Robust and Rapid Air-Borne Odor Tracking without Casting.
Robust and Rapid Air-Borne Odor Tracking without Casting.
复制标题
DOI：
10.1523/eneuro.0102-15.2015
复制
发表时间：
2015-11
期刊：
影响因子：
3.4
通讯作者：
Bhalla US
中科院分区：
文献类型：
--
作者：
Bhattacharyya U;Bhalla US
在野外的空气/液体气味跟踪中，投射行为(在气味流中呈之字形移动)是常见的；然而，陆地气味定位通常涉及到熟悉环境中的路径选择。为了研究这一点，我们训练大鼠在接近层流气流的多选择嗅觉竞技场中跑向气味来源。我们发现，老鼠不是投掷，而是直接跑向气味端口，如果这是不正确的，它们会连续采样其他来源。这种行为在存在扰动的情况下是一致和准确的，例如新气味、背景气味、单侧鼻孔缝合和湍流。我们开发了一个模型，该模型预测，如果在已知位置有少量目标，这种运行扫描跟踪空气中气味的速度比投射更快。因此，最佳猜测目标选择与后备序列采样相结合，为在熟悉的环境中寻找气味来源提供了一种快速而稳健的策略。
Casting behavior (zigzagging across an odor stream) is common in air/liquid-borne odor tracking in open fields; however, terrestrial odor localization often involves path selection in a familiar environment. To study this, we trained rats to run toward an odor source in a multi-choice olfactory arena with near-laminar airflow. We find that rather than casting, rats run directly toward an odor port, and if this is incorrect, they serially sample other sources. This behavior is consistent and accurate in the presence of perturbations, such as novel odors, background odor, unilateral nostril stitching, and turbulence. We developed a model that predicts that this run-and-scan tracking of air-borne odors is faster than casting, provided there are a small number of targets at known locations. Thus, the combination of best-guess target selection with fallback serial sampling provides a rapid and robust strategy for finding odor sources in familiar surroundings.