Posterior cortical atrophy: an investigation of scan paths generated during face matching tasks

Abstract

When viewing a face, healthy individuals focus more on the area containing the eyes and upper nose in order to retrieve important featural and configural information. In contrast, individuals with face blindness (prosopagnosia) tend to direct fixations toward individual facial features-particularly the mouth. Presented here is an examination of face perception deficits in individuals with Posterior Cortical Atrophy (PCA). PCA is a rare progressive neurodegenerative disorder that is characterized by atrophy in occipito-parietal and occipito-temporal cortices. PCA primarily affects higher visual processing, while memory, reasoning, and insight remain relatively intact. A common symptom of PCA is a decreased effective field of vision caused by the inability to "see the whole picture." Individuals with PCA and healthy control participants completed a same/different discrimination task in which images of faces were presented as cue-target pairs. Eye-tracking equipment and a novel computer-based perceptual task-the Viewing Window paradigm-were used to investigate scan patterns when faces were presented in open view or through a restricted-view, respectively. In contrast to previous prosopagnosia research, individuals with PCA each produced unique scan paths that focused on non-diagnostically useful locations. This focus on non-diagnostically useful locations was also present when using a restricted viewing aperture, suggesting that individuals with PCA have difficulty processing the face at either the featural or configural level. In fact, it appears that the decreased effective field of view in PCA patients is so severe that it results in an extreme dependence on local processing, such that a feature-based approach is not even possible.

Keywords: eye movements; face perception; posterior cortical atrophy; scan paths; vision disorders.

Figure 1

Anatomical MRIs revealing atrophy in PCA patients. Sample axial slices from the high-resolution T1 weighed image. Cortical thickness maps obtained in the five patients with PCA and a healthy aged control. Maps are overlaid on inflated brains, so as to display thickness of cortex in sulci. A color spectrum is applied to indicate measured thickness (in mm) indicated on the scale bar.

Figure 2

Facial ROIs, ROIs were defined as areas around anatomical features. The face presented is a sample of the ones used in the experiments.

Figure 3

Eye area ROIs. The area around the eyes was further divided into two sets of three smaller ROIs to examine in more detail which areas received the most focus. In one analysis, the x-axis was divided into three vertical ROIs containing (from left to right on the image) the Right Side of the Face, the Midline, and the Left Side. In a separate analysis, the region was divided into three horizontal rows containing (from top to bottom) the Eye Brow, the Eye Lid, and the Eyes.

Figure 4

Patterns of scanning in healthy controls and PCA Patients during the open-view face matching task. (A) Representative scan patterns observed in young and healthy controls and PCA patient RB, AP, and SS. (B) Face ROIs in which PCA patients RB, AP, and SS attend to more (red) and less (blue) than age-matched controls as measured by proportion of fixation duration.

Figure 5

Attention to eye region ROIs during the open-view face matching watch. (A) The mean percentage of duration observed in Aged healthy controls is overlaid on each ROI. (B) The proportion of fixation durations of PCA patients RB, AP, and SS were compared to the means of Aged healthy controls with red indicating areas of greater duration and blue indicating areas of less duration. The White lines indicate the boundaries of the ROIs.

Figure 6

Patterns of scanning in healthy controls and PCA Patients during the restricted-view face matching task. (A) Representative scan paths observed in young and healthy controls and all PCA patients. (B) Face ROIs in which PCA patients attend to more (red) or less (blue) than age-matched controls as measured by proportion of fixation duration.

Figure 7

Attention to eye region ROIs during the restricted-view face matching task. (A) The mean percentage of duration observed in Aged healthy controls is overlaid on each ROI. (B) The proportion of fixation durations of PCA patients were compared to the means of Aged healthy controls with red indicating areas of greater duration and blue indicating areas of less duration. The white lines indicate the boundaries of the ROIs.