Category difference facilitates oddity learning in honeybees (Apis mellifera).

Performance of honeybees resembles that of vertebrates in a variety of associative learning experiments. Recent work has focused on relational learning phenomena not easily explained by associative principles, including same/different problems, the simplest of which is the oddity problem. Free-flying bees were trained to visit a laboratory window and were rewarded for choice of the odd stimulus among a set of stimuli. There were two stimulus categories, single-color solids and two-color patterns. The training was trial-unique, with new sets of stimuli on each trial. In Experiment 1, 4 groups were trained in a 3-stimulus oddity problem, 2 with solid odd and patterns nonodd and 2 with pattern odd and solids nonodd. For 1 group in each condition, the odd and nonodd stimuli shared a color. The performance of all groups was better than chance. The bees could solve the problem on the basis of oddity (same vs. different) or category (solid vs. pattern). These possibilities were unconfounded in Experiment 2 with 2 groups trained in a 4-stimulus oddity problem. Group 1 was trained with a category difference on each trial; the solid color was odd on half the trials and the pattern odd on the others. Group 2 was trained with no category difference; all stimuli were patterns. Both groups showed better-than-chance performance, and the irrelevant category difference facilitated oddity discrimination for Group 1. The results support previous findings of oddity learning in honeybees, the only invertebrate species for which any relational learning phenomena have been demonstrated. (PsycInfo Database Record (c) 2020 APA, all rights reserved).

Relational learning in honeybees (Apis mellifera): Oddity and nonoddity discrimination.

Honeybee learning is surprisingly similar to vertebrate learning and one implication is that the basic associative learning principles are also similar. This research extends the work to more complex cognitive phenomena. Forager bees were trained individually to visit a laboratory window for sucrose. On each training trial for all experiments, bees found three stimuli, two identical and one different. In Experiments 1 and 2, stimuli were three-dimensional two-color patterns, and in Experiments 3 and 4, stimuli were two-color patterns displayed on a computer monitor. Training was trial-unique, that is, a different triad of stimuli was presented on each trial. In Experiments 1 and 3, choice of odd was rewarded and choice of nonodd was punished. In Experiments 2 and 4, choice of nonodd was rewarded and choice of odd was punished. On every trial, the initial choice was recorded and correction permitted. Honeybees learned to choose the odd stimulus in Experiments 1 and 3 and the nonodd stimuli in Experiments 2 and 4. The results provide compelling evidence of oddity and nonoddity learning, often interpreted as relational learning in vertebrates. Whether the mechanism of such learning in honeybees is similar to that of vertebrate species remains to be determined.

Within-subjects experiments on blocking and facilitation in honeybees (Apis mellifera).

It has long been suspected in the vertebrate literature, but demonstrated only recently in work with honeybees (Apis mellifera), that the different treatments of nontarget stimuli in conventional between-groups blocking experiments may give the appearance of blocking independently of experience with the target stimulus. The same difficulty does not arise in within-subjects experiments, and in a series of such experiments with odors and colors free-flying honeybees gave no evidence of blocking; separate reinforced presentations of one element of a reinforced compound failed to reduce responding to the second. There was, however, clear evidence of facilitation; separate nonreinforced presentations of one element of a reinforced compound increased responding to the second. The implications of the results for further work on compound conditioning in honeybees and other animals are considered.

Delayed symbolic matching in honeybees (Apis mellifera).

In a recent experiment on short-term memory (P. A. Couvillon, T. P. Ferreira, & M. E. Bitterman, 2003), honeybees (Apis mellifera) learned to choose between 2 colors on the basis of immediately preceding experience with 1 of them. Some learned to choose the same color as the sample (perseveration or matching), others to choose the alternative color (alternation or nonmatching). Performance in the 2 problems was very much the same. In the present experiment, honeybees learned no less readily to choose between the 2 colors on the basis of sample stimuli that were different from the colors (symbolic matching). A simple associative interpretation of the results is proposed.

Blocking and pseudoblocking: new control experiments with honeybees.

Prompted by doubts about the adequacy of the various control procedures long used in research on blocking, we repeated some earlier experiments with honeybees that had given the appearance of forward, concurrent, and backward blocking. The new experiments differed from the earlier experiments only in that the target stimulus was omitted during the training and was encountered for the first time in the test. In the new experiments, just as in the earlier experiments, the blocking groups responded less to the target stimulus than did the control groups. The results show that the effects of the different treatments of nontarget stimuli commonly compared in blocking experiments may generalize to the target stimulus and thus affect responding to that stimulus independently of experience with it. Implications for research on blocking in honeybees and other animals are considered.

The role of context in the inhibitory conditioning of honeybees.

Conditioned inhibition or CI training (A+/AB-) was compared with S- training (A+/B-) in three experiments on proboscis-extension conditioning in harnessed honeybees. The purpose was to test the Rescorla-Wagner assumption, widely credited in the vertebrate literature, that a nonreinforced stimulus acquires inhibitory properties in proportion to the excitatory value of the context in which it is presented. In prior work with free-flying honeybees pretrained with sucrose to come of their own accord to the experimental situation, no differences were found in the consequences of CI and S- training, perhaps because A added little to the excitatory value of the context (already very high) in which B occurred. In the new experiments, with harnessed subjects brought involuntarily into the training situation, negative results again were obtained. The possibility is considered that inhibitory conditioning in honeybees is independent of the excitatory value of the context.

Temporal control of conditioned responding in goldfish.

The peak procedure was used to characterize response timing during acquisition and maintenance of conditioned responding in goldfish. Subjects received light-shock pairings with a 5- or 15-s interstimulus interval. On interspersed peak trials, the conditioned stimulus light was presented for 45 s and no shock was delivered. Peaks in the conditioned response, general activity, occurred at about the time of the expected unconditioned stimulus, and variability in the activity distribution was scalar. Modeling of the changes in the activity distributions over sessions revealed that the temporal features of the conditioned response changed very little during acquisition. The data suggest that times are learned early in training, and, contrary to I. P. Pavlov's (1927/1960) concept of "inhibition of delay," that timing is learning when to respond rather than learning when not to respond.

Backward blocking in honeybees.

Three experiments with foraging honeybees were designed to study the effect of experience with A on responding to B after AB+ training. In the first experiment, responding to B was the same whether the AB+ training was preceded or followed by A+ training. In the second experiment, responding to B after AB+ training was less in animals that also had A+ training than in control animals that were equally often reinforced in the absence of A; whether the A+ training preceded, was concurrent with, or followed the AB+ training made no difference. In the third experiment, responding to B after AB+ training was less when the AB+ training was followed by A+C- training than when it was followed by C+/A- training. These results, like those of some recent vertebrate experiments, take us beyond the traditional explanation of blocking in terms of impaired conditioning of B on AB+ trials and support the suggestion that the mechanism, still poorly understood, may nevertheless be a relatively simple one.

Inhibitory conditioning in honeybees.

Honeybees were rewarded with sucrose solution for choosing AX(a grey target, X, labelled with a distinctive stimulus, A) rather than ABX (a grey target labelled both with A and with another distinctive stimulus, B)-AX+/ABX- training. Tests of independent groups made after such training showed a clear preference not only for AX over ABX, but also for ABX over BX, and for X over BX. These experiments, along with some earlier ones to which they bring a new perspective, provide persuasive evidence, previously lacking, of inhibitory conditioning in honeybees.

Delayed alternation in honeybees (Apis mellifera).

Previous experiments with honeybees (Apis mellifera) failed to show learned control of performance by short-ten memory. In this study, honeybees were trained with an improved technique to choose 1 of 2 colors that was either the same as a recently rewarded sample (perseveration) or different (alteration). Because any increase in associative strength stemming from the sample experience would tend to promote perseveration and contravene alternation, the equal difficulty of the 2 tasks suggests that the role played by the sample was primarily discriminative. The animals remembered on each trial the immediately preceding experience with reward and learned to use that information appropriately. These new results extend the list of what may well be fundamental similarities in the learning of vertebrates and honeybees.

The role of nonreinforcement in the learning of honeybees.

Two series of experiments with honeybees were designed to test the assumption that inhibition is generated by nonreinforcement as a function of the excitatory value of the context. In the first series (Experiments 1-3), summation tests with B were made after A+/C-/AB- as compared to A+/C-/CB- training, with precautions taken to minimize the possibility of a masking effect of excitatory within-compound conditioning on AB trials; responding to B did not vary with training procedure. In the second series (Experiments 4-5), retardation tests rather than summation tests were used, in the belief that they might be more sensitive; after A+/AB-/CD- training, acquisition in a B+/D- problem was found to be no less rapid than in a D+/B- problem. A third series of experiments (Experiments 6-9) was designed to test the more general assumption that the effectiveness of nonreinforcement increases with the excitatory value of the context; response to B was found to be no different after A+/B+/C- training followed by A+/AB- training than after A+/B+/C- training followed by A+/CB- training. The results are compatible with the view that the role of nonreinforcement in honeybees is not to generate inhibition, but only to reduce excitation in a manner independent of the excitatory value of the context.

Quantitative tests of an associative theory of risk-sensitivity in honeybees.

Risk-sensitivity was studied in free-flying honeybees trained individually to choose between two scented targets (A and B) with varying amounts and concentrations of sucrose solution as reward. In the first phase of experiment 1, the animals showed "risk-aversion," preferring A, which provided 5 microl of a 40 % sucrose solution on every trial, to B, which provided 30 microl of the same solution once in every six trials (mean amount per trial 5 microl for each alternative). In the second phase, the preference reversed with reversal of the reward assignments. In experiment 2, the consistently rewarded A (5 microl of 40 % sucrose solution per trial) was again preferred, although the inconsistently rewarded B now provided twice the amount of sucrose solution on average (30 microl on two of every six trials, mean amount per trial 10 microl). In experiment 3, with A providing 10 microl of a 15 % sucrose solution on every trial and B providing 10 microl of a 60 % sucrose solution on two of every four trials (mean concentration per trial 30 %), the animals preferred B. In Experiment 4, patterned after experiment 1, similar results were obtained under more natural conditions in which the animals were no longer constrained (as they were in the first three experiments) to go equally often to each alternative. The results of all four experiments were predicted quantitatively and with considerable accuracy by a simple associative theory of discriminative learning in honeybees.

Intermodal blocking in honeybees.

Previous findings of intramodal but not of intermodal blocking in foraging honeybees prompted a new series of experiments with colours, odours, a proximal visual landmark, and a localized geomagnetic anomaly as stimuli. In Experiments 1-2, the landmark was blocked by both colour and odour. In Experiments 3-6, the anomaly was blocked by both colour and odour, but the anomaly failed to block either colour or odour. In Experiments 7-8, the anomaly failed again to block either colour or odour even though it could be shown to develop substantial associative strength in the course of the training. The several instances of intermodal blocking bring the results for honeybees into closer agreement than before with the results for vertebrates. The failures of blocking seem understandable in terms of the relative salience of the stimuli employed without reference to modal relationships. An attentional interpretation is suggested.

Exploratory studies of inhibitory conditioning in honeybees (Apis mellifera).

The best available evidence of inhibitory conditioning in vertebrates comes from experiments in which variants of A+/AB- and A+/B- training were compared in terms of response to B in summation and retardation tests, the results suggesting that inhibition is generated by nonreinforcement as an increasing function of the excitatory value of the setting. We report here 7 experiments with foraging honeybees (Apis mellifera) that failed to show a difference in the effects of the 2 treatments. On the basis of previous experiments as well as supplementary experiments whose results give no reason to doubt the sensitivity of the training techniques and measures used, our consistently negative results may mean either that inhibition in honeybees is generated by nonreinforcement independently of the setting or that there is no inhibitory conditioning at all in honeybees--that the only associative function of nonreinforcement is to reduce excitatory strength.

Quantitative two-process analysis of avoidance conditioning in goldfish.

The shuttlebox performance of goldfish was studied under standardized conditions in a variety of problems--with or without an avoidance contingency, a conditioned stimulus (CS)-termination contingency, and an escape contingency. The effects of CS-only, unconditioned stimulus (US)-only, and explicitly unpaired training were also examined. All the data could be simulated quantitatively with a version of O. H. Mowrer's (1947) 2-process theory expressed in 2 learning equations (1 classical, the other instrumental) and a performance equation. The good fit suggests that the theory is worth developing further with new experiments designed to challenge it.

Learning by honeybees (Apis mellifera) on arrival at and departure from a feeding place.

The question of when in the course of a visit to a feeding place foraging honeybees (Apis mellifera) learn about its location was studied in Experiment 1 by moving the animals a short distance after they arrived and began to feed. A preference for the arrival place developed, although less rapidly than in control animals for which the arrival and departure places were the same. In Experiments 2-5, a distinctive object was used to define the location of the feeding place. When the object was removed after arrival or introduced only after arrival, the animals learned less about its color and shape than did control animals for which it was present throughout each visit. The results contradict the claim that honeybees learn about certain characteristics of a feeding place only on arrival and about others only on departure.

Performance of goldfish (Carassius auratus) in patterned sequences of rewarded and nonrewarded trials.

The performance of goldfish was studied in a series of experiments with patterned sequences of reward (R) and nonreward (N) for response to a given stimulus (S3). Trials with two other, readily discriminable stimuli, response to one always rewarded and response to the second never rewarded, were interpolated in such a way that the outcome of any S3 trial could not be predicted from the events of the immediately preceding trial. The main purpose was to look for control of response to S3 based on memory of N like that found in previous work with pigeons. In Experiments 1 and 2, blocks of R and N trials with S3 were programmed, R trials in the first half of each session and N in the last half, or N trials in the first half and R in the last half. In Experiment 3, successive acquisition and extinction of response to S3 was studied, with the N-to-R and R-to-N transitions either within sessions or between sessions. In Experiment 4, the effects of partial and consistent reinforcement on extinction of response to S3 were compared. The results obtained are similar in important respects to those for pigeons but different in their failure to demonstrate control of performance by memory of N.

The overlearning-extinction effect and successive negative contrast in honeybees (Apis mellifera).

In three experiments with free-flying honeybees, the previously discovered overlearning-extinction effect was replicated under different conditions and shown to depend on magnitude of reinforcement: The effect appeared in training with a 50% sucrose solution but not with a 20% solution. The results prompted a fourth experiment in which successive negative contrast was demonstrated: The animals were disturbed to find the 20% solution on a distinctive target that always before had been baited with the 50% solution. The conclusion is that the overlearning-extinction effect is an instance of contrast and can be understood in terms of frustration engendered by unrealized anticipation of reinforcement.