Shape vision in bees: innate preference for flower-like patterns

Abstract

The bees’ spontaneous preferences toward various black-and-white patterns were studied using a multiple-choice test procedure. The patterns are presented on vertical planes, and the bees’ choices at a fixed distance from the patterns are recorded. To exclude a possible influence of the bees’ previous experience with natural flowers, the bees are trained to randomized checkerboard patterns prior to testing them with sets of other patterns. We find that, when the test patterns are of the same kind, but differ in their spatial frequencies, the bees prefer low over high frequencies. However, when the patterns differ in type, the bees express, regardless of spatial frequency, a positive preference for patterns containing radiating elements, and a negative preference for patterns containing circular elements or elements arranged at random. We find, in addition, that symmetrical patterns are more attractive than less symmetrical or non-symmetrical patterns. We propose that bees respond innately to some features of natural flowers, resulting in a spontaneous preference for radiating, as well as symmetrical patterns.

Getting to the start line: how bumblebees and honeybees are visually guided towards their first floral contact

“…In comparisons of various spatial frequencies for each of several patterns (horizontal gratings, vertical gratings, radial patterns and concentric circles), however, a consistent preference for comparatively low frequencies was obtained, even though the most disrupted patterns were resolvable-i.e. the lines were not perceived as blurred together (Lehrer et al, 1995). Preferences for relatively high spatial frequency patterns have also been reported (Dafni et al, 1997;Plowright et al, 2011).…”

“…'sunburst' patterns: alternating black and white pie shaped segments, all pointing to the centre) over concentric patterns (i.e. 'bull's eye' patterns: alternating black and white circles within each other) has been well documented for honeybees (Lehrer et al, 1995) and bumblebees (Simonds and Plowright, 2004;Plowright et al, 2006;Séguin and Plowright, 2008). Selective landing on radial patterns by free flying bumblebees has also been reported (Orbán and Plowright, 2013).…”

“…data). To circumvent this problem, bees are sometimes trained to ostensibly neutral patterns such as black discs, white discs (Rodríguez et al, 2004) or checkerboards (Lehrer et al, 1995) and subsequently tested for their preferences of new unrewarded patterns. This practice may be innocuous when studying colour preferences: bees do not generalize their experience from pre-training with one colour to testing on others, as long as the colours seem very different to them (Gumbert, 2000).…”

“…A sample of undifferentiated 'bee-choices' (e.g. Lehrer et al, 1995) gives little guide as to whether the results might reflect the behaviour of only a few particularly active bees. Individual differences (e.g.…”

“…In nature, some plants make themselves detectable not by their particularly large floral structures, but by their inflorescences consisting of clusters of small flowers (Lehrer et al, 1995): spatial frequency (i.e. the ''busyness'') of the visual input may be a key variable.…”