What do associations and dissociations between face and object recognition abilities tell us about the domain-generality of face processing?
What do associations and dissociations between face and object recognition abilities tell us about the domain-generality of face processing?
复制标题
人脸和物体识别能力之间的关联和分离告诉我们关于人脸处理的领域通用性的什么信息？
DOI：
10.1080/02643294.2017.1414691
复制
发表时间：
2018
影响因子：
3.4
通讯作者：
Eimer M
中科院分区：
文献类型：
--
作者：
Eimer M
患有先天性或发育性面孔失认症（CP/DP）的人发现很难识别面孔。这种赤字有多具体？Geskin和Behrmann（G&B）对合作原则的研究进行了广泛的回顾，提出了三个主要观点。大部分患有CP的个体也表现出（有时是轻微的）物体识别障碍。面部和物体识别缺陷的严重程度似乎与个体相关。但是大约三分之一的CP没有表现出任何物体识别受损的证据。鉴于CP中面部和对象处理之间的这种复杂的关联和分离模式，G&B在得出有关底层系统的模块化或域通用性的任何明确结论时都保持适当的谨慎。然而，他们观察到，面部和物体识别问题之间的关联似乎比解离更常见，这使他们认为，在大多数情况下，可能涉及一个共享的域一般识别系统。证明这种关联的存在是重要的，需要仔细调查产生这种关联的因素。在这篇评论中，我将讨论为什么个体面部和物体识别障碍之间的关联和相关性并不一定意味着底层系统是领域通用的。G&B审查的数据来自面部和物体识别的行为测试，从这些测试中得出的推论总是可以在方法论上受到质疑。其中一个问题是，对于哪种测试最适合评估CP患者的面部识别障碍，目前还没有普遍的共识。目前正在努力开发一套标准化的CP诊断标准，可在未来的研究中一致应用（例如，巴顿和Corrow，2016）。为了证明CP中面部和物体识别缺陷之间的联系或分离，同样重要的是要商定一套类似的测试程序来评估物体识别能力。在这里，我们面临着一个明显的困难，即把人脸和物体识别测试等同于与人脸和物体识别的核心机制无关的一般因素。正如G&B所指出的，这些测试在对低级视觉机制、记忆和决策相关过程的要求上可能有所不同，尽管这些问题可以通过开发新的人脸和物体识别测试来克服，这些测试尽可能地等同于这些要求。这样的例子是剑桥汽车记忆测试（CCMT; Dennett等人，2012），其被设计为匹配广泛使用的剑桥面部记忆测试（CFMT）的格式。但还有其他因素，如专业知识对视觉处理的影响或特定物体类别吸引注意力的能力，即使不是不可能，也很难完全匹配。在解释行为测试揭示的面孔和物体识别缺陷之间的关联或分离时，考虑这些测量工具引入的系统性偏差的可能性总是谨慎的。为了最大限度地减少这种偏见的风险，我们不仅需要从精心设计和匹配的行为测试中获得更多数据，还需要对完整和受损的面部和物体处理进行不同的测量。例如，眼动追踪研究可用于评估CP如何在视觉上探索面部和物体，并可识别可能特定于或可能不特定于面部的非典型凝视模式（例如，Schwarzer
Individuals with congenital or developmental prosopagnosia (CP/DP) find it hard to recognize faces. How face-specific is this deficit? The extensive review of research on CP by Geskin and Behrmann (G&B) addressed this question, and produced three main insights. A large proportion of individuals with CP also show (sometimes mild) object recognition impairments. The severity of face and object recognition deficits appears to be correlated across individuals. But about one third of all CPs do not show any evidence of impaired object recognition. Given this complex pattern of associations and dissociations between face and object processing in CP, G&B are appropriately cautious in drawing any firm conclusions about the modularity or domain-generality of the underlying systems. However, their observation that associations between face and object recognition problems appear to be more common than dissociations leads them to suggest that a shared domain-general recognition system may be involved in most cases. Demonstrating the existence of such associations is important, and the factors that produce them need to be carefully investigated. In this commentary, I will discuss reasons why associations and correlations between individual face and object recognition impairments do not necessarily imply that the underlying systems are domain-general. The data reviewed by G&B come from behavioural tests of face and object recognition, and inferences drawn from such tests can always be challenged on methodological grounds. One problem is that there is as yet no general agreement as to which tests are best suited to assess face recognition impairments in individuals with CP. Efforts are currently under way to develop a standardized set of diagnostic criteria for CP which can be applied consistently in future investigations (eg, Barton & Corrow, 2016). To demonstrate links or dissociations between face and object recognition deficits in CP, it is equally important to agree on an analogous set of test procedures for the assessment of object recognition abilities. Here, we face the obvious difficulty of equating face and object recognition tests with respect to general factors that are unrelated to the core mechanisms of face and object recognition. As noted by G&B, such tests may differ in their demands on low-level visual mechanisms, memory, and decision-related processes, although these problems could be overcome by developing new face and object recognition tests that equate these demands as closely as possible. On such example is the Cambridge Car Memory Test (CCMT; Dennett et al., 2012) which was designed to match the format of the widely used Cambridge Face Memory Test (CFMT). But there are other factors such as the effects of expertise on visual processing or the ability of particular object categories to engage attention that are harder if not impossible to match fully. When interpreting associations or dissociations between face and object recognition deficits revealed by behavioural tests, it is always prudent to consider the possibility of systematic biases introduced by these measurement tools. To minimize the risk of such biases, we not only need more data from carefully designed and matched behavioural tests, but also different measures of intact and impaired face and object processing. For example, eye tracking studies can be useful to assess how CPs visually explore faces and objects, and can identify atypical gaze patterns that may or may not be specific to faces (eg, Schwarzer
遗传性面部失认症的凝视行为
DOI：
10.1007/s00426-006-0068-0
发表时间：
2007
期刊：
Psychological Research
影响因子：
--
作者：
G. Schwarzer;S. Huber;M. Grüter;T. Grüter;Cornelia Gross;Melanie Hipfel;I. Kennerknecht
通讯作者：
I. Kennerknecht
影响因子：
14.5
作者：
Eimer, Martin;Gosling, Angela;Duchaine, Bradley
通讯作者：
Duchaine, Bradley