Agnosia for scenes in topographagnosia

Abstract

Topographagnosia is most commonly attributed to an agnosia for landmarks. In order to define the nature of this agnosia, we studied a patient with isolated topographic disorientation (TD) after a stroke in the right medial occipitotemporal region. The patient got lost in familiar environments but could readily read and draw maps, describe familiar routes, and provide correct directions. He had normal perceptual test performance and met criteria for topographagnosia rather than for other forms of topographic disorientation. Two ecologically valid route tests assessed the nature of his agnosia. On a familiar route, he could recognize major landmarks. He could not, however, recognize route configurations made up of combinations of visual features each lacking individual distinctiveness. On a test of route learning, he learned landmarks that differed in minor details and could use them to orient himself along a route. He had difficulty, however, recognizing and learning scenes lacking salient landmarks. This agnosia for scenes was worse for semantically-related environments, but improved with semantic knowledge such as street names. In addition, the patient lacked overt prosopagnosia but tended toward semantic errors in the recognition of famous faces. Together these findings suggest that this patient’s inability to recognize a route resulted from an inability of intact perceptual units for scenes, composed of specific visual configurations of individually indefinite features, from accessing stored representations.

Introduction

Topographic disorientation (TD) is the inability to find one’s way in a familiar environments and to learn new routes. The concept of TD dates to Hughlings Jackson (1876) who described environmental difficulty in a patient with a glioma of the right temporal lobe (Jackson, 1932). Since then, clinicians have observed TD in posterior circulation strokes, dementia, and a number of other conditions. Studies examining neuroanatomic correlates of TD have implicated medial occipitotemporal regions, hippocampus, posterior parahippocampal gyrus, and parietal cortex, with right greater than left hemisphere involvement (Barrash, Damasio, Adolphs, & Tranel, 2000; Maguire et al., 1998; McCarthy, Evans, & Hodges, 1996).

The underlying cognitive mechanisms for TD are not entirely clear. TD is dissociable from other cognitive deficits and occurs without other explanatory perceptual or memory problems. Landmark recognition and spatial map construction are two important cognitive operations that assist navigation through familiar surroundings (Paterson & Zangwill, 1945). Accordingly, TD may be largely due to failure to recognize unique landmarks (topographagnosia) or to failure to retrieve spatial maps of an area (topographic amnesia) (De Renzi et al., 1977, Hecaen et al., 1980, Maguire et al., 1996; McCarthy et al., 1996). Questions arise, however, about the role and extent of perceptual or memory changes. Both topographagnosia and topographic amnesia may have at least subtle perceptual changes and both usually involve spatial memory. For example, most patients with TD have both retrograde and anterograde memory difficulty for routes; only mild cases are restricted to impairments in navigating new routes (Habib & Sirigu, 1987).

This study examined a rare patient who lost the ability to find his way in familiar environments, but retained the ability to recognize landmarks. In the absence of agnosia for major landmarks, TD for familiar routes could result from visuospatial difficulties in integrating spatial relationships or an inability to access or retrieve adequate mental maps. Based on this patient’s clinical presentation and history, we hypothesized that problems in recognizing complex visual scenes would prove to be the source of his TD. The study used ecologically valid route tests to determine the cognitive difficulties associated with navigating a familiar route and learning a new route. These tests identified whether problems were more in the recognition of salient landmarks or of non-descript scenes. Further visuoperceptual and visual memory tasks characterized whether the patient’s TD was associated with perceptual problems, such difficulty with visual integration, or with memory problems, such as difficulty with visuospatial learning.