Learning Processes in Matching-to-Sample by Pigeons

鸽子匹配样本的学习过程

• 批准号: 6639193
• 负责人: ANTHONY A WRIGHT
• 金额: $ 20.79万
• 依托单位: UNIVERSITY OF TEXAS HLTH SCI CTR HOUSTON
• 依托单位国家: 美国
• 财政年份: 2001
• 资助国家: 美国
• 起止时间: 2001-05-15 至 2006-04-30
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/6639193
• 关键词: behavior test; behavioral /social science research tag; cognition; computer data analysis; ethology; learning; pigeons; psychometrics; training; visual stimulus

The focus of the experiments of this proposal are the processes by which pigeons learn matching to sample (MTS) and the conditions that lead to different ways that pigeons learn MTS. A newly developed technique will be used to measure if-then rule learning in the MTS task. Another newly developed technique will be used to measure relational-concept learning during acquisition of the MTS task. Tests are proposed to manipulate and confirm the presence of configural learning when evidence indicates that the whole MTS display is learned as a configural pattern. Other evidence suggests that in certain conditions the MTS task may be learned by combinations of these different types of learning. Enhancing and inhibitory interactions between these different learning processes in MTS will be studied. Configural learning will be made more difficult by increasing the number of displays to be learned and by making the display less likely to be configured as a unitary pattern by the pigeons. These manipulations of configural learning. Other manipulations will test the durability of relational-concept learning and explore the type of learning that replaces it. This research will extend our knowledge of how tasks are learned, problems solved, and conditions that lead to either higher-order concept learning or stimulus-specific learning.

本提案实验的重点是鸽子学习样本匹配（MTS）的过程以及导致鸽子学习 MTS 的不同方式的条件。 一项新开发的技术将用于测量 MTS 任务中的 if-then 规则学习。 另一种新开发的技术将用于测量 MTS 任务获取过程中的关系概念学习。 当证据表明整个 MTS 显示被学习为配置模式时，建议进行测试来操纵和确认配置学习的存在。 其他证据表明，在某些条件下，MTS 任务可以通过这些不同类型的学习的组合来学习。我们将研究 MTS 中​​这些不同学习过程之间的增强和抑制相互作用。 通过增加要学习的显示数量以及使显示不太可能被鸽子配置为单一图案，配置学习将变得更加困难。 这些配置学习的操作。 其他操作将测试关系概念学习的持久性，并探索替代它的学习类型。 这项研究将扩展我们对如何学习任务、解决问题以及导致高阶概念学习或特定刺激学习的条件的了解。