Ways of navigation in animals

Both long-distance and short-distance travellers use multiple cues and variety of ways for navigation. Here we will consider briefly the main ways of navigation in animals and illustrate them by several examples based on experimental examinations.

Olfactory navigation. The use of smell to find a goal is widely spread in different taxa, both of invertebrates and vertebrates. One particular case of orienting by smell is the use of pheromones. The term pheromone was introduced by Karlson and Lü scher (1959) to refer to the chemicals that are used for communication, through the sense of smell, among individuals of a given species. Some animals possess scent glands that release pheromones. Pheromone trails can guide animals to food and to their nest sites, to detect boundaries of guarded territories of their neighbours, and to seek sex mates.

One of the most impressive examples of olfactory navigation concerns birds who gain information from atmospheric trace gases. Birds for a long time have been known as weak users of olfactory cues. Human beings could be considered the members of the same club. This choice is determined by that birds are the most prominent and best-known migrants and they possess one of the most intricate navigation systems which are probably based on using multiple sources. The use of olfactory cues may be built into an integrative “cognitive map” in some bird species.

The question of whether birds can derive information on their current position relative to their home loft from airborne substances perceives by the sense of smell have relatively short history of investigation. More than three decades ago, 54 km west of Florence in Italy, 10 homing pigeons were released whose olfactory nerves were sectioned (Papi et al., 1972), and these 10 birds were pioneers marking a breakthrough in avian navigation research (for a review, see: Wallraff, 2004). Less traumatic methods of olfactory deprivation have been then elaborated such as anaesthesia of olfactory epithelia as well as occlusion of nostrils. Further studies based on olfactory deprivation experiments have led to a suggestion that olfaction-dependent/independent homing expresses differently in different species and this may be reflects difference in their “system profiles” of navigation. Fiaschi et al. (1974) applied olfactory deprivation to common swifts Apus apus and released them 47-66 km from home. Of 20 control birds 15 returned, but only three of 23 experimental birds; these three birds had lost their nose plugs. Anosmic European starlings, Sturnus vulgaris, were displaced over varying distances (Wallraff et al., 1994). Up to 60 km, anosmia increased the time required for homing, but did not affect the percentage of birds that finally homed, whereas over 120 and 240 km return rates were drastically reduced. This increased effect of anosmia with increasing distance was similar to that obtained earlier in pigeons, but the starlings were successful in olfaction-independent homing within a larger radius around home. In principle, the range of olfactory navigation is not unlimited. It may extend over distances of some 300-500 km from home and seems dependent on particular geographical, mainly orographical, conditions.

Researchers have carried out plenty of “air filtration” experiments with pigeons. They manipulated with displaced pigeons and suggested them portions of air to filter. In sum, these findings have shown that a pigeon decides in which direction to fly not primarily on the basis of olfactory inputs receive at the moment of departure, but on the basis of inputs gathered a longer period beforehand. Moreover, these experiments suggest that the pigeons develop an “olfactory map” by associating current olfactory sensations with current wind directions. In a model aimed to describe the development of the pigeon’s navigation system, the wind would have to be inserted as an essential link between compasses and grid map (Wallraff, 2004). We will return to these elements of pigeon’s “integral map” in the next issues.

Path integration (dead reckoning). The term “dead reckoning” has been borrowed from humans’ navigation terminology that was historically used by sailors to navigate across featureless open sea. More frequently it has come to be called path integration (PI) to reflect the assumption that the process takes place by the addition of successive small increments of movement onto a continually updated representation of direction and distance from the starting point. Darwin (1873) and Murphy (1873) hypothesised that path integration is based on the integration of interior signals. PI appears to operate in a great number of species with a fixed home base, during the exploration of a new environment or in commuting between home and familiar resource sites. PI functions automatically and constantly, whenever the agent moves in space. Thus, a central place forager may, for instance, interrupt its excursion back home at any place and at any moment of its journey (Etienne, Jeffery, 2004).

Integrating definitions from recent literature, dead reckoning (path integration) can be defined as the process of estimating your position by advancing a known position using course, speed, time and distance to be travelled. In other words figuring out where you will be at a certain time if you keep the speed, time and course.

There is general agreement among those who study animal navigation that dead reckoning is fundamental to the navigation of animals ranging from molluscs and insects to humans. This way of navigation is often considered rather simple. Cornell and Heth (2004) cite Herman Melville’s poetic novel about reverting to elemental level of navigation such as dead reckoning after having ship’s quadrant destroyed by Captain Ahab’s soul. At the same time, Cornell and Heth note that such a method of navigation without fixed references and landmarks has been especially intriguing to comparative and cognitive psychologists. As Pearce (2000) gives this, an obvious problem with navigating in this way is that once an error has entered into calculations as a result of faulty measurement, there is no means of detecting it and the navigator will have no indication of being lost.

Let us consider path integration in desert ants, as a good example. The main subject in question here is the Sahara desert ant Cataglyphis fortis. This is an object of more than 20 years detailed experimental analysis of R.Wehner’s group (see, for example: Wehner and Srinivasan, 1981; Wehner, Michel and Antonsen, 1996). This ant is a solitary forager that scavenges for other arthropods that have succumbed to the physical stress of their desert habitat and what is important in the present context, does so by relying upon visual rather than chemical cues. Its navigation courses can be recorded in full details because it walks rather than flies.

Dead reckoning in the ant and the honeybee is likely to be dependent on information acquired from the movements they do during their journey. How can insects accomplish continuously computing the vector pointing towards its nests? In vertebrates this way of navigation is said to be influenced also by changes that take place in the vestibular system. In rats, for example, lesions of the vestibular system have been shown to disrupt their capacity for dead reckoning (Wallace et al., 2002). It is known that arthropods successfully use path integration but it is still discussible how they could estimate the distances travelled in a given direction.

What is clear now is that a common formula does not exist. For example, energy expenditure as a cue for distance estimation was proposed by Heran and Wanke (1952) on the basis of experiments with honeybees that were trained to forage on steep slopes. This hypothesis has been successfully revisited for bees (Goller and Esch, 1990) but not for ants as an additional load up to four times the body weight did not affect the measurement of walking distance in fortis - ant (Schä fer and Wehner, 1993). The second way to measure the distance traveled is to do this by monitoring locomotor activity (the so called idiothetic orientation). Experiments with spiders such as Cupiennius salei showed that they are able to evaluate travel distance on the basic of idiothetic cues (Seyfarth et al., 1982). In experiments with the fortis ant Ronacher and Wehner (1995) tested the third hypothesis namely that ants are able to use self-induced optic flow components to estimate travel distances.

The animals were trained and tested in transparent channels. One end of the training channel was situated near the ants’ nest, and the ants were induced to enter the channel by a fence around the nest. The channel was equipped with stationary or moving black-and- white gratings of random dots patterns presented underneath a transparent walking platform. The patterns could be moved at different velocities in the same or opposite direction relative to the direction in which the animal walked. Experimental manipulations on the optic flow influenced the ant’s homing distances. The use of different pattern wavelengths showed that distance estimation depends on the speed of self-induced image motion rather than on the contrast frequency. Experiments in which the ants walked on a featureless floor, or in which they wore eye covers showed that they are able to use other cues for assessing travel distance. Hence, even though optic flow cues are not the only cues used by the ants, the experiments show that ants are obviously able to exploit such cues for estimation of travel distance.

Piloting with landmarks and use of geometric relation. Piloting refers to the act of setting a course to a goal on the basis of landmarks that are in a known relation to a goal. In many experiments many species have demonstrated their abilities to make use of landmarks to determine both the direction and the distance that they are going to travel.

The simplest form of piloting would be to navigate towards a feature that was located immediately by the goal - a beacon. But very often landmarks are not conveniently situated by a goal and their use then becomes more complicated.

Many elegant experiments concerning piloting were carried out on insects. This tradition was started by Tinbergen. By means of simple and elegant tests he asked a digger wasp Ammophyla how it finds its way to an “invisible” nest (simply a hole in sand). A mother-wasp digs nest barrow in which she lays an egg. It covers the barrow entrance with pebbles, making it all but invisible, and then it flies off in search of insect prey. Tinbergen and Kruyt (1938) arranged circles if sticks and pine cones around the nest while a wasp was digging it then displaced the circle while she was gone. The returning wasp landed in the centre of the displaced circle. She could not find her nest, even though it was a meter away. This experiment demonstrated that the wasp located the entrance by reference to nearby landmarks, not by the sight, or smell, etc. of the nest itself. It is natural that when two or more landmarks are present, they create a geometric shape. Two landmarks create a line, three a triangle, and so on. Researchers from Tinbergen’s group were first to demonstrate that animals are able to detect this geometric information and to use it to define the location of a goal.

As a variant of experiments with the digger wasp, Van Beusekom (1948) also constructed a circle of pine cones around the hole while the wasp was in it, but as soon as it flew away, he constructed either a square or an ellipse of cones around the circle. Upon returning, the wasp went to the circle rather than to the square, but it selected the circle and the ellipse equally often. This discovery indicates that during their initial flight around the hole the wasp remembers the shape created by the cones. Evidently the representation of this shape is sufficiently precise to permit a distinction between a circle and a square but not between a circle and an ellipse. Luckily nobody punished the wasp with electric shock for its incapability of detecting such a subtle difference between geometric figures so it did not get traumatic neurosis like Pavlov’s dog which also failed to make this distinction (see Chapter 6).

A series of experiments provides a clear demonstration that rats, pigeons, humming birds, hamsters and even newly hatched chicks are sensitive to geometric relations contained in the shape of their test environment and can make use of geometric information in piloting processes when they search for a hidden food (Cheng, 1986; Vallortigara et al., 1990; Brown, 1994; Spetch, 1995; for a review see Cheng, 2005). For instance, Cook and Tauro (1999) investigated how the geometric relation between object/landmarks and goals influenced spatial choice behaviour in rats. Animals searched for hidden food in an object-filled circulate arena containing 24 small poles. The results of tests with removed and rearranged landmarks that were configured in a square suggested that close proximity of objects to goals encourage their use as beacons, while greater distance of objects from goals encourage their use as landmarks.

Magnetic fields and compass. A number of studies have experimentally demonstrated that magnetic properties of the Earth provide a pervasive source of information for travelling both long and short journeys. In support of this proposal, pigeon homing is less accurate when a magnetic storm takes place (Gould, 1982a). It was assumed that birds were able to determine their global position by detecting features of the earth’s magnetic field, such as field intensity and inclination, for reading a bi-coordinate magnetic map based on these features. Besides, the birds apparently use the sky and its celestial bodies as a source of information for navigation. To illustrate that the Sun is an important element of homing the results of experiments with “clock-shifted” pigeons can be cited. Pigeons with a phase-shifted circadian clock, released from home in any direction, deviate at departure from unmanipulated control birds by an angle roughly corresponding to the angle between the observed sun azimuth and that expected according to their shifted time scale (Schmidt-Koenig, 1960; 1979). Kramer (1953) suggested that birds should have some equivalent of a map as well as of a compass. Surely these are conveniently brief metaphors which do not precisely describe the system. The brief term “map” refers to position-indicating factors while “compass” refers to directional references. We must be aware that these terms represent the anthropomorphic. A bird hardly has a map at all and probably does not needs any idea about wide-ranging spatial structures. In other words, it does not necessarily imply that the bird has a representation of a coordinate system or any other wide-ranging spatial configuration in its brain. It merely means that we are dealing with environmental factors whose two-dimentional spatial order involves information on distant positions relative to a bird’s home site. How a bird makes use of such information is not immediately implied (Wallraff, 2004). In particular, the sun-arc hypothesis suggested that birds could determine their position relative to home by comparing the actual movement of the sun along its arc with the remembered one at home. This suggestion rather reflects the paradigm of early research in animal navigation based on representing birds as “true astronavigators”. Since experimental investigations were carried out in 70-s, it has been cleared that birds are not astronomers in the sense that they can not use the sun for taking positional fixes anywhere on the surface of our globe; nor can birds use the stars for fixing their position at night (see Wehner, 1998, for a review). Birds and other animals have not adopted a heliocentric or a general geocentric view of the skylight world surrounding them. Instead, they solve their navigation problems in more immediate ways in which signals from magnetic and celestial cues are included differently for different species and at different stages of their lifetime.