Latty Tanya, Trueblood Jennifer S
School of Life and Environmental Sciences, The University of Sydney, Sydney, NSW, Australia.
Department of Psychology, Vanderbilt University, Nashville, TN, USA.
J Anim Ecol. 2020 Dec;89(12):2750-2762. doi: 10.1111/1365-2656.13347. Epub 2020 Oct 12.
Understanding why animals (including humans) choose one thing over another is one of the key questions underlying the fields of behavioural ecology, behavioural economics and psychology. Most traditional studies of food choice in animals focus on simple, single-attribute decision tasks. However, animals in the wild are often faced with multi-attribute choice tasks where options in the choice set vary across multiple dimensions. Multi-attribute decision-making is particularly relevant for flower-visiting insects faced with deciding between flowers that may differ in reward attributes such as sugar concentration, nectar volume and pollen composition as well as non-rewarding attributes such as colour, symmetry and odour. How do flower-visiting insects deal with complex multi-attribute decision tasks? Here we review and synthesise research on the decision strategies used by flower-visiting insects when making multi-attribute decisions. In particular, we review how different types of foraging frameworks (classic optimal foraging theory, nutritional ecology, heuristics) conceptualise multi-attribute choice and we discuss how phenomena such as innate preferences, flower constancy and context dependence influence our understanding of flower choice. We find that multi-attribute decision-making is a complex process that can be influenced by innate preferences, flower constancy, the composition of the choice set and economic reward value. We argue that to understand and predict flower choice in flower-visiting insects, we need to move beyond simplified choice sets towards a view of multi-attribute choice which integrates the role of non-rewarding attributes and which includes flower constancy, innate preferences and context dependence. We further caution that behavioural experiments need to consider the possibility of context dependence in the design and interpretation of preference experiments. We conclude with a discussion of outstanding questions for future research. We also present a conceptual framework that incorporates the multiple dimensions of choice behaviour.
理解动物（包括人类）为何选择此物而非彼物，是行为生态学、行为经济学和心理学领域的关键问题之一。大多数关于动物食物选择的传统研究聚焦于简单的单属性决策任务。然而，野生动物常常面临多属性选择任务，其中选择集中的选项在多个维度上存在差异。多属性决策对于访花昆虫尤为重要，它们需要在花朵之间做出抉择，这些花朵在奖励属性（如糖分浓度、花蜜量和花粉成分）以及非奖励属性（如颜色、对称性和气味）方面可能存在差异。访花昆虫如何应对复杂的多属性决策任务？在此，我们回顾并综合了关于访花昆虫在进行多属性决策时所采用的决策策略的研究。特别是，我们回顾了不同类型的觅食框架（经典最优觅食理论、营养生态学、启发式方法）如何将多属性选择概念化，并讨论了诸如先天偏好、花忠诚性和情境依赖性等现象如何影响我们对花朵选择的理解。我们发现，多属性决策是一个复杂的过程，可能会受到先天偏好、花忠诚性、选择集的组成以及经济奖励价值的影响。我们认为，为了理解和预测访花昆虫的花朵选择，我们需要超越简化的选择集，转向一种整合非奖励属性作用、纳入花忠诚性、先天偏好和情境依赖性的多属性选择观点。我们进一步提醒，行为实验在偏好实验的设计和解释中需要考虑情境依赖性的可能性。我们最后讨论了未来研究的突出问题。我们还提出了一个包含选择行为多个维度的概念框架。
