B

Memory Professor System

10x Your Memory Power

Get Instant Access

Fig. 2.2 (cont). (b) Memory retrieval is initiated by local and contextual stimuli, which are believed to feed into each memory store directly for the activation of stimulus-specific memory traces. The contents of these traces are integrated into working memory, a memory phase controlling behavior via a neural stage addressed as "expectation". This stage is a necessary assumption from experiments showing that learning follows the difference rule as formulated by Rescorla & Wagner (1972) (see Greggers & Menzel 1993). As a consequence, retrieval is intimately combined with new learning, which leads to an updating of the memory contents according to the mismatch between the expected and the experienced outcome of behavior.

estimated and may vary to a large degree, but they are likely to last longer on average than within-patch choices. The distribution of interbout intervals from training experiments with bees on artificial feeders had a median around 4-5 min (Menzel 1987, Fig. 1a). Much longer intervals are expected during bad weather conditions, e.g., Lindauer (1963) reported that overwintering bees visited the feeding place from the previous autumn at their first flight in the spring.

Different memories may determine choice behavior at these different intervals. Evidence for different memory phases in the honeybee comes from behavioral, neurophysiological, and biochemical studies (Menzel & Müller 1996; Menzel 1999). The concept emerging from these results assumes five sequential stages during the process of memory formation (Fig. 2.2a). Consolidation from early to late memory stages is time- and event-dependent, meaning that both elapsing time and new experience during the process of consolidation define the speed of transfer between the memory phases. Most importantly, the consolidation process changes the content of memory, a general property of memory processing in animals (including invertebrates and humans; Müller & Pilzecker 1900; Milner et al. 1998).

I shall first give a short characterization of the mechanistic basis of the five memory stages and relate them to behavioral measures of retention as revealed by simple forms of associative learning. Then I shall try to incorporate this information into a model of sequential decision-making as it relates to choice behavior of a foraging bee. At this point it will become important to consider two aspects of memory: memory formation and memory retrieval (Fig. 2.2a, b). During foraging, both processes are intimately connected, and it is extremely difficult to assign any particular character of the choice behavior to one process or the other. The central theme here will be the concept of working memory, a form of memory that controls ongoing behavior and retrieves its information from all memory stores.

Memory phases

Associative induction and early short-term memory

An associative learning trial involves the pairing of the stimuli to be learned (conditioned stimuli, CS) with the rewarding stimulus (e.g., sucrose solution; unconditioned stimulus, US). Olfactory conditioning, for example, leads to associative induction and an early form of short-

Fig. 2.3. Temporal dynamics of choice behavior in single honeybees foraging in a patch of four artificial feeders (A, B, C, D). Each feeder provided the same flow rate of sucrose solution. The flow rate was so low that the bee visited each of the feeders. Since the flow rate was similar in all feeders, the frequency of visits was the same for each feeder. Two kinds of subsequent choices were performed: a: next choices called "stay" and "shift" flights, and B: second subsequent choices - here they are called "same" and "different" choices, depending on whether the bee flies back to the same feeder after visiting another feeder or flies to a different feeder. One of the four possible cases is shown in a and b. Reward memory for the last (a) or second-to-last (b) feeder visited was evaluated by correlating lick time at the last or second-to-last

Fig. 2.3. Temporal dynamics of choice behavior in single honeybees foraging in a patch of four artificial feeders (A, B, C, D). Each feeder provided the same flow rate of sucrose solution. The flow rate was so low that the bee visited each of the feeders. Since the flow rate was similar in all feeders, the frequency of visits was the same for each feeder. Two kinds of subsequent choices were performed: a: next choices called "stay" and "shift" flights, and B: second subsequent choices - here they are called "same" and "different" choices, depending on whether the bee flies back to the same feeder after visiting another feeder or flies to a different feeder. One of the four possible cases is shown in a and b. Reward memory for the last (a) or second-to-last (b) feeder visited was evaluated by correlating lick time at the last or second-to-last

feeder visited and the feeder for which reward memory is being evaluated (actual feeder). There is a high correlation when the same feeder is visited, and no correlation when a different feeder is visited (see Greggers & Mauelshagen 1997). Correlation (as a measure of memory, and thus expectancy during the test visit) depends on the interval between visits. a: Next choices follow each other quickly (<1 min). "Stay" and "shift" flights show the same rapidly decaying time dependence of correlation of the respective lick times. This indicates that at short intervals during eSTM, reward memory is unspecific for the feeder visited. b: Second subsequent choices follow each other at intervals of 1-7 min (thus during the lSTM time window). The time course of choices differs for "same" and "different" choices. At different feeders, bees do not expect to get the same reward as during the second-to-last visit at any of the three other feeders; at the same feeder, however, they develop an increasing (consolidating) memory for reward at that feeder. (Data for this figure come from unpublished work by U. Greggers.)

term memory (eSTM) over the time course of a few seconds (Bitterman et al. 1983; Menzel 1990). This memory is strongly dominated by appetitive arousal induced by the US. Arousal sensitizes animals to a broad range of stimuli generally related to the arousal-inducing events, e.g., the unexpected experience of food. eSTM is rather unspecific and imprecise. For example, a bee conditioned to an odor by one trial will respond to other stimuli (a mechanical stimulus to the antenna or to other odor stimuli) much more strongly during the first 30 s than later (Menzel 1990; Smith

Furthermore, in an experiment with free-flying bees choosing among four feeders of continuous but low sucrose solution flow, Greggers & Mauelshagen (1995) found that the lick time during each visit is a measure of what the animals recall about the reward at a particular feeder. Thus, the correlation between actual and last lick time can be taken as a measure of reward memory (Fig. 2.3). If sequential visits over short intervals (<1 min) are considered, memory is unspecific for the four feeders, and decays quickly. The early form of STM is, therefore, characterized by general arousal, which decays within one minute. eSTM covers the time window during which bees can expect to be exposed to the same stimuli. No specific choices need to be performed at this time, and general arousal (depending on the strength of the US) will suffice to control whether the animal stays in the patch or postpones choices for a later time. Affirmative information in this time window will inform the animal whether specific memory storage is worthwhile, and this will lead to quicker formation of longer-lasting forms of memory (see below).

At the level of cellular and neural processes, stimulus association is reflected in the convergence of excitation of the CS and US pathways. The neuroanatomical convergence sites of these pathways are known for olfactory learning in bees (antennal lobe, lip region of mushroom body calyces, lateral protocerebrum; Hammer 1997), and the putative primary transmitters and second messenger pathways have been identified (Menzel & Müller 1996).

Late short-term memory (ISTM)

The transition to the selective associative memory trace (consolidation) during lSTM is a rather slow process after a single learning trial, lasting up to several minutes. It is quicker after multiple learning trials (Menzel 1968; Erber 1975a, b; review by Menzel 1999). Thus, consolidation of the stimulus-specific associative memory trace is both time- and event-dependent. Odor-conditioned animals show increasing retention for intervals >3 min, and during this consolidation period retention becomes more specific for the learned odor. Free-flying bees foraging in the patch of four feeders mentioned above (Greggers & Mauelshagen 1995; see Fig. 2.3) behave identically: retention for the reward quality of each feeder rises after a minimum around 2 min and increases over the next minutes, becoming more and more specific for each feeder. Therefore, consolidation during lSTM is a process that establishes a more precise memory. This memory is more resistant to new information (Menzel 1979, 1990), and is no longer susceptible to amnestic treatment (Menzel 1968; Erber 1975a, b). Transition from eSTM to lSTM is accelerated by multiple learning trials and learning trials in quick succession.

The behavioral relevance of these findings for foraging behavior under natural conditions may be related to the temporal separation between intra- and interpatch visits, and the different memories established for later use. First, memory needs to be highly specific after leaving a patch, because distinctions need to be made between similar and different flowers. Second, such a specific memory trace should also be established after a single learning trial, because in some rare cases a single flower may offer a very high amount of reward. Third, discovering a rewarding flower in a different patch means that the local cues just learned are now presented in a different context (localization within GLM; see above). Consolidation is the process that allocates different memories to different stores, enabling the bee to store many different memories according to the contextual cues related to the separate experiences. How many memories may reside in working memory will be discussed below.

Mid-term memory (MTM) At the beginning of MTM, behavior is controlled by consolidated, highly specific memory. At this stage, memory is more resistant to extinction, conflicting information, and elapsing time, and some context-dependent information may have already been stored. Under natural conditions, bees have usually returned to the hive and departed on a new foraging bout usually within the time window of MTM. Upon arrival at the feeding area, memory for flower cues does not reside in STM any longer, but needs to be retrieved from a more permanent store and put into working memory. Therefore, MTM is a memory stage clearly disconnected from a continuous stream of STMs into working memories, which regulate foraging behavior during the quick successions of intra- and interpatch choices (see below).

MTM is physiologically characterized by the continuing activation of a particular second messenger pathway, the protein kinase C pathway (Grünbaum & Müller 1998). It was concluded that both the sensory neuropil (the antennal lobe, in the case of olfactory learning) and the mushroom bodies are involved in the memory trace. The mushroom body provides the information that relates the memory traces in the primary sensory neuropils to contextual stimuli across modalities. The memories necessary to guide the bee back to the feeding place after returning to the hive involve many different behavioral faculties (e.g., compass orientation to celestial and landmark cues, information about the time of day, sequential landmark appearance, social encounters, and information about the colony's needs). These behavioral faculties can be integrated only across sensory modalities, and are most likely related to mushroom-body function (Menzel et al. 1994; Rybak & Menzel 1998).

Long-term memory (LTM) LTM requires multiple learning trials, indicating that specific information, which can be extracted only from multiple experiences (reliability of signals, context dependence), characterizes its contents. Two forms of LTM are distinguished based on physiological characteristics - early LTM (eLTM, between several hours and 2 days) and late LTM (lLTM, >3 days). Only lLTM depends on protein synthesis, and thus structural changes in the wiring of neurons appear to store memory lasting longer than 3 days

(Grünbaum & Müller 1998). The physiological basis of eLTM is not yet well understood. Since protein-synthesis blockade does not interfere with eLTM formation, and the signaling pathway of protein kinase C is constitutively activated, it is concluded that covalent transformations within the existing neural pathways store eLTM.

The biological circumstances of two forms of LTM may be related to the distinction between those forms of learning that usually lead to lifelong memories (e.g., visual and olfactory cues characterizing the home colony) and those that are stable but need updating on a regular basis (e.g., visual and olfactory cues of feeding places). However, lifelong memories can also be formed for floral cues (e.g., color information; see Menzel 1968, Lindauer 1963), and standard lifelong memories (e.g., localization of the colony) can also be changed by new experience (e.g., swarming).

Sequential choice behavior during foraging: memory dynamics at work

Memories guide choice behavior. Anyone who has trained bees to a feeding station knows that bees choose the signals associated with reward even after long intervals: days, weeks, or - as Lindauer (1963) reported -several months. Therefore, bees activate remote memories stored in long-term form when motivational and contextual conditions are favorable. Similarly, memories kept in MTM and STM will contribute. The active form of all these types of memory is usually called working memory. Working memory is understood to provide the animal with specific expectations about the signals, reward conditions, and manipulatory requirements of a food source for which a bee strives. The choice resulting from an expectation will lead to new experience and thus to new learning, which - in the manner discussed above - will add new information to memory (Fig. 2.2b).

Working memory will be continuously updated by STM during the quick succession of flower visits during a foraging trip, and, therefore, the dynamics of STM will be most important for the actual status of working memory. The content of working memory is certainly limited, as it is in all animals (Baddeley 1986). It is, therefore, likely that the intervals between successive choices and the sequence of experiences will control the expectation, or non-surprisingness as proposed by Wagner (1978), at any moment of the choice process. Indeed, expectation about reward proper ties changes over time (Fig. 2.3). As already pointed out, very short intervals (<1 min) lead to high but rather unspecific expectations, intermediate intervals (1-2 min) to low and unspecific expectations, and long intervals to specific expectations. These observations apply to expectations both for reward properties and learned signals, because the same time dependencies were found in olfactory conditioning experiments and in tests with free-flying bees trained to color signals (Menzel 1999). Some indications that favor this interpretation come from observations by Chittka et al. (1997), who recorded the frequency of intervals between stay and shift flights of bumble bees foraging on more than two plant species. Stay flights appear at shorter intervals than shift flights, indicating that immediate choices are dominated by the most recent and the most effective STM, but reference to more remote memories needs more time, or, to interpret it from another perspective, longer intervals release working memory from the dominant memory of the last visit and allow for contributions from earlier, consolidated memories. The time-scale in the observations of Chittka et al. (range of 1-20 s) differs from that in the experiments with bees foraging in a patch of four artificial feeders mentioned above (Fig. 2.3) (Greggers & Menzel 1993; Mauelshagen & Greggers 1993; Greggers & Mauelshagen 1995), in which the average retrieval between successive visits is in the 1-min range. However, since handling time ranges from 20-40 s in the Greggers et al. experiments, the actual intervals in flight time are rather similar. I conclude, therefore, that the temporal dynamic of STM contributes substantially to working memory, and that initial and later phases (eSTM and lSTM) are an important determinant of working memory.

Capacity of working memory can also be estimated from the experiments with artificial feeders. As Fig. 2.3B indicates, the expectancy during revisits to the same feeder after second subsequent choices is significantly different from expectancies expressed in visits to different feeders. This finding applies to any of the four feeders. Therefore, bees store the reward properties of at least four different feeders in working memory. The same result was found for eight feeders, indicating that the reward properties of eight feeders can also be stored in eight feeder-specific memories. A larger number of feeders has not yet been tested. The capacity of working memory is, therefore, at least eight items, and the time-range for all these specific working memories lies above 6 min (Fig. 3B).

The framework of a mechanistic model of flower choice

It is well documented that memory in both animals and humans is processed in temporal stages. The cellular substrates appear surprisingly similar both among different species (Aplysia, Drosophila, mouse, chick, man) and among different forms of memory (memories consciously addressable and those under automatic control, appetitive and aversive memories, and emotional and non-emotional ones; Milner et al. 1998; Rosenzweig 1998). This has led to the assumption that the memory formation process is determined by its underlying cellular machinery, and that similar time courses for the respective stages are indicative of general mechanisms rather than species-specific and task-specific adaptations. However, studies on memory stages have focused primarily on their neural and cellular substrates, and have not yet asked the question of how these stages are adapted to the needs of an animal behaving in natural surroundings. In particular, very little attention has been paid to the dynamics of natural sequences of behavioral events that simultaneously create new memory and need to be controlled by memory. The notion presented here is that the cellular machinery may not be the defining factor, but, rather, the systems' requirements for the installation, sequence, and character of memory stages. In particular, I favor the view that the similarities in memory stages discovered so far reflect basic and general requirements of the continuous process of concurrently learning, retrieving, storing, and applying information (Menzel 1999).

Species- and task-specific adaptations are to be expected, and these may be the deciding parameters for the dynamics and significance of memory stages. As pointed out above, we need to separate memory states during memory formation and during memory retrieval (Fig. 2.2). During choice performance, memory retrieval guides the next choice, but it is important to keep in mind that any experience will always induce a learning process, which in turn leads to memory formation and alteration of the content of all memory stages. In fast behavioral sequences, such as during foraging within and between flower patches, STM of the last encounter will first feed strongly into working memory, but with time elapsing, the memory from former experiences will gain by a consolidation process. Specifying properties of a food source, such as local signals and contextual cues, will become increasingly important during consolidation. These highly specific memories are stored via multiple experi ences in MTM and LTM, and their contribution to working memory will make expectations rather specific.

A particular aspect of memory specificity is the combination of different memories established independently in LTM. One of these memories, general landscape memory (GLM), has been discussed here. The hypothesis is put forward that food sources are represented along with their properties (signals, rewards, mechanics) in GLM, and these can be chosen by directed flights between them, even over long distances.

The capacity of such a compound LTM is unknown, but it may be safe to assume that multiple locations can be stored that represent different loci in GLM. The spatial resolution of such loci and their maximum number will have to be addressed in future experiments. Future experiments must also evaluate the upper limits of working memory with respect to both the capacity of stored items and the timespan. The ranges found so far for bees (at least eight items, over at least 6 min for a honeybee foraging in a patch of artificial feeders) are already quite impressive when compared with other animals (Baddeley 1986).

Another issue refers to the question of which memory stages (with their accompanying dynamics) are evolutionarily adapted to the task to be solved, namely efficient foraging in an unreliable and scattered food market. Although a mechanistic model need not refer to ultimate causation, it is tempting to speculate in what sense the structure and dynamics of memories might be shaped by evolution. I noted in the introductory section of this chapter that the patchiness of food sources poses different demands on the tasks to be solved in sequence. The decision to stay in the patch is mainly a motivational one controlled by the amount (or quality) of food gained as compared to the amount (or quality) expected. Thus, rather unspecific behavioral control as induced by food arousal is a dominant characteristic of choices occurring in quick succession. I would expect that the dynamic of decay in arousal might reflect the spacing (as measured in flight time) between chosen patches.

The concept of consolidation of associative events includes the notion that different memory items are consolidated separately (Müller & Pilzecker 1900), a notion that has been substantiated for the honeybee. Different memory items are characterized and later retrieved specifically by their contextual cues, which should be mainly those defining different locations in GLM. Thus, lSTM should be adapted to the spacing between patches. It might be an interesting question whether species adapted to different intrapatch distances or with very different flight speeds developed different dynamics of lSTM.

Flower constancy of hymenopteran pollinators results from choice behavior, which is at any moment guided by the memory of former experience. The richness, duration, complexity, and dynamics of memory have been underestimated, and have only recently become clear. Although we still have to learn a large amount about the structures, mechanisms, and contents of the various forms of memory, we certainly can no longer assum that major components of the choice processes are dictated by the limited capacity or duration of memory. Rather, it is the dynamics of the memory stages and their transitions that allow for highly flexible choice behavior and thus for flower constancy.

References

Baddeley AD (1986) Working memory. Oxford University Press, Oxford Becker L (1958) Untersuchungen über das Heimfindevermögen der Bienen. Z vergl Physiol 41:1-25

Bitterman ME, Menzel R, Fietz A & Schäfer S (1983) Classical conditioning of proboscis extension in honeybees (Apis mellifera). J Comp Psychol 97:107-119 Capaldi EA & Dyer FC (1999) The role of orientation flights on homing performance in honeybees. J Exp Biol 202:1655-1666 Capaldi EA, Smith AD, Osborne JL, Fahrbach SE, Farris SM, Reynolds DR, Edwards AS, Martin A, Robinson GE, Poppy GM & Riley JR (2000) Ontogeny of orientation flight in the honeybee revealed by harmonic radar. Nature 403:537-540 Chittka L, Gumbert A & Kunze J (1997) Foraging dynamics of bumble bees: correlates of movements within and between plant species. Behav Ecol 8:239-249 Collett TS (1992) Landmark learning and guidance in insects. Phil Trans R SocLondB

337:295-303

Collett TS & Zeil J (1998) Places and landmarks: an arthropod perspective. In: Healy S (ed) Spatial representation in animals, pp 18-53. Oxford University Press, Oxford Dyer FC & Gould JL (1981) Honey bee orientation: a backup system for cloudy days. Science 214:1041-1042

Erber J (1975a) The dynamics of learning in the honeybee (Apis mellifera carnica). I. The time dependence of the choice reaction. J Comp Physiol 99:231-242 Erber J (1975b) The dynamics of learning in the honeybee (Apis mellifera carnica). II.

Principles of information processing. J Comp Physiol 99:243-255 Esch HE & Burns JE (1995) Honeybees use optic flow to measure the distance of a food source. Naturwiss 82:38-40 Frisch K von (1965) Tanzsprache und Orientierung der Bienen. Springer-Verlag, Heidelberg Gould JL (1986) The locale map of honey bees: do insects have cognitive maps? Science 232:861-863

Greggers U & Mauelshagen J (1995) The temporal composition of task-specific and reward related memory components in free-flying honeybees. In: Elsner N &

Menzel R (eds) Learning and memory, Proceedings of the 23rd Göttingen Neurobiology Conference, p 13. Thieme Verlag, Stuttgart Greggers U & Mauelshagen J (1997) Matching behavior of honeybees in a multiple-

choice situation: the differential effect of environmental stimuli on the choice process. Anim Learn Behav 25:458-472 Greggers U & Menzel R (1993) Memory dynamics and foraging strategies of honeybees.

Behav Ecol Sociobiol 32:17-29 Grünbaum L & Müller U (1998) Induction of a specific olfactory memory leads to a long-lasting activation of protein kinase C in the antennal lobe of the honeybee. J Neurosci 18:4384-4392 Hammer M (1997) The neural basis of associative reward learning in honeybees. Trends

Neurosci 20:245-252 Koltermann R (1971) 24-Std-Periodik in der Langzeiterinnerung an Duft- und

Farbsignale bei der Honigbiene. Z vergl Physiol 75:49-68 Lehrer M (1999) Dorsoventral asymmetry of colour discrimination in bees. J Comp Physiol A 184:195-206

Lindauer M (1963) Allgemeine Sinnesphysiologie. Orientierung im Raum. Fortschr Zool 16:58-140

Mauelshagen J & Greggers U (1993) Experimental access to associative learning in honeybees. Apidol 24:249-266 Menzel R (1968) Das Gedächtnis der Honigbiene für Spektralfarben. I. Kurzzeitiges und langzeitiges Behalten. Z vergl Physiol 60:82-102 Menzel R (1979) Behavioral access to short-term memory in bees. Nature 281:368-369 Menzel R (1985) Learning in honey bees in an ecological and behavioral context. In:

Hölldobler B & Lindauer M (eds) Experimental behavioral ecology, pp 55-74. Gustav Fischer Verlag, Stuttgart Menzel R (1987) Memory traces in honeybees. In: Menzel R & Mercer AR (eds)

Neurobiology and behavior of honeybees, pp 310-325. Springer-Verlag, Berlin Menzel R (1990) Learning, memory, and "cognition" in honey bees. In: Kesner RP & Olton DS (eds) Neurobiology of comparative cognition, pp 237-292. Erlbaum, Hillsdale, NJ

Menzel R (1999) Memory dynamics in the honeybee. J Comp Physiol A 185:323-340 Menzel R & Giurfa M (2001) Cognitive architecture of a minibrain: the honeybee. Trends Cognite Sci, in press.

Menzel R & Müller U (1996) Learning and memory in honeybees: from behavior to neural substrates. Ann Rev Neurosci 19:379-404 Menzel R, Durst C, Erber J, Eichmüller S, Hammer M, Hildebrandt H, Mauelshagen J, Müller U, Rosenboom H, Rybak J, Schäfer S & Scheidler A (1994) The mushroom bodies in the honeybee: from molecules to behavior. In: Schildberger K & Elsner N (eds) Neural basis of behavioral adaptations, Fortschritte der Zoologie vol. 39, pp 81-102. Gustav Fischer Verlag, Stuttgart Menzel R, Geiger K, Chittka L, Joerges J, Kunze J & Müller U (1996) The knowledge base of bee navigation. J Exp Biol 199:141-146 Menzel R, Gumbert A, Kunze J, Shmida A & Vorobyev M (1997) Pollinators' strategies in finding flowers. Israel J Plant Sci 45:141-156 Menzel R, Geiger K, Müller U, Jörges J & Chittka L (1998) Bees travel novel homeward routes by integrating separately acquired vector memories. Anim Behav

55:139-152

Menzel R, Giurfa M, Gerber B & Hellstern F (1999) Elementary and configural forms of memory in an insect: the honeybee. In: Friederici AD & Menzel R (eds) Learning: rule extraction and representation, pp 259-282. Walter de Gruyter, Berlin Menzel R, Brandt R, Gumbert A, Komischke B & Kunze J (2000) Two spatial memories for honeybee navigation. Proc R Soc Lond B 267:961-968 Milner B, Squire LR & Kandel ER (1998) Cognitive neuroscience and the study of memory. Neuron 20:445-468 Müller GE & Pilzecker A (1900) Experimentelle Beiträge zur Lehre vom Gedächtnis. Z Psychol 1:1-288

Rescorla RA & Wagner AR (1972) A theory of classical conditioning: variations in the effectiveness of reinforcement and non-reinforcement. In: Black AH & Prokasy WF (eds) Classical conditioning II: Current research and theory, pp 64-99. Appleton-Century-Crofts, New York Riley JR, Smith AD, Reynolds DR, Edwards AS, Osborne JL, Williams IH, Carreck NL, &

Poppy GM (1996) Tracking bees with harmonic radar. Nature 379:29-30 Rosenzweig MR (1998) Some historical background of topics in this conference.

Neurobiol Learn&Mem 70:3-13 Rybak J & Menzel R (1998) Integrative properties of the Pe1-neuron, a unique mushroom body output neuron. Learn & Mem 5:133-145 Smith BH (1991) The olfactory memory of the honeybee Apis mellifera. I. Odorant modulation of short- and intermediate-term memory after single-trial conditioning. J Exp Biol 161:367-382 Srinivasan MV, Zhang SW, Lehrer M & Collett TS (1996) Honeybee navigation en route to the goal: visual flight control and odometry. J Exp Biol 199:237-244 Vollbehr J (1975) Zur Orientierung junger Honigbienen bei ihrem ersten

Orientierungsflug. Zool Jb Physiol 79:33-69 Wagner AR (1978) Expectancies and the priming of STM. In: Hulse SH, Fowler H, & Honig WK (eds) Cognitive processes in animal behavior, pp 224-251. Erlbaum, Hillsdale, NJ

Wehner R (1992) Arthropods. In: Papi F (ed) Animal homing, pp 45-144. Chapman & Hall, London

Wehner R & Menzel R (1990) Do insects have cognitive maps? Ann RevNeurosci

13:403-414

Was this article helpful?

0 0

Post a comment