Back

Bumblebee visual learning: simple solutions for complex stimuli

Guiraud, M.-G.; Gallo, V.; Quinsal-Keel, E.; MaBouDi, H.

2024-03-17 animal behavior and cognition
10.1101/2024.03.16.585132 bioRxiv
Show abstract

Natural visual stimuli are typically complex. This presents animals with the challenge of learning the most informative aspects of these stimuli while not being confused by variable elements. How animals might do this remains unclear. Here, we tested bumblebees ability to learn multicomponent visual stimuli composed of a simple constant bar element and a grating element that was consistent in orientation but varied in width and number of gratings. Bees rapidly and successfully learned these compound stimuli. Tests revealed learning of the single bar element was more robust than learning of the grating element. Our study highlights how even small-brained invertebrates can rapidly learn multicomponent stimuli and prioritise the most consistent elements within them. We discuss how the learning phenomena of generalisation and overshadowing may be sufficient to explain these findings, and caution that complex cognitive concepts are not necessary to explain the learning of complex stimuli. HighlightsO_LIBumblebees are highly efficient in prioritising the most consistent elements in multicomponent visual stimuli. C_LIO_LIBees trained on horizontal and vertical cues exhibit differences in how they memorise visual cues. C_LIO_LITwo phenomena can explain how bees preferentially select, memorise and use visual cues in this experiment: generalisation and overshadowing. C_LIO_LIBumblebees as generalist foragers are well-suited to study visual cognition. C_LI

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.