Conservation biological control (CBC) has been an active research topic for the last two decades and is now one of the key ways being explored to develop agroecological production systems. Using broad concepts and indicators, recent reviews and meta-analyses have highlighted major inconsistencies in the responses of CBC to landscape structure, revealing their context-dependent nature. To decipher these relations, we reviewed the scientific literature (50 articles) using (1) an original ontology allowing us to navigate across the different terms and concepts used in this literature and (2) a network-based methodology to describe the scattering, completeness, and generalizability of scientific knowledge on CBC. An interactive version of this network is available online. Our results highlight the strong information scattering caused by the variety of indicators used to describe both landscape structure and CBC. We observe trade-offs between the use of coarse concepts classically used in meta-analysis (e.g., landscape complexity) and the non-convergence of results (ambiguity). The network analysis points out consistently less information ambiguity when considering sub-networks focused on trophic chains than in the full information network, without losing connectance. We suggest that effects of landscape structure may be different between trophic chains because of specific selection pressures associated with cropping systems. Our novel review procedure offers a relatively simple but powerful complementary approach to classical meta-analysis to explore ecological patterns. It highlights that crop trophic chain probably represents the adequate ecological unit to investigate the landscape–CBC relationship. Designing pest suppressive landscapes while favoring farmland biodiversity will imply considering multiple crop trophic chains responding differently to landscape structure. Therefore, we recommend assessing the level of CBC at both crop field and landscape scales to inform decisions on the best individual or collective strategy to adopt.

中文翻译：

---

保护生物防治对景观结构的不一致反应：基于网络的审查的新见解

保护性生物控制（CBC）在过去的二十年里一直是一个活跃的研究课题，现在是探索发展农业生态生产系统的关键方法之一。使用广泛的概念和指标，最近的评论和荟萃分析强调了 CBC 对景观结构的反应中的主要不一致之处，揭示了它们与环境相关的性质。为了破译这些关系，我们回顾了科学文献（50 篇文章），使用（1）原始本体允许我们浏览该文献中使用的不同术语和概念，以及（2）基于网络的方法来描述散射、完整性, 以及关于 CBC 的科学知识的普遍性。该网络的交互式版本可在线获得。我们的结果突出了用于描述景观结构和 CBC 的各种指标造成的强烈信息分散。我们观察了元分析中经典使用的粗略概念（例如，景观复杂性）与结果的不收敛（模糊性）之间的权衡。网络分析指出，在考虑专注于营养链的子网络时，信息模糊性始终低于完整信息网络，而不会丢失连接。我们认为，由于与种植系统相关的特定选择压力，营养链之间景观结构的影响可能不同。我们新颖的审查程序为经典荟萃分析提供了一种相对简单但功能强大的补充方法，以探索生态模式。它强调作物营养链可能代表了研究景观-CBC关系的适当生态单元。在有利于农田生物多样性的同时设计害虫抑制景观将意味着考虑多种作物营养链对景观结构的不同反应。因此，我们建议在农田和景观尺度上评估 CBC 水平，以便为决定采用最佳个人或集体策略的决策提供信息。