Scene Structure and Surfaces: Viewing strategies effects on attention, learning, and memory

视觉搜索是许多任务不可或缺的一部分；当我们在城市中导航、做三明治或修理漏水的水龙头时，我们会搜索地标、物体和工具。传统上，视觉搜索期间的注意机制是使用视觉阵列来研究的，在视觉阵列中，项目在人工的、统一的视野中等间距分布。*我和我的受训者研究注意力控制机制如何在更复杂、更自然的环境中运作。我们发现，使用人工阵列开发的基本理论不容易推广到真实世界的场景，部分原因是场景环境基于知识和经验提供了强烈的偏见和预期。解释这些差异的一种方法是对象和场景如何在空间上联系在一起(Castelhano&Heaven，2011)，而不考虑场景类别或其他语义信息。本文提出的研究将通过我们新引入的理论框架开始填补这一空白。*表面制导框架假定注意力和眼球运动被定向到与目标对象最相关的场景表面。使用这些关联，我们可以将场景划分为相关和不相关的表面区域，并检查注意力如何在场景中分配、信息如何处理以及当预期达到或违反时这些过程如何随时间变化。在三个项目中，我们的目标是研究如何：*1.注意力在整个场景中的分布以及在无关区域中的处理是否受到积极抑制(项目1)，*2.策略随着时间的推移而改变(从利用先前的经验和知识转变为当那些期望未达到时探索环境)以及它们如何影响对场景信息的学习和记忆(项目2)*3.3D环境中的视觉搜索变化，瞬间决策的成本收益比随着人们的走动而改变(项目3)。*我的研究计划的目标是开发一个更详细的理论描述，说明感知信息、一般世界知识和以前的经验是如何结合在一起的，从而允许在现实世界环境中进行高效和有效的视觉搜索。表面制导框架通过产生可检验的假设，使人们能够对视觉复杂环境中的注意机制和信息处理提出新的概念、范例和新的理论理解。*这项拟议的工作对于理解真实世界场景信息的基本认知过程以及理解这些过程如何组合和随时间变化都是具有变革性的。我和我的学员将利用这个新颖的理论框架，作为更全面地理解复杂自然环境中的认知过程的第一步。

Visual search is an integral part of many tasks; while navigating through a city, making a sandwich or fixing a leaky faucet, we search for landmarks, objects, and tools. Traditionally, attentional mechanisms during visual search have been investigated using visual arrays, in which items are equally spaced across an artificial, uniform visual field. ******My trainees and I study how mechanisms of attentional control operate in more complex, naturalistic environments. We have found that basic theories, developed using artificial arrays, do not easily generalize to the real-world scenes, in part because the scene context provides strong biases and expectations based on knowledge and experience. One way to account for these differences lies in how object and scenes are linked spatially (Castelhano & Heaven, 2011), regardless of scene category or other semantic information. The research proposed here will begin to fill this gap, through our newly introduced theoretical framework. ******The Surface Guidance Framework assumes that attention and eye movements are directed to the scene surfaces most associated with a target object. Using these associations, we can divide the scene into relevant and irrelevant surface regions, and examine differences in how attention is distributed across the scene, how information is processed and how these processes change over time when expectations are met or violated. Across three projects, we aim to examine how:***1. attention is distributed across the scene and whether processing in irrelevant regions is actively suppressed (Project 1), ***2. strategies shift over time (from one of exploiting previous experience and knowledge to one exploring the environment when those expectations are not met) and how they affect learning and memory for scene's information (Project 2) ***3. visual search changes in a 3D environment, where the cost-benefit ratio of moment-to-moment decisions change as people walk around (Project 3).******The goal of my research program is to develop a more detailed theoretical account of how perceptual information, general world knowledge, and previous experience is combined to allow for efficient and effective visual search within real-world environments. The Surface Guidance Framework, by generating testable hypotheses, enables the development of new concepts, paradigms and a new theoretical understanding of mechanisms of attention and the processing of information in visually complex environments. ******This proposed work is transformational both for the understanding of basic cognitive processes of real-world scene information and for understanding how these processes combine and change over time. My trainees and I will use this novel theoretical framework as a first step in developing a fuller understanding of cognitive processing in complex, naturalistic environments.