Suppr超能文献

大黄蜂对多种习得目标类型的视觉搜索。

Bumblebee visual search for multiple learned target types.

机构信息

Biological and Experimental Psychology, School of Biological and Chemical Sciences, Queen Mary University of London, Mile End Road, London E1 4NS, UK.

出版信息

J Exp Biol. 2013 Nov 15;216(Pt 22):4154-60. doi: 10.1242/jeb.085456. Epub 2013 Aug 15.

Abstract

Visual search is well studied in human psychology, but we know comparatively little about similar capacities in non-human animals. It is sometimes assumed that animal visual search is restricted to a single target at a time. In bees, for example, this limitation has been evoked to explain flower constancy, the tendency of bees to specialise on a single flower type. Few studies, however, have investigated bee visual search for multiple target types after extended learning and controlling for prior visual experience. We trained colour-naive bumblebees (Bombus terrestris) extensively in separate discrimination tasks to recognise two rewarding colours in interspersed block training sessions. We then tested them with the two colours simultaneously in the presence of distracting colours to examine whether and how quickly they were able to switch between the target colours. We found that bees switched between visual targets quickly and often. The median time taken to switch between targets was shorter than known estimates of how long traces last in bees' working memory, suggesting that their capacity to recall more than one learned target was not restricted by working memory limitations. Following our results, we propose a model of memory and learning that integrates our findings with those of previous studies investigating flower constancy.

摘要

视觉搜索在人类心理学中得到了很好的研究,但我们对非人类动物类似的能力知之甚少。人们有时假设动物的视觉搜索一次只能限制在一个目标上。例如,在蜜蜂中,这种限制被用来解释花的恒定性,即蜜蜂专门针对单一花型的倾向。然而,很少有研究在控制先前视觉经验的情况下,调查蜜蜂对多种目标类型的视觉搜索,经过长时间的学习。我们在单独的辨别任务中对颜色幼稚的熊蜂(Bombus terrestris)进行了广泛的训练,以识别穿插块训练课程中的两种奖励颜色。然后,我们在存在干扰颜色的情况下同时用两种颜色对它们进行测试,以检查它们是否以及多快能够在目标颜色之间切换。我们发现蜜蜂能够快速且经常地在视觉目标之间切换。切换目标的中位数时间短于已知的蜜蜂工作记忆中痕迹持续时间的估计值,这表明它们回忆多个学习目标的能力不受工作记忆限制的限制。根据我们的结果,我们提出了一个记忆和学习的模型,该模型将我们的发现与之前研究花恒定性的研究结果结合在一起。

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验