Training-induced plasticity of multitasking in everyday-like motor behavior

日常生活中的许多活动都涉及感觉运动和认知多任务处理。经常被引用的例子是开车，在超市购物和在忙碌的交通中过马路。先前的研究表明，多任务处理能力随着年龄的增长而下降，这可能会损害老年人在日常生活中的安全和独立性。目前尚不清楚应如何设计培训制度，以切实有效地利用培训所产生的多任务处理的可塑性，从而促进老年人参与日常活动。理解这个问题及其可能的解决方案的一个主要障碍是，我们几乎所有关于多任务处理的知识都来自典型的实验室范例，这些范例在两个重要方面与正常生活不同。首先，他们同时或交替管理不超过两项任务，而日常活动通常涉及不断变化的任务序列。其次，他们管理的任务没有实际目的，而日常生活通常是由目的驱动的。为了克服这些方法上的障碍，我们在这个项目的第一阶段开发和评估了两个新的和更现实的实验范式，用于研究多任务处理。在虚拟现实环境中，参与者要么开车，要么穿过一条忙碌街道，同时对不断变化的模拟日常活动的加载任务序列做出反应;加载任务使用不同的感觉方式，参与不同的认知过程，需要不同的运动输出。我们发现，与年龄相关的缺陷也出现在这些现实的多任务模式中，并且当加载任务需要注视变化和并行视觉处理时，它们会显着增加。SPP第二阶段的研究目标1（RO 1）将确定我们的现实范例中的多任务处理性能是否比只涉及两个并发或两个交替任务的经典范例更好，相等或更差。RO 2将探索一种新的训练机制，它建立在现有的关于行动控制的中央机制的理论基础上。这一制度考虑到，这些机制的现有能力对所有人来说并不相同，因此，不同的人应以不同的方式接受培训。我们将比较这种个性化训练与传统的固定训练的好处。RO 3将研究现实多任务处理的神经基础。为此，我们将记录参与者在现实或经典范例中的大脑激活。总之，这项研究应该扩大我们对现实多任务处理及其适应性可塑性的认识，并应该突出老年人多任务处理技能培训的新途径。

Many activities of daily life involve sensorimotor and cognitive multitasking. Often-cited examples are driving a car, shopping in a supermarket and crossing a street in busy traffic. Previous research documented that multitasking skills decline with advancing age, which could compromise older persons’ safety and independence in everyday life. It is still not clear how training regimes should be designed to effectively and efficiently exploit the training-induced plasticity of multitasking and thus to facilitate older persons’ participation in everyday activities. A main obstacle in understanding the problem and its possible solutions is that nearly all our knowledge about multitasking comes from typical laboratory paradigms which differ from normal life in two important ways. First, they administer not more than two tasks concurrently or alternately, while everyday activities usually involve an ever-changing sequence of tasks. Second, they administer tasks with no practical purpose, while everyday life is usually driven by purpose. To overcome these methodological barriers, we developed and evaluated in phase 1 of this project two new and more realistic experimental paradigms for the study of multitasking. In a virtual-reality setup, participants either drive a car or cross a busy street while concurrently responding to an ever-changing sequence of loading tasks which simulate everyday activities; the loading tasks use different sensory modalities, engage different cognitive processes and require different motor outputs. We found that age-related deficits emerge in these realistic multitasking paradigms as well, and that they are dramatically high when loading tasks require gaze changes and parallel visual processing. Research objective 1 (RO1) for the second SPP phase will determine whether multitasking performance in our realistic paradigms is better, equal or worse than that in classical paradigms that involve only two concurrent or two alternating tasks. RO2 will explore a new training regime, which builds upon existing theories about the central mechanisms of action control. This regime takes into account that the available capacity of those mechanisms is not the same for all individuals, and different persons therefore should be trained in different ways. We will compare the benefits of such personalized training to those of conventional, fixed training. RO3 will investigate the neuronal underpinning of realistic multitasking. To this end, we will register participants’ brain activation while they engage in realistic or classical paradigms. Taken together, the proposed research should expand our knowledge about realistic multitasking and its adaptive plasticity, and should highlight new avenues for the training of older adults’ multitasking skills.