Evidence for associative learning in newly emerged honey bees (Apis mellifera)

The honey bee is a model organism for studies on the neural substrates of learning and memory. Associative olfactory learning using sucrose rewards is fast and reliable in foragers and older hive bees. However, researchers have so far failed to show any significant learning in newly emerged bees. It is generally argued that in these bees only part of the brain structures important for learning are fully developed. Here we show for the first time that newly emerged honey bees are capable of associative learning, if they are sufficiently responsive to sucrose. Responsiveness to sucrose, which can be measured using the proboscis extension response (PER), increases with age. Newly emerged bees are on average very unresponsive to sucrose. We show that if newly emerged bees displaying a PER to 10% sucrose or lower sucrose concentrations are conditioned to an odour, they show significant associative learning and early long-term memory. Nevertheless, the level of acquisition is still lower than in foragers. The general assumption that newly emerged honey bees are incapable of associative learning must therefore be reconsidered. Further, our study suggests that an age-dependent increase in responsiveness to rewarding stimuli is directly related to the development of early learning abilities. The decisive influence of responsiveness to rewarding stimuli in associative learning of newly emerged bees has far reaching consequences for studies on the development of associative learning capabilities in insects and vertebrates.     

Sensory responsiveness and the effects of equal subjective rewards on tactile learning and memory of honeybees

In tactile learning, sucrose is the unconditioned stimulus and reward, which is usually applied to the antenna to elicit proboscis extension and which the bee can drink when it is subsequently applied to the extended proboscis. The conditioned stimulus is a tactile object that the bee can scan with its antennae. In this paper we describe the quantitative relationships between gustatory antennal stimulation, gustatory proboscis stimulation, and tactile learning and memory. Bees are 10-fold more responsive to sucrose solutions when they are applied to the antenna compared to proboscis stimulation. During tactile conditioning, the sucrose solution applied to the proboscis determines the level of acquisition, whereas antennal input is of minor importance. Bees differing in their gustatory responsiveness measured at the antenna differ strongly in their tactile acquisition and memory. We demonstrate how these differences in tactile acquisition and memory can be greatly reduced by calculating equal subjective rewards, based on individual gustatory responsiveness.     

Learning in honey bees with brain lesions: how partial mushroom-body ablations affect sucrose responsiveness and tactile antennal learning

Honey bees are ideal organisms for studying associative learning, because they can rapidly learn different sensory cues, even under laboratory conditions. Classical olfactory learning experiments have shown that the mushroom bodies (MBs), a prominent neuropil of the central nervous system of the bee, are involved in olfactory learning and memory formation. We tested whether the MBs are also involved in tactile antennal learning. As in olfactory learning, bees use the antennae during tactile learning to sense tactile cues. We produced specific MB ablations by applying the mitotic blocker hydroxyurea (HU). In Drosophila, HU-induced brain lesions of the MBs strongly impaired olfactory memory formation. As treatment with HU might also interfere with the processing of the reward stimulus, sucrose, we measured the responsiveness to sucrose stimuli in these bees. Treatment with HU led to partial ablations of the median MB sub-units on one or both sides of the brain. We analysed side-specific effects in double-blind tests, testing sucrose responsiveness separately for each antenna, and conditioning first one antenna and then the other in a reversal learning assay. HU-treated bees without detectable ablations were less responsive to sucrose and had a poorer learning performance than untreated controls. Partial MB ablation did not additionally affect responsiveness to sucrose or tactile antennal learning. Interestingly, bees with bilateral MB ablations did not differ from untreated controls in their learning performance during the first learning phase. During reversal learning, acquisition in these bees was significantly lower than that in untreated controls. It is concluded that HU treatment affects sucrose responsiveness and tactile learning even without detectable ablation of neuropils. The effects of MB ablations on tactile learning are not side-specific and not correlated with the volume of the ablated neuropil. Accepted after revision: 15 January 2001 ❚ Electronic Publication     

Acute ethanol ingestion impairs appetitive olfactory learning and odor discrimination in the honey bee

Invertebrates are valuable models for increasing our understanding of the effects of ethanol on the nervous system, but most studies on invertebrates and ethanol have focused on the effects of ethanol on locomotor behavior. In this work we investigate the influence of an acute dose of ethanol on appetitive olfactory learning in the honey bee (Apis mellifera), a model system for learning and memory. Adult worker honey bees were fed a range of doses (2.5%, 5%, 10%, or 25%) of ethanol and then conditioned to associate an odor with a sucrose reward using either a simple or differential conditioning paradigm. Consumption of ethanol before conditioning significantly reduced both the rate of acquisition and the asymptotic strength of the association. Honey bees also exhibited a dose dependent reduction in arousal/attention during conditioning. Consumption of ethanol after conditioning did not affect recall 24h later. The observed deficits in acquisition were not due to the affect of ethanol on gustatory sensitivity or motor function. However, honey bees given higher doses of ethanol had difficulty discriminating amongst different odors suggesting that ethanol consumption influences olfactory processing. Taken together, these results demonstrate that an acute dose of ethanol affects appetitive learning and olfactory perception in the honey bee.     

Development of an ethanol model using social insects: III. Preferences for ethanol solutions

Experiments are designed to assess whether free-flying honey bees have an aversion to an ethanol solution when given a choice between targets containing an ethanol solution in sucrose or sucrose only. Animals given a choice between a 1% ethanol solution and sucrose only show no aversion to the ethanol solution either in acquisition or extinction. Honey bees given a choice between a 5% ethanol solution and sucrose only show no differences in the initial choice of targets but some ees do switch over to the sucrose-only target. Performance during extinction indicates that bees landed on the previously reinforced sucrose-only target more than the target previously containing the 5% ethanol solution. An experiment in which bees were given a single 5%, ethanol target showed that of 20 bees, 11 returned for the entire 12 trials of the experiment. All bees returned at least 6 times to the 5% ethanol target. Additional experiments were run on harnessed foragers in a palatability study of alcoholic beverages consumed by humans. The results of the palatability experiment indicate that in general, bees prefer more sweet drinks with less alcohol.     

Peak shift in honey bee olfactory learning

If animals are trained with two similar stimuli such that one is rewarding (S+) and one punishing (S?), then following training animals show a greatest preference not for the S+, but for a novel stimulus that is slightly more different from the S? than the S+ is. This peak shift phenomenon has been widely reported for vertebrates and has recently been demonstrated for bumblebees and honey bees. To explore the nature of peak shift in invertebrates further, here we examined the properties of peak shift in honey bees trained in a free-flight olfactory learning assay. Hexanal and heptanol were mixed in different ratios to create a continuum of odour stimuli. Bees were trained to artificial flowers such that one odour mixture was rewarded with 2 molar sucrose (S+), and one punished with distasteful quinine (S?). After training, bees were given a non-rewarded preference test with five different mixtures of hexanal and heptanol. Following training bees’ maximal preference was for an odour mixture slightly more distinct from the S? than the trained S+. This effect was not seen if bees were initially trained with two distinct odours, replicating the classic features of peak shift reported for vertebrates. We propose a conceptual model of how peak shift might occur in honey bees. We argue that peak shift does not require any higher level of processing than the known olfactory learning circuitry of the bee brain and suggest that peak shift is a very general feature of discrimination learning.     

Continuous, fixed-ratio, and fixed-interval reinforcement in honey bees

Bees learned to enter a Plexiglas tube and to suck small portions of sugar solution; every entry or every fifth entry was reinforced. During an extinction phase, the bees on the fixed-ratio schedule emitted twice as many responses as did those given continuous reinforcement. Bees on a fixed-interval schedule of reinforcement emitted lower response rates than did those given fixed-ratio reinforcement. By extending the conditioning procedure for several days, it was possible to maintain responding with fixed-ratio schedules requiring 30 responses per reinforcement and with fixed-interval values up to 90 sec. Under fixed-interval schedules, response rates did not increase toward the end of the reinforcement intervals.     

Differences in performance on a reversal learning test and division of labor in honey bee colonies

We studied the association between honey bee (Apis mellifera) division of labor and performance on an olfactory reversal-learning test. Manipulations of colony age structure and flight experience were used to test whether differences in performance are associated with age, current behavioral state, or flight experience. Nurse bees showed significantly faster rates of extinction to a learned odor than did foragers. This difference was associated primarily with differences in behavioral state, rather than age; it was seen when comparing nurses and foragers from typical colonies and normal-age nurses and precocious foragers from single-cohort colonies. Differences in extinction rate were not related to differences in flight experience; there was no difference between foragers and foraging-age bees denied flight experience. These results suggest that changes in learning and memory occur in association with division of labor. We speculate on the possible functional significance of the difference in extinction rate between nurses and foragers. Received: 15 January 2000 / Accepted: 22 August 2000     

Successive olfactory reversal learning in honeybees

Honeybees Apis mellifera can associate an originally neutral odor with a reinforcement of sucrose solution. Forward pairings of odor and reinforcement enable the odor to release the proboscis extension reflex in consecutive tests. Bees can also be conditioned differentially: They learn to respond to a reinforced odor and not to a nonreinforced one. They can also learn to reverse their choice. Here we ask whether honeybees can learn successive olfactory differential conditioning tasks involving different overlapping pairs of odors. The conditioning schedules were established in order to train the animals with 3, 2, 1, or 0 reversals previous to a last differential conditioning phase in which two additional reversals were present. We studied whether or not successive reversal learning is possible and whether or not learning olfactory discrimination reversals affects the solving of subsequent discrimination reversals. Therefore we compared the responses of bees that had experienced reversals with those of bees that had not experienced such reversals when both are confronted with a new reversal situation. In experiment 1 we showed that bees that had experienced three previous reversals were better in solving the final reversal task than bees with no previous reversal experience. In experiment 2, we showed that one reversal learning is enough for bees to perform better in the final reversal task. The successive different reversals trained in our experiments resemble the natural foraging situation in which a honeybee forager has to switch successively from an initial floral species to different ones. The fact that experiencing such changes seems to improve a bee's performance in dealing with further new exploited food sources has therefore an adaptive impact for the individual and for the colony as a whole.     

The comparator hypothesis of conditioned response generation: manifest conditioned excitation and inhibition as a function of relative excitatory strengths of CS and conditioning context at the time of testing

In the present research water-deprived rats were used in a conditioned lick suppression paradigm to test and further develop Rescorla's (1968) contingency theory, which posits that excitatory associations are formed when a conditioned stimulus (CS) signals an increase in unconditioned stimulus (US) likelihood and that inhibitory associations develop when the CS signals a decrease in US likelihood. In Experiment 1 we found that responding to a CS varied inversely with the associative status of the context in which the CS was trained and that this response was unaltered when testing occurred in a distinctively dissimilar context with a different conditioning history, provided associative summation with the test context was minimized. These results suggest that manifest excitatory and inhibitory conditioned responding is modulated by the associative value of the training context rather than that of the test context. In Experiment 2 it was demonstrated that postconditioning decreases in the associative value of the CS training context reduced the effective inhibitory value of the CS even when testing occurred outside of the training context. Moreover, this contextual deflation effect was specific to the CS training context as opposed to any other excitatory context. Collectively, these studies support the comparator hypothesis, which states that conditioned responding is determined by a comparison of the associative strengths of the CS and its training context that occurs at the time of testing rather than at the time of conditioning. This implies that all associations are excitatory and that responding indicative of conditioned inhibition reflects a CS-US association that is below (or near) the associative strength of its comparator stimulus. It is suggested that response rules which go beyond a monotonic relation between associative value and response strength can partially relieve learning theories of their explanatory burdens, thereby allowing for simpler models of acquisition.     

Aversive learning overcomes appetitive innate responding in honeybees

Despite their miniature brain, honeybees have emerged as a powerful model for the study of learning and memory. Yet, they also exhibit innate responses to biologically relevant social signals such as pheromones. Here, we asked whether the bees’ developed learning capabilities allow them to overcome hardwired appetitive responses. Can they learn that attractant pheromones, that are not normally associated with a noxious stimulation in nature, predict the punishment of an electric shock? Immobilized honeybees were trained to discriminate two odorants, one that was paired with a shock and another that had no consequences. We measured whether they learned to produce aversive sting extension responses to the punished but not the non-punished odorant. One odorant was a neutral odor without innate value while the other was either an attractive pheromone (geraniol or citral) or an attractive floral odorant (phenylacetaldehyde). In all cases, bees developed a conditioned aversive response to the punished odorant, be it pheromone or not, and efficiently retrieved this information 1 h later. No learning asymmetries between odors were found. Thus, associative aversive learning in bees is strong enough to override preprogrammed responding, thus reflecting an impressive behavioral flexibility.     

The acquisition of conditioned responding

This report analyzes the acquisition of conditioned responses in rats trained in a magazine approach paradigm. Following the suggestion by Gallistel, Fairhurst, and Balsam (2004), Weibull functions were fitted to the trial-by-trial response rates of individual rats. These showed that the emergence of responding was often delayed, after which the response rate would increase relatively gradually across trials. The fit of the Weibull function to the behavioral data of each rat was equaled by that of a cumulative exponential function incorporating a response threshold. Thus, the growth in conditioning strength on each trial can be modeled by the derivative of the exponential--a difference term of the form used in many models of associative learning (e.g., Rescorla & Wagner, 1972). Further analyses, comparing the acquisition of responding with a continuously reinforced stimulus (CRf) and a partially reinforced stimulus (PRf), provided further evidence in support of the difference term. In conclusion, the results are consistent with conventional models that describe learning as the growth of associative strength, incremented on each trial by an error-correction process.     

Timing at the Start of Associative Learning

Some theories of associative learning imply that time plays a fundamental role in the acquisition process. Consistent with these theories, this paper presents evidence that the time from the onset of a conditioned stimulus (CS) until presentation of the unconditioned stimulus (US) is learned very rapidly at the start of training. We report two autoshaping studies and a study on aversive conditioning in goldfish in which we examine timing at the start of conditioning. We also review data from a number of other conditioning preparations, including fear-potentiated startle, appetitive conditioning in rats, and eyeblink conditioning in rabbits, that report conditioned response (CR) timing early in training. Acquisition speed and the very first expressions of conditioned responding often show sensitivity to the time of US presentation. In instances where temporal control is slowly expressed, it is likely due to performance factors, not to slow learning about time. In fact, the learning about time may be a necessary condition for associative learning.     

Facilitation of instrumental behavior by a Pavlovian appetitive conditioned stimulus

Three experiments examined appetitive Pavlovian-instrumental interactions by presenting separately trained conditioned stimuli (CSs) during reinforced instrumental responding in rabbits. Intra-oral reinforcement was used to minimize interference from peripheral responses such as magazine approach. In experiment 1, the rabbits were first trained to perform an instrumental head-raising response for sucrose reward. A conditioned jaw movement response was then established to a 2-sec CS by pairing it with sucrose; a control stimulus was unpaired with sucrose. Instrumental responding maintained by a variable-interval 40-sec schedule was enhanced during 10-sec presentations of the paired, but not the unpaired, CS. Responding on a variable-ratio 15 schedule was unaffected except on trials on which the pre-CS baseline response rate was low; in such cases the paired CS caused a long-lasting acceleration of responding. Noncontingent presentation of the sucrose reinforcer itself briefly suppressed responding but had no long-term effect. In Experiment 2, a CS that had been conditioned at a 10-sec duration produced the same pattern of effects as in the first study, indicating that facilitation resulted from CS presentation rather than from the frustrative effects of non-reinforcement of the CS. In Experiment 3 an inhibitory CS blocked facilitation by the excitatory CS but did not itself affect instrumental responding. These results support the view that Pavlovian processes play a positive role in instrumental performance and suggest that previous findings of suppression by a short-duration CS reflect peripheral interference. The dependence of facilitation on the baseline level of responding is discussed in terms of associative and motivational theories of Pavlovian mediation.     

Control of appetitive and aversive taste-reactivity responses by an auditory conditioned stimulus in a devaluation task: a FOS and behavioral analysis

Through associative learning, cues for biologically significant reinforcers such as food may gain access to mental representations of those reinforcers. Here, we used devaluation procedures, behavioral assessment of hedonic taste-reactivity responses, and measurement of immediate-early gene (IEG) expression to show that a cue for food engages behavior and brain activity related to sensory and hedonic processing of that food. Rats first received a tone paired with intraoral infusion of sucrose. Then, in the absence of the tone, the value of sucrose was reduced (Devalue group) by pairing sucrose with lithium chloride (LiCl), or maintained (Maintain group) by presenting sucrose and LiCl unpaired. Finally, taste-reactivity responses to the tone were assessed in the absence of sucrose. Devalue rats showed high levels of aversive responses and minimal appetitive responses, whereas Maintain rats exhibited substantial appetitive responding but little aversive responding. Control rats that had not received tone-sucrose pairings did not display either class of behaviors. Devalue rats showed greater FOS expression than Maintain rats in several brain regions implicated in devaluation task performance and the display of aversive responses, including the basolateral amygdala, orbitofrontal cortex, gustatory cortex (GC), and the posterior accumbens shell (ACBs), whereas the opposite pattern was found in the anterior ACBs. Both Devalue and Maintain rats showed greater FOS expression than control rats in amygdala central nucleus, GC, and both subregions of ACBs. Thus, through associative learning, auditory cues for food gained access to neural processing in several brain regions importantly involved in the processing of taste memory information.     

Analysis of an ambiguous-feature discrimination

An autoshaping experiment with pigeons and three appetitive Pavlovian conditioning experiments with rats investigated the course of an ambiguous-feature discrimination in which trials with stimuli A and BC were followed by food, and trials with B and AC were not. The discrimination between A and B was acquired more rapidly than the discrimination between BC and AC in all experiments. Furthermore, the acquisition of conditioned responding with A was faster than that with BC for the three rat experiments. The results of these experiments are discussed in terms of elemental and configural theories of associative learning.     

Analysis of an ambiguous-feature discrimination

An autoshaping experiment with pigeons and three appetitive Pavlovian conditioning experiments with rats investigated the course of an ambiguous-feature discrimination in which trials with stimuli A and BC were followed by food, and trials with B and AC were not. The discrimination between A and B was acquired more rapidly than the discrimination between BC and AC in all experiments. Furthermore, the acquisition of conditioned responding with A was faster than that with BC for the three rat experiments. The results of these experiments are discussed in terms of elemental and configural theories of associative learning.     

Side-specificity of olfactory learning in the honeybee: generalization between odors and sides

Honeybees (Apis mellifera) can be trained to associate an odor stimulus with a sucrose reward. The neural structures involved in the detection and integration of olfactory stimuli are represented bilaterally in the brain. Little is known about the respective roles of the two sides of the brain in olfactory learning. Does each side learn independently of the other, or do they communicate, and if so, to what extent and at what level of neural integration? We addressed these questions using the proboscis extension response (PER) conditioning paradigm applied in a preparation that allows the separation of the two input sides during olfactory stimulations. Bees conditioned to two odorants A and B, one being learned on each side (A+/B+ training), showed in extinction tests rather unspecific responses: They responded to both odorants on both sides. This could be attributable to either a transfer of the learned information between sides, or to a generalization between odorants on each side. By subjecting bees to conditioning on one side only (A+/0 training), we found that the learned information is indeed transferred between sides. However, when bees were trained explicitly to give opposite values to the two odorants on the two sides (A+B-/B+A- training), they showed clear side-specific response patterns to these odorants. These results are used in the elaboration of a functional model of laterality of olfactory learning and memory processing in the honeybee brain.     

Associative mechanosensory conditioning of the proboscis extension reflex in honeybees

The present work introduces a form of associative mechanosensory conditioning of the proboscis extension reflex (PER) in honeybees. In our paradigm, harnessed honeybees learn the elemental association between mechanosensory, antennal stimulation and a reward of sucrose solution delivered to the proboscis. Thereafter, bees extend their proboscis to the antennal mechanosensory stimulation alone. We show that bees can learn such an association in a side-specific manner, that is, they learn the association on the antennal side that was rewarded and not on the side that was not rewarded. Responding produced by the paired training does likely contain a substantial Pavlovian component. Responding is only elicited by mechanosensory stimulation and not by spurious cues such as olfactory, visual, and contextual ones. The interstimulus interval (ISI) affects one-trial mechanosensory learning: a bell-shaped curve with a maximum of responding approximately 4 sec ISI was obtained. Mechanosensory memory is still operative 24 h after conditioning. Apart from absolute conditioning in which mechanosensory stimulation of one antenna is paired with sucrose, differential, side-specific, mechanosensory conditioning using two mechanosensory stimulations, one rewarded and the other not, is also possible. This paradigm constitutes, therefore, a new standard procedure for further learning studies in honeybees.     

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.