Vanderbilt University, USA; Simon Fraser University, Canada.
Vanderbilt University, USA.
Cognition. 2017 Dec;169:1-14. doi: 10.1016/j.cognition.2017.07.014. Epub 2017 Aug 9.
Imagined perspective switches are notoriously difficult, a fact often ascribed to sensorimotor interference between one's to-be-imagined versus actual orientation. Here, we demonstrate similar interference effects, even if participants know they are in a remote environment with unknown spatial relation to the learning environment. Participants learned 15 target objects irregularly arranged in an office from one orientation (0°, 120°, or 240°). Participants were blindfolded and disoriented before being wheeled to a test room of similar geometry (exp.1) or different geometry (exp.2). Participants were seated facing 0, 120°, or 240°, and asked to perform judgments of relative direction (JRD, e.g., imagine facing "pen", point to "phone"). JRD performance was improved when participants' to-be-imagined orientation in the learning room was aligned with their physical orientation in the current (test) room. Conversely, misalignment led to sensorimotor interference. These concurrent reference frame facilitation/interference effects were further enhanced when the current and to-be-imagined environments were more similar. Whereas sensorimotor alignment improved absolute and relative pointing accuracy, sensorimotor misalignment predominately increased response times, supposedly due to increased cognitive demands. These sensorimotor facilitation/interference effects were sustained and could not be sufficiently explained by initial retrieval and transformation costs. We propose that facilitation/interference effects occurred between concurrent egocentric representations of the learning and test environment in working memory. Results suggest that merely being in a rectangular room might be sufficient to automatically re-anchor one's representation and thus produce orientation-specific interference. This should be considered when designing perspective-taking experiments to avoid unintended biases and concurrent reference frame alignment effects.
想象中的视角转换是出了名的困难，这一事实通常归因于想象中的自己与实际朝向之间的感觉运动干扰。在这里，即使参与者知道他们处于一个与学习环境没有已知空间关系的远程环境中，我们也展示了类似的干扰效应。参与者从一个方向（0°、120°或 240°）学习不规则排列在办公室中的 15 个目标物体。参与者在被推到具有相似几何形状的测试室（实验 1）或不同几何形状的测试室（实验 2）之前被蒙住眼睛并迷失方向。参与者被安排面向 0°、120°或 240°，并要求进行相对方向判断（JRD，例如，想象面向“笔”，指向“电话”）。当参与者在学习室中的想象方向与当前（测试）室中的物理方向对齐时，JRD 表现得到改善。相反，错位会导致感觉运动干扰。当当前和想象环境更相似时，这些并发参考系促进/干扰效应得到进一步增强。虽然感觉运动对准提高了绝对和相对指向准确性，但感觉运动错位主要增加了响应时间，这可能是由于认知需求增加所致。这些感觉运动促进/干扰效应是持续的，并且不能通过初始检索和转换成本得到充分解释。我们提出，促进/干扰效应发生在工作记忆中学习和测试环境的并发自我中心表示之间。结果表明，仅仅处于矩形房间中就足以自动重新锚定一个人的表示，从而产生特定于方向的干扰。在设计视角转换实验时应考虑到这一点，以避免意外的偏差和并发参考系对齐效应。
