of stimuli associated with equal variable interval (VI) schedules but originally trained in alternative concurrent pairs (one with a richer schedule, the other with a poorer schedule—VI 20 sec vs. VI 40 sec and VI 40 sec vs. VI 80 sec), the stimulus originally paired with the poorer schedule was preferred. But Belke’s results may have been obtained because the pigeons had been trained to peck the VI 40 sec paired with the poorer schedule and they had been trained not to peck the VI 40 sec paired with the richer schedule. In the present experiment, we avoided this bias by training pigeons on two concurrent schedules in which the tested stimuli both had been associated with the poorer schedule of the pair [A(VI 20 sec) vs. B(VI 80 sec) and C(VI 40 sec) vs. D(VI 80 sec)]. Evidence for value transfer was demonstrated when on probe trials pigeons preferred B over D.     

Blocking with serial and simultaneous compounds in a trace conditioning procedure

Blocking of conditioned suppression in rats was studied in three experiments using serial and simultaneous compounds in Pavlovian trace conditioning procedures. Experimental groups were first given trace conditioning trials with a 2-sec stimulus (A) presented at least 60 sec before an electric grid shock US. Next, both experimental and control groups received reinforced trials with a compound stimulus (AB). Both A and B were 2 sec in duration and were presented at least 60 sec before the US. For some groups during AB training, the A stimulus preceded the B stimulus; for others, B preceded A; for still others, A and B occurred simultaneously. Conditioning was subsequently assessed separately to both A and B. The results were as follows: First, varying the interval between the onset of A and the US during A training appeared to produce significantly different levels of conditioning to A but did not detectably affect A’s ability to block conditioning to B. Second, blocking was observed in both simultaneous and serial procedures. Third, in the serial procedure, A blocked conditioning to B whether it preceded B or followed B in the AB compound. Fourth, in tests given after AB conditioning, the experimental and control groups suppressed similarly to A. The relevance of these results to the conditioning model of Rescorla and Wagner (1972) and to Mackintosh’s (1975b) theory of attention are discussed.     

Benzodiazepine administration prevents the use of error-correction mechanisms during fear extinction

Three experiments examined the effect of systemic administration of the benzodiazepine midazolam on extinction and re-extinction of conditioned fear. Experiment 1 demonstrated that midazolam administration prior to extinction of a conditioned stimulus (CS) impaired that extinction when rats were subsequently tested drug free; however, extinction was spared if rats were extinguished, reconditioned, and re-extinguished under midazolam. Experiment 2 provided a replication of this effect within-subjects; rats were conditioned to two CSs (A and B), extinguished to one (A-), reconditioned to both, and then extinguished/re-extinguished to both stimuli in compound (AB-), under either vehicle or midazolam. On the drug-free test, rats given midazolam froze more to the CS that had been extinguished (B) than the one that been re-extinguished (A). The final experiment examined whether extinction under midazolam was regulated by prediction error. Rats were trained with three CSs (A, B, C) and extinguished to two (A-, C-). These stimuli then underwent additional extinction under midazolam or vehicle, with one CS now presented in compound with the non-extinguished CS (AB-, C-). Rats were then tested for fear of A relative to C. Rats given vehicle showed a deepening of extinction to A relative to C, as is predicted from error-correction models; however, rats given midazolam failed to show any such discrepancy in responding. The results are interpreted to indicate that the drug reduced prediction error during extinction by reducing fear, and rats were able to re-extinguish fear via a retrieval mechanism that is independent of prediction error.     

Stimulus competition in the absence of compound conditioning

Most associative theories have assumed that stimulus competition occurs only between conditioned stimuli (CSs) that are trained in compound. The present research investigated the possibility of competition between two CSs that were individually paired to the same unconditioned stimulus (US). We used human subjects in an anticipatory suppression analogue to Pavlovian conditioning. Experiment 1 showed that X+ training followed by A+ training resulted in impaired responding to X. This did not occur when A+ training preceded X+ training. Experiment 2 replicated the basic effect and showed that it did not occur when the Phase 2 training consisted of A? instead of A+ nor when the A+ pairings occurred in a second context. Experiment 3 showed that A+ pairings occurring in a second context could still produce the effect when X was tested in the context in which the A+ pairings had occurred, but not when X was tested in a context different from that used for A+ training. Collectively, these results suggest that individually trained CSs may compete with each other when one of those CSs is more strongly activated by the test context than the other one.