ALBERT

Notes: Abstract Experiments by Fabre (1915), Thorpe (1950), Chmurzynski (1964), and most recently Gould (1986) suggest that insects have “maps” of their terrain which enable them to find their way directly to a goal when they are displaced several hundred metres from it. This paper discusses what might constitute an insect's map in terms of a two-part computational model. The first part describes how an insect reaches a goal when the insect is sufficiently close that it can see some of the landmarks which are visible from the goal. The second part considers the problem of navigating when there is no similarity between the view from the release-site and the view from the goal. We start from a model designed to explain how a bee might return to a goal using a two-dimensional “snapshot” of the landscape seen from the goal (Collett and Cartwright 1983). To guide its return, the model bee continuously compares its snapshot with its current retinal image and moves so as to reduce the discrepancy between the two. Bees can only be guided in the right direction by the difference between current retinal image and snapshot when there is some resemblance between the two. In a realistically cluttered world, snapshot and retinal image become very dis-similar only a short distance from the goal. To increase the distance from which a model bee can return, the bee takes two snapshots at the goal. The first snapshot excludes landmarks near to the goal and the second snapshot includes them. With close landmarks filtered from both snapshot and retinal image, the match between the two deteriorates gradually as the bee moves away from the goal. A model bee using a filtered snapshot and image finds its way back to the neighbourhood of the goal from a relatively long distance (Fig. 2). The bee then switches to the second snapshot and is guided to the precise spot by its memory of the close landmarks. For longer range guidance, the model bee is equipped with an album of snapshots, each taken at a different location within the terrain. Linked to each snapshot is a vector encoding the distance and direction from the place where the snapshot was taken to the hive. When the bee is displaced to a new position, it selects the snapshot which best matches its current image and follows the associated home-vector back to the hive (Fig. 3). Such a hive-centred map can also be used to devise novel routes to places other than the hive. For instance, a bee can reach a foraging site from anywhere in its terrain by adding the home-vector recalled at the starting position to a vector specifying the distance and direction of the foraging site from the hive. The sum of these two vectors defines a direct trajectory to the foraging site.

Notes: Summary 1. The experiments described here were undertaken to discover how bees use nearby landmarks to guide their way to a food source. Two major questions are raised. First, what do bees learn about the spatial layout of landmarks and food source? Secondly, how might this information help them reach their destination? 2. Single, marked bees were trained to collect sugar solution from a small and inconspicuous reservoir in a room in which extraneous visual cues had been reduced to a minimum. The position of the reservoir was defined by an array of one or more matt black landmarks. After bees had been trained, their flight path was recorded on videotape when the landmarks were present, but the food source absent. During such tests bees spent most of their time searching where the food source should have been. 3. Thus, if bees were trained to a reservoir whose position was specified by a single cylindrical landmark and tested with the same landmark, they searched at the expected site of the reservoir. However, when the size of the landmark was changed between training and testing, the area in which bees searched was displaced to one where the landmark appeared roughly the same size as the training landmark when viewed from the reservoir. These experiments suggest that bees learn no more than the apparent size and bearing of the landmark as seen from the food source, and that to return there they move to a position where their retinal image matches their remembered image of the landmark. 4. Experiments with more complex arrays of landmarks support the same hypothesis. A simple rule predicts a bee's search area when it is trained to a food source defined by the position of three landmarks and tested either with the same array, or with landmarks of different sizes, or with landmarks placed at different distances from the reservoir. The bee then always searches where the compass bearings of the landmarks on its retina were the same as they had been when it was stationed at the food source. 5. Tests with bees trained to either one or three landmarks suggest that the bearings of landmarks on the retina are learnt with respect to external compass bearings. Thus, a single, cylindrical landmark does not define direction. Nonetheless, bees searched in one location and not in a circle centred on the landmark. Bees trained to three landmarks only learnt the site of the reservoir if the array was kept in a constant orientation during training. 6. Computer models were devised to discover how bees might use a remembered image of the landmark array to direct their flight path to their destination. The models simulated a situation in which a bee takes a 2-dimensional snapshot of its surroundings from the position it wishes to retrieve and continuously compares this with its current retinal image. It then uses the difference between the two to guide its way. Different models of increasing complexity were explored until one was found which closely mimicked the bee's behaviour.

Notes: Summary 1. The aim of this study is to understand what a rodent (Meriones unguiculatus) learns about the geometrical relations between a goal and nearby visual landmarks and how it uses this information to reach a goal. Gerbils were trained to find sunflower seeds on the floor of a light-tight, black painted room illuminated by a single light bulb hung from the ceiling. The position of the seed on the floor was specified by an array of one or more landmarks. Once training was complete, we recorded where the gerbils searched when landmarks were present but the seed was absent. In such tests, gerbils were confronted either with the array of landmarks to which they were accustomed or with a transformation of this array. 2. Animals searched in the appropriate spot when trained to find seeds placed in a constant direction and at a constant distance from a single cylindrical landmark (Fig. 1). Since gerbils look in one spot and not in a circle centred on the landmark, the direction between landmark and goal must be supplied by cues external to the landmark array. Distance, on the other hand, must be measured with respect to the landmark. Tests in which the size of the landmark was altered from that used in training suggest that distance is not learned solely in terms of the apparent size of the landmark as seen from the goal (Fig. 3). 3. Gerbils can still reach a goal defined by an array of landmarks when the room light is extinguished during their approach (Figs. 4, 5). This ability implies that they have already planned a trajectory to the goal before the room is darkened. In order to compute such a trajectory, their internal representation of landmarks and goal needs to contain information about the distances and bearings between landmarks and goal. 4. For planning trajectories, each landmark of an array can be used separately from the others (Fig. 7). Gerbils trained to a goal specified by an array of several landmarks were tested with one or more of the landmarks removed or with the array expanded. They then searched as though they had computed an independent trajectory for each landmark. For instance, gerbils trained with an array of two landmarks were tested with the distance between two landmarks doubled. The animals then searched for seeds in two positions, which were at the correct distance and in the right direction from each landmark. 5. If an internal representation of an array of landmarks is to be used to plan a trajectory, landmarks seen on the ground must be matched to those held in memory. One way in which gerbils do this is by learning properties of individual landmarks, such as their shape, size or surface markings (Figs. 10, 11, 13). For example, gerbils were able to locate seeds defined by a single relevant landmark while ignoring an irrelevant landmark with different features which was placed randomly with respect to the goal. 6. Several experiments (Figs. 4, 12, 13, 14) suggested that, although landmarks may be used independently for computing trajectories, the process of matching landmarks to the gerbil's representation requires a knowledge of the distances and directionsbetween landmarks. 7. We conclude that a gerbil's representation of its environment is complete in that it stores explicitly or can compute from what it has stored the geometric arrangement of landmarks and goal. We discuss the possibility that its spatial memories consist of a set of vectors describing the distance and direction from the goal to each landmark (Fig. 18) and consider the advantages and disadvantages of such a goal-centred memory.

Notes: [Auszug] Bees trained to forage at a place specified by landmarks do not construct a cartesian map of the arrangement of landmarks and food source. Instead they store something like a two-dimensional snapshot of their surroundings taken from the food source. To return there, bees move so as to reduce ...