Name: 作者聚焦：训练实验动物进行轻柔、无压力的处理
Uploaded: 2024-02-16T13:50:00
Duration: 1 min 45 s
Description: Watch this Scientific Journal Video about Author Spotlight: Training of Laboratory Animals for Gentle and Stress-Free Handling at JoVE.com

作者感谢动物看护人对兔子的照顾以及他们富有同情心和敬业精神的工作。感谢 Kai Diederich 对手稿的校对。

动物的维护得到了柏林国家当局的批准（“Landesamt für Gesundheit und Soziales”，许可证编号：ZH3 - 德国联邦风险评估研究所，柏林）。培训协议是在常规畜牧业的框架内制定的，以完善处理、健康检查和称重等程序。由于训练协议没有造成任何痛苦、痛苦或伤害，并且所有观察都是在常规畜牧业的背景下进行的，因此根据欧洲立法，这项研究不被视为动物实验。
一、培训的一般要求
注意：以下几点通常是动物训练所必需的，独立于本议定书。
<ol><li>清楚地标记动物以进行识别。</li>
	<li>确保获得高质量的奖励，例如，非常可口的食物。</li>
	<li>定义训练计划，包括实现目标行为的所有步骤，即动物应该表现出的期望行为。 注意：目标行为是逐步形成的，而每个步骤都由目标标准定义，应该很容易完成。</li>
	<li>在进行下一步之前，定义动物是否应该成功执行 80% 或 100% 的训练步骤。</li>
	<li>准备文件（例如， 补充表1）。</li>
	<li>不要在匆忙时训练动物，并注意举止、语气和运动速度，即轻声细语、平静地说话，缓慢而有预见性地移动。</li>
	<li>测试在上一次训练中训练的步骤。</li>
	<li>当动物的一般健康状况受损和/或动物对与训练师互动不感兴趣时，不要训练动物。</li>
</ol>2. 本议定书的训练程序
注：为本协议定义了以下训练程序。
<ol><li>在动物室/围栏（图2）中设置训练场，并根据要训练的目标行为，将秤或运输箱（地板30厘米×50厘米;门27厘米×25.5厘米;盖子开口16厘米×39厘米）放置在竞技场中。</li>
	<li>将秤和运输箱放在竞技场外，让兔子熟悉这些新奇的物体，然后再训练目标行为“称重”和“运输箱”。</li>
	<li>对于目标行为（参见步骤 6-9），按照步骤 5.7 中所述，引导兔子进入训练场。</li>
	<li>在训练场上坐在地板上或使用矮凳/台阶坐下。</li>
	<li>按照协议中描述的顺序一个接一个地训练目标行为，而不是同时进行。如果动物连续两天未能学习目标行为的一个步骤，则破例 - 在这种情况下，同时训练下一个目标行为。如果这有利于个人的训练过程，请更改目标行为的顺序。</li>
	<li>在习惯和答题器介绍之后，开始在目标行为的第 1 步 “跟随目标”（即，目标杆末端的球被定义为目标）训练所有动物。请注意，根据他们的个人表现，他们可能会在这一天达到不同的训练步骤。在接下来的几天里，根据动物的个体进展继续进行训练。</li>
	<li>在一周的每个工作日安排一次训练课程（例如，对于有六到七只动物的小组，大约30分钟）。 注意：当训练师在动物房间/围栏内并准备开始训练时，训练课程就开始了。当没有一只兔子（不再）参与训练，或者所有兔子都达到了要在此会话中训练的目标行为的最后训练步骤时，会话结束。从协议的最后一个成功步骤开始每个培训课程。一个疗程包括每只动物至少一次或多次试验。</li>
	<li>通过要求动物执行特定的训练步骤（即，在兔子学会用鼻子/嘴巴触摸目标之后的步骤3.1）开始试验，并在动物不再参与训练（不再）或动物达到要训练的目标行为的最后训练步骤时结束试验。训练结束时，打开训练场，引导兔子离开场;然后握住目标棒（目标朝向天花板）以及奖励碗放在胸前。</li>
	<li>在喂食、清洁、治疗或实验程序之前安排训练。</li>
	<li>使用带有集成答题器的奖励碗（图 3， 补充编码文件 1），对于某些目标行为，使用额外的目标杆。</li>
	<li>在奖励碗中装满可口的食物，例如葵花籽和以谷物为基础的觅食零食。 注意：兔子是食叶动物，这些奖励示例不是最佳的;然而，由于实验动物设施的卫生限制，它们是这些动物的最佳替代品。如果可能的话，应该优先考虑新鲜的草药作为食物奖励。请注意，只有当兔子真的喜欢它时，奖励才是奖励。</li>
	<li>确保始终将奖励碗保持在相同的位置，例如，在胸部前面，并且只有在点击后才开始将奖励碗向兔子移动。除了目标行为“饲养”和“跳腿”外，一只手拿着目标棒，另一只手拿着奖励碗。</li>
	<li>在正确显示所需行为时，在兔子的鼻子/嘴巴所在的位置展示奖励。</li>
	<li>注意兔子在赠送奖励时只吃少量食物。</li>
	<li>如果一只兔子在五次尝试中至少表现出四次所需的行为（80% 的成功率），请继续下一步。如果动物无法执行下一步（即多次未能执行该步骤），请返回到协议的上一步。如果需要返回上一步或兔子不参加训练，请修改训练协议（例如，定义较小的训练步骤），选择另一种（更可口的）食物奖励或检查训练师的训练技能。</li>
	<li>如果一只兔子正确地执行了目标行为的最后训练步骤，请在同一天完成这只兔子的训练课程。</li>
	<li>打印文档表（补充文件1）、详细的培训协议以及简化的培训协议（补充文件2），并将它们附加到剪贴板或在平板电脑上打开它们。使用文档表记录每个会话中单个动物所达到的训练步骤。</li>
</ol>3. 情境条件反射/习惯化
注意：在连续五天（周一至周五）内，训练师和动物看护人每天对每组兔子进行约 15 分钟的情境调节/适应课程。
<ol><li>轻声和兔子说话。</li>
	<li>提供可口的食物奖励。 注意：请注意，饮食的突然变化可能会导致兔子的消化问题。因此，慢慢引入新颖的食物奖励。</li>
	<li>将装满美味食物的奖励碗移向兔子，让它们从中吃东西。</li>
	<li>轻轻触摸兔子的脖子、肩膀、背部和臀部，不要束缚它们。</li>
</ol>4. 答题器简介
注意： 目标行为：走近教练，并在 clic k 之后从碗中吃掉奖励。目标是让兔子将食物奖励与点击相关联，即兔子在点击后吃掉食物奖励。
<ol><li>让兔子靠近。</li>
	<li>单击一次，然后将奖励碗移向兔子（称为单击和治疗或 C 和 T）。</li>
	<li>重复步骤 4.2 十次。当第十次点击时，等待一秒钟，然后再显示奖励，并观察兔子的行为。 注意：如果动物在点击后立即寻求奖励，则可以假设辅助强化器的调节是成功的。</li>
	<li>重复步骤 4.1.-4.3。第二天，如果一只兔子在点击后至少八次没有吃下食物，则奖励。</li>
</ol>5. 遵循目标
注意： 目标行为：跟随并用鼻子/嘴巴触摸目标。第 6-9 节中的目标行为是对第 5 节中提出的目标行为的修改。
<ol><li>将目标放在靠近兔子的鼻子/嘴巴的地方。 注意： 目标杆末端的球被定义为目标。<ol><li>当兔子用鼻子/嘴巴接触目标时，C 和 T。 注意：确保兔子的鼻子/嘴巴不会在任何未定义的位置接触棍子，而是球。</li>
		<li>在奖励的同时将目标从兔子的视野中移除。</li>
		<li>重复上述所有步骤五次，直到兔子成功执行此步骤至少五次中的四次（成功率为 80%）。</li>
	</ol></li>
	<li>将目标放置在不同的位置，在那里可以通过头部运动用鼻子/嘴巴触摸目标，而无需移动整个身体。<ol><li>当兔子用鼻子/嘴巴接触目标时，C 和 T 并按照步骤 5.1.2 和 5.1.3 进行操作。</li>
	</ol></li>
	<li>将目标放置在离兔子更远的地方，让兔子向它伸展。<ol><li>当兔子用鼻子/嘴巴接触目标时，C 和 T 并按照步骤 5.1.2 和 5.1.3 进行操作。</li>
	</ol></li>
	<li>进一步增加目标和兔子之间的距离，要求动物跳一跳才能到达目标。<ol><li>当兔子用鼻子/嘴巴接触目标时，C 和 T 并按照步骤 5.1.2 和 5.1.3 进行操作。</li>
	</ol></li>
	<li>将目标放在兔子的鼻子/嘴巴前面，以便兔子可以触摸目标，然后慢慢将目标远离兔子。确保兔子必须进行可变数量的跳跃才能跟随目标。 注意：在五次尝试中，第 1次、第 2次、第 3次、第 4次和第 5次 分别在 2、1、2、3 和 2 次跳跃后获得奖励（平均：2 次跳跃后）。<ol><li>C 和 T，当兔子进行所需的跳跃数并遵循步骤 5.1.2 和 5.1.3 时。</li>
	</ol></li>
	<li>重复步骤 5.5 并将目标移动到两个对象之间的走廊中。 注意：在此步骤中，两个对象（如隧道）彼此平行放置并形成一个走廊。<ol><li>C 和 T，当兔子进行所需的跳跃数并按照步骤 5.1.2 和 5.1.3 进行操作时。</li>
	</ol></li>
	<li>重复步骤 5.5 并将目标移向训练场，以便兔子进入训练场。<ol><li>C 和 T，当兔子进行所需的跳跃数并按照步骤 5.1.2 和 5.1.3 进行操作时。</li>
	</ol></li>
</ol>6. 称重
注意： 目标行为：在体重秤上停留至少一秒钟。 如果秤的表面太滑，可以在顶部放一个防滑垫。然而，有些兔子往往会啃垫子。
<ol><li>按照步骤 5.5 中所述，将目标棒引导兔子朝向秤。<ol><li>当兔子用鼻子/嘴巴接触目标时，C 和 T。</li>
		<li>在奖励的同时将目标从兔子的视野中移除。</li>
		<li>重复上述所有步骤五次，直到兔子成功执行此步骤至少五次中的四次（成功率为80%）。</li>
	</ol></li>
	<li>将目标稍微移到秤上方，使兔子用前爪踩在上面，用鼻子/嘴巴接触目标。<ol><li>C 和 T，并按照步骤 6.1.2 和 6.1.3 操作。</li>
	</ol></li>
	<li>将目标稍微移出刻度中心，使兔子用前爪踩在上面，然后伸展到刻度上，用鼻子/嘴巴接触目标。<ol><li>C 和 T，并按照步骤 6.1.2 和 6.1.3 操作。</li>
	</ol></li>
	<li>将目标进一步移动到刻度中心之外，使兔子用前爪和后爪踩在刻度上，用鼻子/嘴巴接触目标。<ol><li>C 和 T，并按照步骤 6.1.2 和 6.1.3 操作。</li>
	</ol></li>
	<li>按照 4.4 中的描述将兔子引导到秤上。并在 C 和 T 之前等待几秒钟。 注意：在此步骤中，在秤上坐下可变时间后奖励兔子。在五次尝试中，第 1次、第 2次、第 3次、第 4次和第 5次 ，兔子分别在 0.5、1.0、2.0、0.5 和 1.0 秒后获得奖励（平均：1.2 秒）。虽然兔子在此期间保持静止，但可以从秤上读取重量。如果秤需要更多时间来稳定和测量重量，请增加动物在秤上的坐姿时间。<ol><li>C 和 T，并按照步骤 6.1.2 和 6.1.3 操作。</li>
	</ol></li>
</ol>7. 进入运输箱
注意： 目标行为：进入和离开运输箱。对于培训课程，使用一个运输箱（地板 30 厘米× 50 厘米;门 27 厘米× 25.5 厘米），盖子上有一个开口（16 厘米× 39 厘米），底部覆盖着脏污的垫料。
<ol><li>按照步骤 5.5 中所述，将目标杆引导兔子朝向盒子。<ol><li>C 和 T 当兔子在盒子入口前时。</li>
		<li>在奖励的同时将目标从兔子的视野中移除。</li>
		<li>重复上述所有步骤五次，直到兔子成功执行任务至少五次中的四次（成功率为80%）。</li>
	</ol></li>
	<li>将目标棒穿过盖子开口，将其移入盒子中，从而激励兔子用前爪进入盒子。<ol><li>C 和 T，并按照步骤 7.1.2 和 7.1.3 操作。</li>
	</ol></li>
	<li>按照步骤 7.2 将兔子带入盒子（即，所有前爪和后爪都进入盒子）。<ol><li>C 和 T，并按照步骤 7.1.2 和 7.1.3 操作。</li>
	</ol></li>
	<li>按照步骤 7.3 并关闭盒子的门几秒钟。然后再次打开门。<ol><li>C 和 T，并按照步骤 7.1.2 和 7.1.3 操作。</li>
	</ol></li>
	<li>按照步骤 7.3 并关闭盒子的门。小心地将盒子稍微抬起（即，离地面几厘米）。将盒子放回地板上，然后打开门。<ol><li>C 和 T，并按照步骤 7.1.2 和 7.1.3 操作。</li>
	</ol></li>
	<li>按照步骤 7.5 操作。然后将带有目标棒的兔子从盒子中引出。<ol><li>C 和 T，并按照步骤 7.1.2 和 7.1.3 操作。</li>
	</ol></li>
</ol>8. 饲养
注意： 目标行为：在饲养时将前爪放在手掌上。在训练课程中，用一只手握住目标棒和奖励碗。
<ol><li>将目标放在兔子的头顶上方，使兔子将鼻子向上伸展，用鼻子/嘴巴接触目标，同时所有爪子都保持在地面上。<ol><li>C 和 T。</li>
		<li>在奖励的同时将目标从兔子的视野中移除。</li>
		<li>重复上述所有步骤五次，直到兔子成功执行此步骤至少五次中的四次（成功率为80%）。</li>
	</ol></li>
	<li>将目标放在兔子的头顶上方，如步骤 8.1 中所述。兔子向上伸展鼻子后，将目标进一步向上移动，使兔子将前爪抬离地面并用鼻子/嘴巴接触目标。<ol><li>C 和 T，并按照步骤 8.1.2 和 8.1.3 操作。 注意：如果兔子无法执行 8.2，可以在 8.2 之后添加一个临时步骤。为此，请按照 8.2 进行操作，然后用空闲的手掌接近并触摸兔子的前爪（手掌表面）。C 和 T，并按照步骤 8.1.2 和 8.1.3 操作。</li>
	</ol></li>
	<li>将空出的手掌放在兔子面前，将目标移动到兔子的头顶上方，使兔子将前爪放在手上，用鼻子/嘴巴接触目标。<ol><li>C 和 T 一秒钟后，并按照步骤 8.1.2 和 8.1.3 操作。 注意：虽然前爪放在训练师的手上，但可以目视检查兔子的腹部。</li>
	</ol></li>
</ol>9.跳上圈
注意： 目标行为：跳到教练的腿上，在被触摸时保持坐姿。对于步骤 9.5-9.9，用同一只手握住目标棒和奖励碗，因为一只手需要触摸兔子。
<ol><li>按照协议步骤 5.5 中所述，将目标棒靠近训练器引导兔子。<ol><li>当兔子用鼻子/嘴巴接触目标时，C 和 T。</li>
		<li>在奖励的同时将目标从兔子的视野中移除。</li>
		<li>重复上述所有步骤五次，直到兔子成功执行此步骤至少五次中的四次（成功率为80%）。</li>
	</ol></li>
	<li>将目标移动到训练器的腿上方，使兔子将前爪放在腿上，用鼻子/嘴巴接触目标。<ol><li>C 和 T，并按照步骤 9.1.2 和 9.1.3 操作。</li>
	</ol></li>
	<li>将目标移动到腿上方，稍微远离兔子，使兔子将前爪放在腿上并向前伸展以用鼻子/嘴巴接触目标。<ol><li>C 和 T，并按照步骤 9.1.2 和 9.1.3 操作。 注意：如果兔子无法执行 9.4，可以在 9.3 之后添加一个临时步骤。为此，请按照步骤 9.3 操作，然后用奖励碗将兔子引诱到膝盖上。当兔子将所有四只爪子放在膝盖上时，C 和 T 并按照步骤 9.1.2 和 9.1.3 进行操作。</li>
	</ol></li>
	<li>将目标移动到腿上方，甚至离兔子更远，这样兔子就会跳到腿上，用鼻子/嘴巴触摸目标。<ol><li>C 和 T，并按照步骤 9.1.2 和 9.1.3 操作。</li>
	</ol></li>
	<li>按照步骤 9.4 操作。并小心地用空出的手从颅骨到尾部（从肩膀开始）抚摸兔子，同时让兔子坐在腿上。<ol><li>C 和 T，并按照步骤 9.1.2 和 9.1.3 操作。</li>
	</ol></li>
	<li>按照步骤 9.4 操作，当兔子坐在腿上时，用空闲的手从颅到尾小心地抚摸兔子的耳朵。<ol><li>C 和 T，并按照步骤 9.1.2 和 9.1.3 操作。</li>
	</ol></li>
	<li>按照步骤 9.4 操作。轻轻地抓住兔子的一只耳朵，目视检查耳朵。<ol><li>C 和 T，并按照步骤 9.1.2 和 9.1.3 操作。</li>
	</ol></li>
	<li>按照步骤 9.4 操作。当兔子坐在腿上时，用空出的手抚摸前爪或后爪。 注意： 如果兔子的右侧身体朝向，左侧身体背对训练器，则可以轻轻触摸左爪（反之亦然）。确保交替训练双方。<ol><li>C 和 T，并按照步骤 9.1.2 和 9.1.3 操作。</li>
	</ol></li>
	<li>按照步骤 9.4 操作。当兔子坐在腿上时，用空出的手小心地抬起前爪或后爪。 注意： 当爪子抬起时，可以检查爪子。确保交替训练双方。<ol><li>C 和 T，并按照步骤 9.1.2 和 9.1.3 操作。</li>
	</ol></li>
</ol>

基于正强化的常规兔子饲养程序训练

作者声明与本出版物没有利益冲突。

2015 年至 2019 年间，欧盟（和挪威）有超过 4900 万只动物被用于科学目的;其中 1,745,037 （3.5 %） 是兔子1.家兔主要用于监管研究（例如，质量控制、毒性和其他安全测试，包括药理学）和生物物质（例如，血液制品）的常规生产2。尽管努力用无动物方法取代动物实验，但出于其中一些目的，仍然需要使用动物。每当动物实验无法被替代时，减少动物数量并改进实验以及住房和饲养条件以尽量减少痛苦至关重要。Russel 和 Burch 在 1959 年将这一策略描述为 3R 原则（替换、减少、改进）3，并于 2010 年将其实施到关于保护用于科学目的的实验动物的指令 2010/63/EU中 4。精致的作用，有时被称为 3Rs5 的“灰姑娘”，已经变得越来越重要。虽然替代和减少是实现指令完全替代动物实验的最终目标的长期战略，但改进可以立即改善动物福利5，这反过来又有可能提高科学质量6。正如指令 2010/63/EU 附件 III 中指出的那样，改进的关键部分是动物处理和培训。该指令规定，动物设施应为实验动物设计适应物种、程序和项目4 的适应和培训计划。ETS 第 123 号附录 A 建议实验动物人员花时间“与动物交谈、处理、训练和梳理动物”7。
一些处理程序会引起压力，尤其是对于兔子等陆生猎物。例如，被逼入绝境、束缚（即被人抓住、抱在人的手臂上或被设备固定）和抬起（即从地板上捡起）进行健康检查或实验程序8.一项针对宠物兔子的调查显示，其中 57-61% 的兔子在被举起时会挣扎 9,10，有些甚至表现出与恐惧相关的攻击性（即咬人）11。这强调了温和的处理和培训技术的必要性。对于其他小型哺乳动物（如小鼠）来说，温和的处理技术已经证明可以减少焦虑12.此外，温和的处理与训练相结合可以减少小鼠的压力、焦虑和类似抑郁的行为13。在恒河猴和黑猩猩中，当它们之前接受过程序训练时，与压力相关的生理措施受到的影响较小14,15。可以假设，如果动物可以选择自愿参加训练，它就可以积极参与并控制这种情况，从而提高动物的福利，也可能有助于获得更稳健和可靠的科学数据16.
许多动物训练程序依赖于操作性条件反射原则，特别是使用正强化和塑形 17,18,19。操作性条件反射要求动物引发行为，然后可以增加或减少20。有四种常见的意外事件用于描述操作性条件反射：积极和消极的强化以及积极和消极的惩罚21,22.在兔子中，当一个人接近（不想要的刺激）动物以将其移离或朝向特定位置时，会使用积极的惩罚23.在处理检查台上的兔子时，负强化的一个例子是当兔子保持静止时，消除对兔子身体的压力（不需要的刺激）;负面惩罚的一个例子是在兔子开始挣扎时撤回食物奖励（期望的刺激）。在动物训练中，既不建议进行积极/消极的惩罚，也不建议进行消极的强化。惩罚会导致负面的情绪影响24，例如狗25 所表现出的痛苦。相反，在训练动物时，正强化应该是首选的方法25,26。积极强化训练 （PRT） 意味着在动物表现出期望的行为后提供理想的刺激（例如，食物奖励），这增加了动物在未来再次表现出期望行为的可能性27.PRT 通常涉及使用有条件的增强剂。条件强化物（次级强化物）通常是中性刺激，例如声音，动物学会将其与无条件强化物（主要强化物，例如食物奖励）相关联28。在动物表现出期望的行为后，立即呈现有条件的强化物，然后呈现食物奖励26。PRT的一个著名例子是咔嗒声训练，其中咔嗒声器用于产生咔嗒声作为辅助增强器26。
训练应遵循可预测的模式，这是通过设计训练协议来实现的。训练协议描述了如何逐步达到目标行为（即行为塑造），而每个步骤都由客观标准定义，并且应该易于完成26.如果训练遵循特定的协议，实验动物工作人员可能会交替训练动物。需要注意的重要一点是，人员需要了解学习和培训技术的原则，以便正确应用它们26.动物训练通常包含行为塑造。斯金纳及其同事在 1940 年代初期首次故意使用手工塑造行为塑造29.相比之下，在以前的工作中，斯金纳更专注于“在物理环境中进行微小的改变，以实现逐次逼近程序”29。在《科学美国人》上发表的一篇文章中，他解释说，强化行为会增加行为重复的可能性，这“使塑造动物的行为成为可能，几乎就像雕塑家塑造一块粘土一样”30。从那时起，塑形对圈养动物的重要性越来越大 29,30,31,32。为了复制和改进整形协议，以系统和客观的方式评估协议至关重要 33,34,35,36。
Pfaller-Sadovsky 等人最近发表的一项系统评价和荟萃分析发现，条件强化最常在马和狗中研究，尽管猫、牛、鱼、山羊和非人类灵长类动物也进行了研究37。在实验动物科学中，PRT 广泛用于非人灵长类动物 38,39,40 和狗 41,42。此外，猪43、山羊44，甚至小型哺乳动物，如小鼠13 和大鼠 45,46,47，在实验室环境中使用答题器成功训练。据作者所知，将PRT引入实验室兔的报道很少，但已有48例;几十年前发表的现有同行评议文献已经过时，不符合现代伦理研究的标准。尽管很少有未经同行评审的信息表明可以使用 PRT49 成功训练兔子，但似乎缺乏关于饲养在实验动物设施中的兔子的合适训练方案的知识。根本原因可能各不相同。例如，所使用的培训技术可能没有在研究文章中详细描述，或者数据是为监管研究而生成的，因此不会发布。此外，兔子可能不是精细化研究的主要焦点，因为其他物种（如小鼠和大鼠）在研究中更常用。由于兔子可以很好地掩盖压力和不适，因此它们在被处理时的负面情绪状态可能经常被忽视。兔子的大小使人类能够限制这些动物进行手术，而其他物种（如非人类灵长类动物）可能无法用手握住进行类似的干预，而不会受到严重伤害50.然而，这些事实不应阻止实验室兔子PRT协议的开发或共享。
关于PRT的潜在好处，首先尝试填补实验室兔子适当训练方案的知识空白。为了促进处理、健康检查和称重等常规程序，制定了培训协议，并评估了它们对雌性新西兰白兔的可行性。进行了比较，以评估训练方案中描述的替代处理技术是否比常规处理技术更令人厌恶，以及训练方案是否可以作为改进措施。 图 1 显示了本研究的时间表。
动物维护
13只雌性新西兰白兔是从商业来源获得的（在大约7周大时到达动物设施）。这些动物没有FELASA建议51中列出的所有病毒、细菌和寄生虫病原体。在执行本议定书的步骤时，穿戴了防护服和设备（手套、口罩、发网）。一组六只和一组七只动物分别被饲养在2.8 m×2.8 m（地板住房）的地板围栏中。地板上覆盖着细小而片状的木制床上用品（高压灭菌）。此外，筑巢材料和纸棉散落在垫料的顶部。三座塑料房屋，各有两个开口（37厘米×60厘米;高30厘米）和一个塑料隧道（长：58厘米;直径：16厘米）作为庇护所。此外，还提供了丰富的物品，如木制啃块、干草和柳球，以及装满食物颗粒的零食球，这些物品在移动球时会掉落。兔子可以自由使用自来水、颗粒状食物、高压灭菌的干草和稻草。有关食品和浓缩物项的更多信息，请参阅 材料表。动物房在星期四打扫干净。
在标准条件下（室温和相对湿度为平均值±标准差：20.3 ± 0.4 °C 和 50 ± 5%），在 12：12 小时的光暗循环（人造光具有额外的日光入射）中保持。在光相开始前30分钟打开收音机约8小时，以使兔子适应环境噪音。

<table border="1" fo:keep-together.within-page="1" fo:keep-with-next.within-page="always">
	<tbody>
		<tr>
			<td>Bed-r&#8217;Nest</td>
			<td>The Andersons, Inc., Maumee, USA</td>
			<td></td>
			<td>The successful implementation of the protocol does not depend on this particular product. You could use comparable products by other companies.&#160;</td>
		</tr>
		<tr>
			<td>Finger Clicker</td>
			<td>ASCO</td>
			<td></td>
			<td>https://ascoshop.de/products/fingerclicker</td>
		</tr>
		<tr>
			<td>G*Power</td>
			<td></td>
			<td>version 3.1.9.7</td>
			<td></td>
		</tr>
		<tr>
			<td>Heu-Proppen</td>
			<td>Ssniff Spezialdi&#228;ten GmbH, Soest, Germany</td>
			<td></td>
			<td>The successful implementation of the protocol does not depend on this particular product. You could use comparable products by other companies.&#160;</td>
		</tr>
		<tr>
			<td>IBM Corp. Released 2020. IBM SPSS Statistics for Windows,&#160; Armonk, NY: IBM Corp</td>
			<td>IBM Corp</td>
			<td>Version 27.0.</td>
			<td>in the manuscript referred to as &#34;software for advanced statistical analysis&#34;</td>
		</tr>
		<tr>
			<td>Kaninchen Haltung + 25 mg/kg Salinomycin-Natrium-Sondermischung</td>
			<td>Ssniff Spezialdi&#228;ten GmbH, Soest, Germany</td>
			<td></td>
			<td>The successful implementation of the protocol does not depend on this particular product. You could use comparable products by other companies.&#160;</td>
		</tr>
		<tr>
			<td>Lignocel block</td>
			<td>Lignocel J. Rettenmaier &#38;S&#246;hne GmbH + Co KG, Rosenberg, Germany</td>
			<td></td>
			<td>The successful implementation of the protocol does not depend on this particular product. You could use comparable products by other companies.&#160;</td>
		</tr>
		<tr>
			<td>Lillebro gesch&#228;lte Sonnenblumenkerne&#8221;, Martina GmbH, M&#252;nchen, Germany) and Rainbow Foraging Bits (Bio-Serv, Flemington, USA)&#160;</td>
			<td>Martina GmbH, M&#252;nchen, Germany</td>
			<td></td>
			<td>The successful implementation of the protocol does not depend on this particular product. You could use comparable products by other companies.&#160;</td>
		</tr>
		<tr>
			<td>Microsoft Excel for Microsoft 365 MSO</td>
			<td>Microsoft Corporation</td>
			<td>Version 2305</td>
			<td>in the manuscript referred to as &#34;spread sheet&#34;</td>
		</tr>
		<tr>
			<td>New Zealand White rabbits&#160;</td>
			<td>Charles River Laboratories France</td>
			<td></td>
			<td></td>
		</tr>
		<tr>
			<td>Plattformwaage DE 35K5DL</td>
			<td>KERN &#38; SOHN GmbH, Balingen-Frommern, Germany</td>
			<td></td>
			<td>The successful implementation of the protocol does not depend on this particular product. You could use comparable products by other companies.&#160;</td>
		</tr>
		<tr>
			<td>Premium scientific bedding (fine &#38; flake)</td>
			<td>SAFE, Augy, France</td>
			<td></td>
			<td>The successful implementation of the protocol does not depend on this particular product. You could use comparable products by other companies.&#160;</td>
		</tr>
		<tr>
			<td>Rainbow Foraging Bits</td>
			<td>Bio-Serv, Flemington, USA</td>
			<td></td>
			<td>The successful implementation of the protocol does not depend on this particular product. You could use comparable products by other companies.&#160;</td>
		</tr>
		<tr>
			<td>Snackball f&#252;r Kaninchen &#8211; 7 cm</td>
			<td>TRIXIE Heimtierbedarf GmbH &#38; Co. KG, Tarp, Germany</td>
			<td></td>
			<td>The successful implementation of the protocol does not depend on this particular product. You could use comparable products by other companies.&#160;</td>
		</tr>
		<tr>
			<td>Target stick with integrated clicker</td>
			<td>TRIXIE Heimtierbedarf GmbH &#38; Co. KG, Tarp, Germany</td>
			<td></td>
			<td>The successful implementation of the protocol does not depend on this particular product. You could use comparable products by other companies.&#160;</td>
		</tr>
		<tr>
			<td>Trixi Transportbox Capri Open Top</td>
			<td>TRIXIE Heimtierbedarf GmbH &#38; Co. KG, Tarp, Germany</td>
			<td></td>
			<td>The successful implementation of the protocol does not depend on this particular product. You could use comparable products by other companies.&#160;</td>
		</tr>
		<tr>
			<td>Weiden-Heuball</td>
			<td>JR FARM GmbH, Holzheim-Pessenburgheim, Germany</td>
			<td></td>
			<td>The successful implementation of the protocol does not depend on this particular product. You could use comparable products by other companies.&#160;</td>
		</tr>
	</tbody>
</table>

training laboratory rabbits to refine routine husbandry procedures

需要对实验动物进行非厌恶的处理和培训技术，以促进实验和常规饲养程序，提高动物福利和科学质量。答题器训练被用来为兔子制定训练方案，以完善通常与举起（即从地板上被捡起）/约束（即被抱在人类的手臂中）相关的压力大的常规饲养程序。13只雌性新西兰白兔接受了为期三周的训练。所有兔子都学会了预定的目标行为：它们跟随目标棒，跳到秤上，进入运输箱，并在将前爪放在训练员的手上时饲养。此外，十只动物从地板上跳到坐着的训练师的腿上，让训练师坐在训练师的腿上时将爪子从表面抬起。对于一些人来说，必须通过额外的临时步骤来调整这些协议。在训练结束时，兔子可靠地表现出预期的目标行为，即使在短暂和长时间的训练休息之后也是如此。除了少数例外，除了训练师之外，一个熟悉的人可以从兔子那里引出目标行为（泛化），尽管需要进一步的课程才能进行泛化。在自愿接近测试中，兔子更喜欢在第 1次 试验中与训练师互动，但在第 2 次试验中，与不熟悉的人相处的时间与与训练师相处的时间一样多。行为观察表明，按照协议中的描述，用运输箱捡起兔子，而不是用脖子的束缚来束缚它们并将它们举在手臂上，这样不那么令人厌恶。总而言之，训练方案是可行的，可以作为实验动物设施的改进策略。为了动物福利，应尽可能应用训练规程。

Between 2015 and 2019, more than 49 million animals were used for scientific purposes in the European Union (and Norway); 1,745,037 (3.5 %) of them were rabbits1. Rabbits are mostly used for regulatory research (e.g., quality control, toxicity, and other safety testing, including pharmacology) and routine production of biological substances (e.g., blood-based products)2. Although efforts are made to replace animal experiments with animal-free methods, the use of animals is still necessary for some of these purposes. Whenever an animal experiment cannot be replaced, it is crucial to reduce the number of animals and refine experimental as well as housing and husbandry conditions to minimize suffering. Russel and Burch described this strategy as the 3R principle (replace, reduce, refine) in 19593 and, in 2010, it was implemented into the Directive 2010/63/EU on the protection of laboratory animals used for scientific purposes4. The role of refinement sometimes referred to as &#39;Cinderella&#39; of the 3Rs5, has gained growing importance. While replacement and reduction are long-term strategies towards the ultimate goal of the Directive to fully replace animal experiments, refinement enables the immediate improvement of animal welfare5, which in turn has the potential to improve scientific quality6. Essential parts of refinement are animal handling and training, as pointed out in Annex III of the Directive 2010/63/EU. The Directive stipulates that animal facilities should design habituation and training programs for the laboratory animals adapted to species, procedure, and project4. Appendix A of the ETS No. 123 suggests that the laboratory animal personnel spends time &#34;talking to, handling, training and grooming animals&#34;7.
Some handling procedures cause stress, especially for terrestrial prey animals such as rabbits. Examples are being cornered, restrained (i.e., grabbed by a human, held in the arms of a human, or immobilized by a device), and lifted (i.e., picked up from the floor) for a health inspection or experimental procedures8. A survey focusing on pet rabbits revealed that 57-61% of them struggled when being lifted9,10 and some even showed fear related aggression (i.e., biting)11. This emphasizes the need for gentle handling and training techniques. It has already been demonstrated for other small mammals such as mice that gentle handling techniques decreases anxiety12. Moreover, gentle handling in combination with training reduces stress, anxiety, and depression-like behavior in mice13. In rhesus macaques and chimpanzees, physiological measures related to stress were less affected when they were previously trained on a procedure14,15. It may be assumed that if an animal can choose to participate in the training voluntarily, it can actively engage in and control the situation, which improves the animal&#39;s welfare and may also contribute to more robust and reliable scientific data16.
Many animal training procedures are reliant on operant conditioning principles, particularly the use of positive reinforcement and shaping17,18,19. Operant conditioning requires an animal to elicit behaviors that can then be increased or decreased20. There are four common contingencies that are used to describe operant conditioning: positive and negative reinforcement as well as positive and negative punishment21,22. In rabbits, positive punishment is used when a person approaches (undesired stimulus) an animal to move it away from or towards a particular location23. In the context of handling a rabbit on the examination table, an example for negative reinforcement is the removal of pressure (undesired stimulus) on the rabbit&#39;s body when the rabbit keeps still; an example for negative punishment is withdrawing a food reward (desired stimulus) when the rabbit starts struggling. Neither positive/negative punishment nor negative reinforcement are recommended in animal training. Punishment causes negative emotional affects24, such as distress as shown in dogs25. Instead, positive reinforcement should be the method of choice when training animals25,26. Positive reinforcement training (PRT) means that a desirable stimulus (e.g., a food reward) is provided after the animal showed a desired behavior, which increases the likelihood that the animal will display the desired behavior again in the future27. PRT often involves the application of a conditioned reinforcer. A conditioned reinforcer (secondary reinforcer) usually is a neutral stimulus such as a sound, which the animal learns to associate with an unconditioned reinforcer (primary reinforcer, e.g., food reward)28. After the animal displays the desired behavior, the conditioned reinforcer is immediately presented, followed by the presentation of the food reward26. A well-known example of PRT is clicker training, where a clicker is used for producing a click as a secondary reinforcer26.
The training should follow predictable patterns, which is achieved by designing a training protocol. The training protocol describes how a goal behavior can be reached step by step (i.e., behavioral shaping) while each step is defined by objective criteria and should be easily accomplished26. If the training follows a particular protocol, it is possible that the laboratory animal staff alternates with training the animals. Important to note is that the personnel needs to know the principles of learning and training techniques in order to correctly apply them26. Animal training often incorporates behavioral shaping. Behavioral shaping by hand was deliberately used for the first time by Skinner and colleagues in the early 1940s29. In contrast, in previous work, Skinner rather focused on &#34;making small changes in the physical environment in order to implement a program of successive approximation&#34;29. In an article published in Scientific American, he explained that reinforcing a behavior increases the likelihood that the behavior will be repeated which &#34;makes it possible to shape an animal&#39;s behavior almost as a sculptor shapes a lump of clay&#34;30. Since then, shaping has experienced growing importance for captive animals29,30,31,32. For reproducing and improving shaping protocols, it is crucial to assess the protocols in a systematic and objective manner33,34,35,36.
A recently published systematic review and meta-analysis by Pfaller-Sadovsky et al. found that conditioned reinforcement was most frequently studied in horses and dogs, though cats, cattle, fish, goats, and non-human primates were also investigated37. In laboratory animal science, PRT is widely used for non-human primates38,39,40 and dogs41,42. In addition, pigs43, goats44, and even small mammals, such as mice13 and rats45,46,47, were successfully trained using a clicker in the laboratory setting. To the authors&#39; knowledge, the introduction of PRT for laboratory rabbits has only been rarely reported yet48; existing peer-reviewed literature published several decades ago is outdated and does not meet the criteria of modern ethical research. Although few non-peer-reviewed pieces of information suggest that rabbits can be successfully trained using PRT49, there seems to be a lack of knowledge on suitable training protocols for rabbits that are housed in laboratory animal facilities. The underlying reasons may be varying. For instance, training techniques used may not be described in detail in research articles or data are generated for regulatory research and therefore not being published. Moreover, rabbits may not be the main focus of refinement research since other species such as mice and rats are more frequently used in research. Since rabbits can mask stress and discomfort extremely well, their negative emotional state when being handled may often remain unrecognized. The size of a rabbit allows humans to restrain these animals for procedures while other species such as non-human primates may not be held with the hands for similar interventions without being seriously injured50. However, these facts should not prevent the development or sharing of PRT protocols for laboratory rabbits.
Regarding the potential benefits of PRT, first attempts were made to fill the knowledge gap on suitable training protocols for laboratory rabbits. To facilitate routine procedures such as handling, health inspections, and weighing, training protocols were developed and their feasibility for female New Zealand White rabbits was evaluated. Comparisons were carried out to assess whether the alternative handling technique described in the training protocols was less aversive than the conventional handling technique and whether the training protocols can serve as refinement measures. Figure 1 demonstrates the time schedule of the present study.
Animal maintenance
Thirteen female New Zealand White rabbits were obtained from commercial sources (arrival at animal facility at the age of approximately seven weeks of age). The animals were free of all viral, bacterial, and parasitic pathogens listed in the FELASA recommendations51. Protective clothing and equipment (gloves, masks, hair nets) were worn when conducting the steps of the present protocol. A group of six and a group of seven animals were housed together in a floor pen of 2.8 m &#215; 2.8 m (floor housing), respectively. The floor was covered with fine and flaky wooden bedding material (autoclaved). Moreover, nesting material and paper wool were scattered on top of the bedding. Three plastic houses with two openings each (37 cm &#215; 60 cm; height 30 cm) and a plastic tunnel (length: 58 cm; diameter: 16 cm) served as shelters. Additionally, enrichment items such as wooden gnawing blocks, hay and willow balls, and snack balls filled with food pellets, that fall out when moving the ball, were provided. The rabbits had free access to tap water, pelleted food, and autoclaved hay as well as straw. Further information on food and enrichment items are listed in the Table of Materials. The animal room was cleaned on Thursdays.
The rabbits were maintained under standard conditions (room temperature and relative humidity given as mean &#177; standard deviation: 20.3 &#177; 0.4 &#176;C and 50 &#177; 5 %) on a light:dark cycle of 12:12 h (artificial light with additional daylight incidence). A radio was turned on 30 min before the beginning of the light phase for approximately 8 h to habituate the rabbits to ambient noises.

作者聚焦：训练实验动物进行轻柔、无压力的处理

训练实验兔改进常规饲养程序

We intend to improve animal welfare for as long as animal experiments are carried out. It is our moral responsibility and duty to ensure that the laboratory animals suffer as less as possible. This also includes caring for and handling the animals in the gentlest way possible.For instance, by using positive reinforcement training. Our research suggests that gentle handling techniques and positive reinforcement training reduces stress in laboratory animals. researchers are currently confronted with the requirements to train their laboratory animals.However, this is extremely challenging for someone without any previous experiences in animal training. Therefore, our protocol provides step-by-step instructions that should allow everyone to get the first positive experiences in training a rabbit. First of all, it is animal friendly.We compared the behavior of the rabbits after they were picked up in a transport box, or with the conventional handling techniques. The transport box was less aversive and therefore, could serve as an effective refinement measure.

Watch this Scientific Journal Video about Author Spotlight: Training of Laboratory Animals for Gentle and Stress-Free Handling at JoVE.com

已经制定并证明，涉及对兔子常规饲养程序进行积极强化的训练方案是成功的，因为兔子在每周的一般检查中可靠地展示了经过训练的行为。与传统的处理技术相比，用运输箱捡起兔子的厌恶程度要小。

Non-aversive handling and training techniques for laboratory animals are required to facilitate experimental and routine husbandry procedures, improving ...

只要进行动物实验，我们就打算改善动物福利。我们的道德责任和义务是确保实验动物尽可能少地遭受痛苦。这还包括以最温和的方式照顾和处理动物。例如，通过使用积极的强化训练。我们的研究表明，温和的处理技巧和积极的强化训练可以减轻实验动物的压力。研究人员目前面临着训练实验动物的要求。然而，对于没有任何动物训练经验的人来说，这是非常具有挑战性的。因此，我们的协议提供了分步说明，应该让每个人都能获得训练兔子的第一次积极体验。首先，它是动物友好的。我们比较了兔子在运输箱中被捡起后的行为，或者用传统的处理技术来做。运输箱不那么令人厌恶，因此可以作为一种有效的改进措施。

training-laboratory-rabbits-to-refine-routine-husbandry-procedures

Research

JoVE Journal

Behavior

Training Laboratory Rabbits to Refine Routine Husbandry Procedures

Non-aversive handling and training techniques for laboratory animals are required to facilitate experimental and routine husbandry procedures, improving both animal welfare and scientific quality. Clicker training was utilized to develop training protocols for rabbits to refine stressful routine husbandry procedures usually associated with lifting (i.e., being picked up from the floor)/restraining (i.e., being held in the arms of a human) them. Thirteen female New Zealand White rabbits were trained over three weeks. All rabbits learned the predefined goal behaviors: they followed the target stick, jumped onto the weighing scale, entered a transport box, and reared while placing their front paws onto the trainer&#39;s hand. In addition, ten animals jumped from the floor onto the sitting trainer&#39;s lap and allowed the trainer to lift their paws off the surface while sitting on the trainer&#39;s lap. For some individuals, the protocols had to be adapted by additional interim steps. At the end of the training, the rabbits reliably showed the expected goal behaviors, even after short and long training breaks. With few exceptions, a familiar person other than the trainer could elicit the goal behaviors from the rabbits (generalization), though further sessions were required for generalization. In the voluntary approach test, the rabbits preferred interacting with the trainer in the 1st trial but spent as much time with an unfamiliar person as with the trainer in the 2nd trial. The behavioral observations suggested that picking the rabbits up with the transport box, as described in the protocol, instead of restraining them with the scruff of their neck and lifting them on the arm, was less aversive. All in all, the training protocols were feasible and can serve as a refinement strategy in laboratory animal facilities. In the interest of animal welfare, the training protocols should be applied wherever possible.

Training protocols involving positive reinforcement for routine husbandry procedures in rabbits were developed and proved to be successful, as the rabbits reliably showed the trained behaviors at the weekly general examination. Picking the rabbits up with the transport box was less aversive than the conventional handling technique.