The role of mate recognition in speciation: a neural network approach

知识价值。新物种是如何产生的？当来自不同种群的雄性和雌性不能认出对方是配偶时，就会产生新的物种。当不同种群中的个体不再将彼此视为配偶时，它们就不会再进行杂交。因此，种群可能成为单独的物种。因此，了解不同种群之间的交配行为如何变得如此不同，对于理解新物种何时以及如何产生至关重要。与其他物种的相互作用是种群可能发展不同交配行为的一种方式。因为与错误的物种交配可能导致很少或没有后代，所以男性和女性都应该拥有防止他们与错误的物种交配的交配行为。如果种群在相互作用的物种上存在差异，那么种群在交配行为上就会出现分歧。本研究将通过研究与不同异种物种的相互作用是否会导致雄性和雌性交配行为的多样化来评估这一问题。为此，人工神经网络模型——模拟神经功能的数学模型——将被开发出来，以模拟不同物种种群中雌性配偶偏好和雄性交配信号的发展。此外，这些模型的预测将使用从一个掌足蟾蜍（Spea multiplicata）研究系统收集的经验数据进行评估。在实现这些目标的过程中，这项研究将帮助我们了解生物多样性的起源。更广泛的影响。随着人工神经网络的使用越来越普遍，评估该技术在不同学科中的有效性至关重要。由于将使用经验数据来测试这些模型的预测，因此该项目将评估人工神经网络模型作为预测自然系统行为的研究工具的使用。该项目还将通过提供获得研究机会的途径，寻求增加代表性不足的群体在科学界的代表性。拟议的研究将在历史上少数民族大学（hmu）的演讲中使用，以说明如何将计算技术应用于动物行为问题。这些演讲将用于从hmu招募学生参与与资助提案相关的研究项目。最终的目标是向少数族裔学生传达基础研究是一条可行的职业道路。

Intellectual Merit. How do new species arise? New species may arise when males and females from different populations fail to recognize each other as mates. When individuals in different populations no longer recognize each other as mates, they will no longer interbreed. Consequently, the populations may become separate species. Understanding how mating behaviors become so distinct among populations is therefore critical to understanding when and how new species are generated. Interactions with other species are one way that populations may develop different mating behaviors. Because mating with the wrong species can result in few or no offspring, males and females are expected to possess mating behaviors that prevent them from mating with the wrong species. If populations differ in the species with which they interact, then populations may become divergent in their mating behaviors. This research will evaluate this issue by examining whether interactions with different heterospecific species can cause the diversification of both male and female mating behaviors. To that end, artificial neural network models - mathematical models that mimic neural function - will be developed that simulate the development of female mate preferences and male mating signals in populations that occur with different species. Moreover, the predictions of these models will be evaluated using empirical data collected from a spadefoot toad (Spea multiplicata) study system. In addressing these goals, this research will help us understand the origins of biodiversity. Broader Impacts. As the use of artificial neural networks becomes increasingly common, it is critical to evaluate the effectiveness of this technique across different disciplines. Because empirical data will be used to test the predictions of these models, this project will evaluate the use of artificial neural network models as a research tool for predicting the behavior of natural systems. This project will also seek to increase representation of underrepresented groups in science by providing access to research opportunities. The proposed research will be used in presentations at historically minority universities (HMUs) to illustrate how computational techniques can be applied to questions in animal behavior. These presentations will be used to recruit students from the HMUs into research projects associated with the funded proposal. Ultimately, the goal is to convey to minority students that basic research is a viable career path.