Dim-light colour vision in the facultatively nocturnal Asian giant honeybee, Apis dorsata.

We discovered nocturnal colour vision in the Asian giant honeybee Apis dorsata-a facultatively nocturnal species-at mesopic light intensities, down to half-moon light levels (approx. 10-2 cd m-2). The visual threshold of nocturnality aligns with their reported nocturnal activity down to the same light levels. Nocturnal colour vision in A. dorsata is interesting because, despite being primarily diurnal, its colour vision capabilities extend into dim light, while the 'model' European honeybee Apis mellifera is reported to be colour-blind at twilight. By employing behavioural experiments with naturally nesting A. dorsata colonies, we show discrimination of the trained colour from other stimuli during the day, and significantly, even at night. Nocturnal colour vision in bees has so far only been reported in the obligately nocturnal carpenter bee Xylocopa tranquebarica. The discovery of colour vision in these two bee species, despite differences in the extent of their nocturnality and the limitations of their apposition compound eye optics, opens avenues for future studies on visual adaptations for dim-light colour vision, their role in pollination of flowers at night, and the effect of light pollution on nocturnal activity in A. dorsata, a ubiquitous pollinator in natural, agricultural and urban habitats in the Asian tropics and sub-tropics.

Spatial resolution and sensitivity of the eyes of the stingless bee, Tetragonula iridipennis.

Stingless bees are important pollinators in the tropics. The tremendous variation in body size makes them an excellent group to study how miniaturization affects vision and visual behaviours. Using direct measurements and micro-CT, we reconstructed the eye structure, estimated anatomical spatial resolution and optical sensitivity of the stingless bee Tetragonula iridipennis. T. iridipennis is similar in size to the Australian stingless bee Tetragonula carbonaria and is smaller than honeybees. It has correspondingly small eyes (area = 0.56 mm2), few ommatidia (2451 ± 127), large inter-facet (3.0 ± 0.6°) and acceptance angles (2.8°). Theoretical estimates suggest that T. iridipennis has poorer spatial resolution (0.17 cycles degree-1) than honeybees, bumblebees, and T. carbonaria. Its optical sensitivity (0.08 µm2 sr), though higher than expected, is within the range of diurnal bees. This may provide them with greater contrast sensitivity, which is likely more relevant than the absolute sensitivity in this diurnal bee. Behaviourally determined detection thresholds for single targets using y-maze experiments were 11.5° for targets that provide chromatic contrast alone and 9.1° for targets providing chromatic and achromatic contrast. Further studies into microhabitat preferences and behaviour are required to understand how miniaturization influences its visual ecology.

Visual detection thresholds in the Asian honeybee, Apis cerana.

To understand how insect pollinators find flowers against complex backgrounds in diverse natural habitats, it is required to accurately estimate the thresholds for target detection. Detection thresholds for single targets vary between bee species and have been estimated in the Western honeybee, a species of bumblebee and in a stingless bee species. We estimated the angular range of detection for coloured targets in the Asian honeybee Apis cerana. Using a Y-maze experimental set up, we show that targets that provided both chromatic and green receptor contrast were detected at a minimum visual angle of 7.7°, while targets with only chromatic contrast were detected at a minimum angle of 13.2°. Our results thus provide a robust foundation for future studies on the visual ecology of bees in a comparative interspecific framework.

Complex multi-modal sensory integration and context specificity in colour preferences of a pierid butterfly.

Innate colour preferences in insects were long considered to be a non-flexible representation of a floral 'search image' guiding them to flowers during initial foraging trips. However, these colour preferences have recently been shown to be modulated by multi-sensory integration of information. Using experiments on the butterfly Catopsilia pomona (common emigrant), we demonstrate that cross-modal integration of information not only affects colour preferences but also colour learning, and in a sex-specific manner. We show that spontaneous colour preference in this species is sexually dimorphic, with males preferring both blue and yellow while females prefer yellow. With minimal training (two training sessions), both males and females learned to associate blue with reward, but females did not learn green. This suggests that the aversion to green, in the context of foraging, is stronger in females than in males, probably because green is used as a cue to find oviposition sites in butterflies. However, females learned green after extensive training (five training sessions). Intriguingly, when a floral odour was present along with green during training, female colour preference during the subsequent choice tests resembled their innate preference (preference for yellow). Our results show that multi-sensory integration of information can influence preference, sensory bias, learning and memory in butterflies, thus modulating their behaviour in a context-specific manner.

Foraging strategies and physiological adaptations in large carpenter bees.

Large carpenter bees are charismatic and ubiquitous flower visitors in the tropics and sub-tropics. Unlike honeybees and bumblebees that have been popular subjects of extensive studies on their neuroethology, behaviour and ecology, carpenter bees have received little attention. This review integrates what is known about their foraging behaviour as well as sensory, physiological and cognitive adaptations and is motivated by their versatility as flower visitors and pollinators. This is evident from their extremely generalist foraging and adeptness at handling diverse flower types as legitimate pollinators and as illegitimate nectar robbers. They purportedly use traplining to forage between isolated patches and are long-distance flyers over several kilometres suggesting well-developed spatial learning, route memory and navigational capabilities. They have a broad range of temperature tolerance and thermoregulatory capabilities which are likely employed in their forays into crepuscular and nocturnal time periods. Such temporal extensions into dim-light periods invoke a suite of visual adaptations in their apposition optics. Thus, we propose that carpenter bees are an excellent though understudied group for exploring the complex nature of plant-pollinator mutualisms from ecological and mechanistic perspectives.

A comparative analysis of colour preferences in temperate and tropical social bees.

The spontaneous occurrence of colour preferences without learning has been demonstrated in several insect species; however, the underlying mechanisms are still not understood. Here, we use a comparative approach to investigate spontaneous and learned colour preferences in foraging bees of two tropical and one temperate species. We hypothesised that tropical bees utilise different sets of plants and therefore might differ in their spontaneous colour preferences. We tested colour-naive bees and foragers from colonies that had been enclosed in large flight cages for a long time. Bees were shortly trained with triplets of neutral, UV-grey stimuli placed randomly at eight locations on a black training disk to induce foraging motivation. During unrewarded tests, the bees' responses to eight colours were video-recorded. Bees explored all colours and displayed an overall preference for colours dominated by long or short wavelengths, rather than a single colour stimulus. Naive Apis cerana and Bombus terrestris showed similar choices. Both inspected long-wavelength stimuli more than short-wavelength stimuli, whilst responses of the tropical stingless bee Tetragonula iridipennis differed, suggesting that resource partitioning could be a determinant of spontaneous colour preferences. Reward on an unsaturated yellow colour shifted the bees' preference curves as predicted, which is in line with previous findings that brief colour experience overrides the expression of spontaneous preferences. We conclude that rather than determining foraging behaviour in inflexible ways, spontaneous colour preferences vary depending on experimental settings and reflect potential biases in mechanisms of learning and decision-making in pollinating insects.

Motion cues improve the performance of harnessed bees in a colour learning task.

The proboscis extension conditioning (PER) is a successful behavioural paradigm for studying sensory and learning mechanisms in bees. Whilst mainly used with olfactory and tactile stimuli, more recently reliable PER conditioning has been achieved with visual stimuli such as colours and looming stripes. However, the results reported in different studies vary quite strongly, and it remains controversially discussed how to best condition visual PER. It is particularly striking that visual PER leads to more limited performance as compared to visual conditioning of free-flying bees. It could be that visual PER learning is affected by the lack of movement and that the presence of visual motion cues could compensate for it. We tested whether bees would show differences in learning performances when conditioned either with a colour and motion stimulus in combination or with colour alone. Colour acquisition was improved in the presence of the motion stimulus. The result is consistent with the idea that visual learning might be tightly linked to movement in bees, given that they use vision predominantly during flight. Our results further confirm recent findings that successful visual PER conditioning in bees is achievable without obligatorily removing the antennae.