Odor sequence learning in honeybees: insights from olfactory classical conditioning paradigm

The ability to extract regularities from the sensory input is important for adaptation to a complex environment and to support behavioral flexibility. Honeybees have sophisticated cognitive and learning capacities, including categorization of stimuli on the basis of perceptual (e.g., symmetry) and abstract (e.g., sameness/difference) concepts. Here we investigated whether bees could extrapolate the temporal regularity of a sequence. By exploiting the olfactory conditioning of the proboscis extension response (PER), we investigated whether honeybees could learn and generalize a particular sequence structure composed of different odors presented in a specific temporal order. Different conditioning paradigms (i.e., absolute, differential, generalization) were employed. Bees used different strategies, such as an early tendency to encode the single odor properties, instead of learning the entire sequence pattern. The use of a generalization paradigm potentially uncovered a spontaneous tendency to transfer over similar structures one hour after the training irrespective of the single-element properties, though this finding is not yet conclusive. These results shed light on the strategies used by bees to solve an odor abstraction task, highlighting the crucial role of the type of conditioning to let them emerge.

从感觉输入中提取规律性的能力，对于适应复杂环境并支撑行为灵活性具有重要意义。蜜蜂拥有复杂的认知与学习能力，包括基于知觉（perceptual）层面（例如对称性）与抽象（abstract）层面（例如相同/差异）的概念对刺激物进行分类。本研究旨在探究蜜蜂是否能够外推序列的时间规律性。我们借助伸喙反射（proboscis extension response, PER）的嗅觉条件化方法，探究蜜蜂能否学习并泛化由特定时间顺序呈现的不同气味所构成的特定序列结构。本研究采用了三类条件化范式，即绝对条件化、辨别条件化与泛化条件化。蜜蜂会采取不同的策略，例如在早期阶段倾向于编码单一气味的属性，而非直接学习完整的序列模式。尽管该发现尚未形成定论，但泛化条件化范式的应用，在训练一小时后潜在揭示了一种自发倾向：蜜蜂会将习得的规则迁移至相似的序列结构中，且该迁移过程不受单一气味元素属性的影响。本研究结果阐明了蜜蜂在解决气味抽象任务时所采用的策略，并凸显了条件化范式类型在促使这些策略得以显现过程中的关键作用。