Mapping in the context of natural histories

Some insights concerning intricate mechanisms of spatial orientation and navigation came from careful study of species for which some aspect of their natural history makes spatial navigation particularly important for biological success (Kamil and Cheng, 2001). We have already mentioned homing in birds and insects. Many bird species from which pigeons have been studied best of all are known to use multiple ways of orientation such as magnetic fields, sun compass, mosaic of local landmarks, and odour. The cleverest insects, Hymenopterans, use in general the same set of ways of orientation. Here we consider several special examples concerning foraging processes in these two groups of organisms that require excellent spatial memory and orientation.

One amazing example of a navigating system is that of foraging hymenopterans. One explanation of how a honey bee finds a goal is that they follow a sequence of place-finding servomechanisms, following a vector to the vicinity of the target homing on a landmark located near the target and then flying in a stereotypical direction to match the image they see to a remembered image (Cheng, 2000).

As in many other animals, complex navigating behaviour in honeybees can be divided into several sequences that in turn are characterised with more or less specific features. Some of these features can be explained with the use of common properties of insects’ physiology, in particular, their visual systems. For insects, an image of three-dimensional world on the retina is only two-dimensional. Whereas vertebrates have evolved several mechanisms for depth perception, such as stereotypic vision, convergence of the eyes or accommodation of the lens, most insects lack these mechanisms (Mazokhin-Porshnyakov, 1969; Mazokhin-Porshnyakov et al., 1977; Frantsevich, 1993.).

The first behavioural pattern the bee performs when it has found a novel and rich food source is the so-called circling behaviour. When bees depart from a novel feeding site for the first several times, they fly around it in wide and high circles. During these flights they memorise the landmark panorama around the feeding place (Frisch, 1967; Gould, 1986). The circling behaviour does not provide information about such details as shape and colour of the feeding site. For these purposes other patterns serve which has been discovered by M. Lehrer and co-authors (Lehrer, 1987, 1993; Lehrer et al., 1985, 1988; Srinivasan et al., 1989) and called subsequently “ scanning behaviour ” and “ turn-back-and-look-behaviour ” (TBL). Scanning behaviour is displayed by the bees when they leave a novel feeder and try to remember its visual pattern. In experiments with the use of artificial black-and-white patterns the bees followed the contours of the patterns in front of which they were flying. This behaviour is effective to discriminate models that are placed on a vertical plane. In experiments in which depth cues were important as the main feature of the novel food source, the bees, upon leaving a reward box, turned around to view the entrance, approaching it again and again. For example, in one series of experiments bees were trained to collect food on a white paper “meadow” surrounded by a white wall. Six black discs of different size, each carrying a trop of water, were placed flat on the ground; while one disc, placed on a stalk above the ground, offered a reward of sugar water. The positions and the sizes of all seven discs varied, so the only cue which could be used by the bees was its height above the ground, which was the only parameter that was kept constant. In other experiments the bees were trained to collect sugar water at an edge between two surfaces covered with different visual patterns. Colour learning tasks have been involved in other series of tests (Lehrer, 1996, 1998; Lehrer and Bianko, 2000). It has been revealed that the TBL-phase of remembering spatial cues is based on specific of bee’s eye. The bees use the speed of image motion in a variety of tasks involving small-scale navigation. These insects use different cues not only under different circumstances but also in different eye regions. For example, colour discrimination in the lower half of the frontal eye region is better than in the upper half, and estimation of the distance flown is not a function of the ventral visual field. At the same time, the bees effectively employ many options in a process of small-scale navigation depending on the context of their vital tasks.

Knowing different behavioural patterns and specificity of their visual system, we now can predict what the bee will do in different situations. For example, J.L.Gould and C.G. Gould (1995) observed that honey bees learn the colours of flowers only in the last two seconds before landing, and fix landmarks in memory only when they leave. A bee transported to a source of nectar, and later transported back to the hive, cannot find the place again. Even the location and appearance of their own hive is learned only when leaving the first time each morning. If the hive is relocated after the first flight, the bee that leaves the relocated hive cannot find it again. As the authors sum up, the bees learn exactly what they were programmed to learn, exactly when they are programmed to learn it.

The distance information acquired during the TBL is used only in the initial phase of visiting the novel feeding site. Experienced bees arriving at a familiar food source use the size of the landmark as a cue to distance. Lehrer and Collett (1994) propose that cues to distance are particularly important in the initial phase of learning, because near landmarks are more useful for pin-pointing the goal than are more distant ones. Thus, the insect must first of all determine which marks are near and should, therefore, be memorised.

It is easier for us to imagine a process of small-scale navigation during foraging trips in pedestrian hymenopterans. Recent studies have shown that ants take several “snapshots” of a familiar beacon from different vantage points. This mechanism reminds the TBL in bees and works as if the ants had acquired an eidetic image – a “photographic snapshot” – of the landmark panorama around their nesting site and then move in the new area to match their individually acquired template as closely as possible with their current retinal image (Wehner, 1997).

As it has been obtained by Judd and Collett (1998) from their laboratory experiments on red wood ants presented with a sucrose reward placed near artificial landmarks (black cones), foraging individuals take several snapshots of a familiar beacon from different vantage points. An ant leaving a newly discovered food source at the base of a landmark performs a tortuous walk back to its nest during which it periodically turns back and faces the landmark. The ant, on revisiting the familiar landmark, holds the edges of the landmark’s image steady at several discrete positions on its retina. These preferred retinal positions tend to match the positions of landmark edges that the ant captured during its preceding “learning walks”. So, as Srinivasan (1998) notes, red wood ants match as they march during their foraging journeys.

Another biologically significant situation, in which spatial navigation plays a central role, is the recovery of stored food by several species of birds and small mammals. These animals are highly dependent on stored resources for survival and reproduction. We have already met with these masterly creators/detectors of treasures in Part III and will return to them in Part VII to investigate constituents of their special education. Here a central problem is what special ways these animals use for spatial navigation in order to recover thousands caches hidden by themselves.

Shettleworth and Krebs (1982) carried out experiments with marsh tits. These birds hide seeds in tree branches, and are able to remember months later the location of thousands of hidden seeds. The seeds are found using a memory for spatial coordinates. When landmarks (for example rocks) are moved, the birds look for seeds in the wrong places; they search where the seeds would be if they had been moved together with the rocks.

Researchers carried out long-term experiments with another cashier - Cark’s nutcracker (Nucifraga columbiana). In an autumn, with a good pine seed crop, an individual nutcracker will cash ten of thousands of pine seeds in thousands of locations, subsequently returning to them throughout winter and spring (Tomback, 1980). Birds recover their caches with great precision. Many field and laboratory studies have strongly supported the cash site memory hypothesis for the accurate cash recovery of nutcrackers (reviews in: Vander Wall, 1990; Kamil and Balda, 1990). What makes this story puzzling is that many caches are located relatively far from large landmarks. As far as small objects or ground markings, experimenters found no effect of using them by birds. It sounds naturally that nutcrackers ignore small landmarks because these markings can change dramatically with the changes of seasons. A major puzzle for a student of animal navigation is how a nutcracker can possibly achieve the precision required to relocate a small cache while digging with its small-diameter beak? (Kamil and Cheng, 2001).

Kamil and Jones (2000) have demonstrated that nutcrackers use spatial memory to retrieve their caches with accuracy and precision and that the birds can learn directional relationships between a goal and two landmarks. They tested birds in a rectangular room. In one of the settings of their experiments, birds were presented with two landmarks with varying distances between them. The birds always buried a seed at a third point relatively to two landmarks. The distance between the goal and the landmarks varied with the inter-landmark distance so as to maintain constant directional relationships: the seed was always buried north-west of one landmark and south-west of the other. The birds thus demonstrated an ability to use directional relationships. Another series of experiments revealed the evidence that the direction from the goal to a landmark is a more potent cue than the distance between them. Two groups of birds were trained to find the third point of the triangle. The goal location was defined by two landmarks whose inter-landmark distance varied from trial to trial. For one group, the third point of the triangle was defined by bearings. The goal was always buried at the intersection of two fixed bearings whose value was constant (requiring the goal-landmark distance to vary). For the second group, the goal was always buried at the same distance from each landmark (requiring the goal-landmark bearings to vary). The constant-bearing group learned to solve this problem much more rapidly than the constant-distant group. Under these conditions, bearings provided a more useful cue to location than the distance did.

Basing on these and many other experimental results and on computer simulations Kamil and Cheng (2001) have come to the multi-bearings hypothesis of way-finding in nutcrackers. These birds use the metric relationships between a goal and multiple landmarks. They prefer to bind the direction up with the distant landmarks rather than with the nearest one. The researchers propose that nutcrackers use a set of bearings, each a measure of the direction from the goal to a different landmark, when searching for that goal. Increasing the number of landmarks used – within certain limits - results in increasingly precise searching.

Once more impressive example of essential function of spatial representation concerns meerkats, cooperatively breeding mongooses (see Part X for details) that live under high predation pressure, and thus need to know what from more than 1000 boltholes on their territory is closest to find shelter from predators quickly. Manser and Bell (2004) used observations and manipulation experiments to investigate how meerkats succeed to find shelter immediately when an alarm call is given. Experimenters played back alarm calls to foraging mongooses and dug new boltholes and covered existing ones to see whether location or other cues were used. Meerkats almost always ran to the bolthole closest to them. This was not done by a simple rule of running back to a bolthole they had just passed, nor by escaping in random direction and finding a bolthole by chance. Mongooses nearly always ignored the boltholes that researchers dug but ran to those they had covered up. Adult animals seemed to have an accurate knowledge of the distance and direction to the closest shelter in relation to their own position in their territory at any time (Fig. IV-1).Although meerkats sometimes failed to find shelter, they still appeared to know a large proportion of the boltholes available in their territory, which still leaves several hundred or more locations to remember. This ability to memorise many locations might, as the authors give this, be comparable to the skills of food-storing birds.

12. TO WHAT DEGREE MAPPING IS COGNITIVE IN ANIMALS?