Honeybee vision and floral displays from detection to closeup recognition

In a social insect such as the honeybee, the survival of the colony depends on the success of its foragers. The bee optimizes its foraging success by returning to flowers of the species at which it has previously found food. This so-called flower constancy (see Chittka et al. 1999 for references) is based on the bee's capacity to learn and memorize specific flower signals (Menzel et al. 1993; Menzel & Müller 1996; Menzel 1999 and this volume) and to discriminate among different species by their different signals.

A bee returning to the feeding site in search of a flower, be it natural or artificial, must first detect the target from a distance. Once the flower has been detected, the bee will approach it up to a distance at which it is able to recognize whether or not the flower is similar to that stored in memory. Among the different sensory cues used, visual cues are of fundamental importance. In the rich market of coexisting and competing flower species, flower colors, shapes, and patterns are the visual cues that allow bees to recognize and discriminate profitable species.

Here we review studies concerned with the bee's use of visual signals for detecting and recognizing food sources. In the first part of the chapter, we examine the role of the bee's color vision in these tasks. In the second part, we look at the role of several spatial parameters contained in achromatic (black-and-white) stimuli. In both cases, we ask whether or not the visual cues that serve the bee for detecting a target at some distance are identical with those that serve for recognizing it at a closer range. We further ask whether the bee's use of particular visual cues in particular situations can be modified through experience. Flexibility of behavior based on learning may indicate some cognitive capacities on the part of the animal.

Fig. 4.1. The spectral sensitivity curves of the three photoreceptor types of the honeybee, Apis mellifera: S (UV receptor, solid line), M (blue receptor, dashed line), and L (green receptor, dotted line).

Detection and recognition of colored targets

The bee's color vision system

Behavioral (von Frisch 1914; Daumer 1956; Menzel 1967; Helversen 1972) and electrophysiological studies (Autrum & Zwehl 1964; Menzel & Blakers 1976; Peitsch et al. 1992) have shown that the honeybee possesses trichromatic color vision, with three spectral types of photoreceptors. Their sensitivity peaks are at 344 nm in the short-wave (ultra violet) region of the spectrum (S receptor), 436 nm in the middle-wave (blue) region (M receptor), and 544 nm in the long-wave (green) region of the spectrum (L receptor), respectively (Fig. 4.1).

An individual receptor can determine the amount of light it absorbs, but not the spectral composition of that light. Because the eye has three spectral types of photoreceptors, however, two juxtaposed color areas (a target and its background) will produce against each other three types of contrast: (1) Color contrast. Color is encoded in the comparison among the inputs arriving from the three spectral types of receptor, i.e., the three receptor types interact at a higher neural level. Several models of these interactions have been proposed, but are not discussed in detail here (for reviews, see Menzel & Backhaus 1991; Vorobyev & Brandt 1997). The output of the interactions is the perceptual color distance (i.e., the animal-subjective color difference) between stimuli. In the studies reviewed here, the calculations of color contrast are based on the neural interactions postulated in the color-opponent model proposed by Backhaus (1991) (see Giurfa et al. 1997 for formulas). (2) Receptor-specific contrasts, defined as the difference in excitations produced by two stimuli in each of the three receptor types, S (UV), M (blue), and L (green). We thus differentiate among "UV contrast," "blue contrast," and "green contrast". Because only one receptor type is involved in each case, receptor-specific contrasts are achromatic, i.e., they do not give rise to the perception of a color. (3) Intensity contrast, defined as the difference between the two stimuli in the sum of excitations that they evoke in all three receptor types. Summing the excitations prevents the comparison among the individual inputs. Therefore, intensity contrast is achromatic. In color discrimination tasks, bees make no use of intensity differences (Daumer 1956; Helversen 1972; Backhaus 1991; Menzel & Backhaus 1991).

The detection phase: the role of chromatic and achromatic cues and the minimum detectable angle

Traditionally, flower colors have been considered to constitute long-distance signals that enable detection of the flower from afar (von Frisch 1967; Kevan & Baker 1983; Chittka & Menzel 1992). However, the distance at which a flower is first detected cannot measure the detectability of its color, because that distance depends on the size of the flower. Therefore, the relevant measure of detectability of a distant target is the visual angle that it subtends on the insect's eye. This angle depends on both distance and size.

To find the minimum visual angle for detection of a colored target, bees were trained to collect a food reward at a colored disc presented against an achromatic (gray) background (Giurfa et al. 1996b). The experimental set-up consisted of a Y-shaped, dual-arm apparatus (Fig. 4.2). In one of the arms, termed positive, a colored disc was presented against the gray background. A bee entering that arm received a reward of sucrose solution when it arrived at the colored disc. The alternative arm, termed negative, displayed the gray background only, with no reward. Thus, bees

Decision Chamber

Fig. 4.2. View of the Y-maze apparatus. The decision between the arm with the training disc and that with the background alone could be made by the bee only after it has entered the decision chamber, from which the back walls of both arms could be viewed simultaneously. The decision point was defined as the center of the decision chamber. A choice of either arm was scored when the bee crossed the imaginary line between the decision chamber and one of the arms. The visual angle subtended by the rewarding disc at the bee's eye was calculated from the distance of the disc from the decision point, and the diameter of the disc. For more details, see Giurfa etal. (1996b).

Decision Chamber

Fig. 4.2. View of the Y-maze apparatus. The decision between the arm with the training disc and that with the background alone could be made by the bee only after it has entered the decision chamber, from which the back walls of both arms could be viewed simultaneously. The decision point was defined as the center of the decision chamber. A choice of either arm was scored when the bee crossed the imaginary line between the decision chamber and one of the arms. The visual angle subtended by the rewarding disc at the bee's eye was calculated from the distance of the disc from the decision point, and the diameter of the disc. For more details, see Giurfa etal. (1996b).

were trained to distinguish between the presence and the absence of a colored spot.

Six different colored discs were used in six different experiments. The visual angle subtended by the disc was varied by changing the disc diameter or the distance between the back wall and the decision point. Because changing the disc's distance had the same effect as changing its diameter (see also Lehrer & Bischof 1995), the results were pooled. The minimum visual angle at which a given stimulus is detectable was defined as the angle at which the bees chose the positive arm with a frequency of 60%.

The six stimuli presented comparable amounts of color contrast against the gray background. The results, however, differed among the six experiments (Fig. 4.3). Four of the stimuli rendered a threshold of detectability at a minimum visual angle of 5°. These stimuli produced, in addition to color contrast, (achromatic) green contrast against the background. The other two stimuli were not detected unless they subtended a visual angle of at least 15°. In these stimuli, green contrast was absent.

Was this article helpful?

0 0

Post a comment