What Processing Is Impaired in Apperceptive Agnosia? Evidence from Normal Subjects
Total Page:16
File Type:pdf, Size:1020Kb
What Processing Is Impaired in Apperceptive Agnosia? Evidence from Normal Subjects Shaun P. Vecera and Kendra S. Gilds University of Utah Downloaded from http://mitprc.silverchair.com/jocn/article-pdf/10/5/568/1758424/089892998562979.pdf by guest on 18 May 2021 Abstract ■ Visual agnosia is a neuropsychological syndrome charac- jects. Sensory-deªcit accounts of the syndrome predict that terized by a failure of object identiªcation. Apperceptive ag- degrading visual processing would make normal subjects per- nosia, an object identiªcation deªcit caused by damage to early form like patients; grouping-deªcit accounts predict that re- perceptual processes, has been explained by appealing to both moving perceptual organization cues from visual displays damaged early sensory processes and to damaged preattentive would make normal subjects perform like patients. We were grouping processes. Which of these two explanations best able to simulate the behavior of an apperceptive agnosic pa- accounts for the behavior of these patients? We present results tient by removing perceptual organization cues, consistent from two experiments designed to distinguish rival theoretical with a grouping-deªcit account of this syndrome. The implica- accounts of apperceptive agnosia. In our studies, we attempted tions for understanding both apperceptive agnosia and normal to simulate apperceptive agnosia in neurologically intact sub- visual functioning are discussed. ■ INTRODUCTION respond to two types of visual agnosia that were sug- gested by Lissauer. Associative agnosics, Lissauer sug- Among the problems that the visual system must solve, gested, appear to have the ªrst type of damage; they are object representation and identiªcation are perhaps the typically characterized by an inability to recognize ob- most difªcult. Attempts to understand object repre- jects, despite having intact early-level perceptual repre- sentation processes have come from several disciplines, sentations. The quality of these perceptual processes including computational vision (e.g., Marr, 1982), cogni- typically is assessed with a copying task; associative ag- tive psychology (e.g., Biederman, 1987; Tarr, 1995), and nosics often can copy visually presented objects, al- neuroscience (e.g., Tanaka, 1996; Tanaka, Saito, Fukada, & though they are unable to recognize those same objects. Moriya, 1991). Neuropsychological studies of the visual Apperceptive agnosics, in contrast, appear to have the agnosias, disorders in object recognition, also have con- second type of damage; they have damaged early-level tributed to the understanding of object representation perceptual processes. Unlike associative agnosics, apper- processes. One signiªcant contribution of neuropsy- ceptive agnosics cannot copy visually presented objects, chological accounts of object processing has been the suggesting damage to lower-level processing. Although identiªcation of multiple systems/processes that are im- Lissauer’s original scheme is limited in that it cannot portant for object representation and recognition. account for all of the recent agnosic patients reported Most neuropsychological approaches to object repre- (see Farah, 1990; Grailet, Seron, Bruyer, Coyette, & sentation have followed the theorizing of Lissauer Frederix, 1990; Humphreys & Riddoch, 1987; Riddoch & (1890/1988), one of the earliest theorists of the agnosias. Humphreys, 1987), the two forms of agnosia suggested Lissauer suggested that there were two ways in which by Lissauer remain in current neuropsychological theo- object recognition could be impaired following brain ries of object processing. damage. One form of damage was to the object repre- More recent theorists have explored what cognitive sentations themselves, which would prevent recognition processes are impaired in both associative and appercep- because of lost visual memory representations. The sec- tive agnosia. Associative agnosia has been explained by ond form of damage was to earlier perceptual process- postulating a disconnection between intact visual proc- ing, processing Lissauer termed “apperception.” Damage esses and verbal processes (e.g., Geschwind, 1965), as to the process of apperception would prevent recogni- well as by postulating damage to internally stored visual tion because of poor visual input to the otherwise intact memories or templates (e.g., Mesulam, 1985). Appercep- object representations. These two forms of damage cor- tive agnosia has been explained as being caused by © 1998 Massachusetts Institute of Technology Journal of Cognitive Neuroscience 10:5, pp. 568–580 Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/089892998562979 by guest on 02 October 2021 viewing the world through a “peppery mask” caused by within a 20-year period. None of the agnosic cases they multiple scotomas (e.g., Campion, 1987) and as a loss of reviewed were free of either low-level sensory deªcits perceptual grouping processes (e.g., Farah, 1990). or of general mental dysfunction, consistent with Bay’s Within the two broadly deªned forms of visual ag- (1953) analysis of visual agnosia. nosia, apperceptive agnosia is particularly interesting be- The terminology used with the sensory-deªcit ac- cause multiple hypotheses have arisen to explain it count is somewhat vague and lacks a cognitive mecha- despite a great deal of homogeneity among these nism that could have been damaged. This approach neuropsychological patients. Although the multiple ac- could be updated by appealing to spatial frequency counts of associative agnosia might easily be explained channels (see Ginsburg, 1986; Regan, 1982); the visual by the different lesion locations and etiologies in these agnosias could potentially be explained as a selective Downloaded from http://mitprc.silverchair.com/jocn/article-pdf/10/5/568/1758424/089892998562979.pdf by guest on 18 May 2021 patients, apperceptive agnosics, when the syndrome is loss of high spatial frequency channels, which would narrowly deªned (see Farah, 1990), exhibit many simi- result in a loss of perception for ªne detail. Indeed, some larities in both etiology and locus of damage: Most cases patients with lesions to temporal lobe visual areas have that have been reported have been the result of an difªculty performing spatial frequency discriminations anoxic episode, which leads to diffuse damage over (Greenlee, Rischewski, Mergner, & Seeger, 1993), and posterior cortices, most notably the occipital lobes (see apperceptive agnosics have been found to have abnor- Adler, 1944; Benson & Greenberg, 1969; Campion & mal contrast sensitivity functions (Campion & Latto, Latto, 1985; Efron, 1968; Milner et al., 1991; Vecera & 1985). Behrmann, 1997).1 Although the sensory-deªcit account might explain Given this apparent homogeneity of apperceptive ag- some of the neuropsychological data, this explanation of nosia, a single cognitive function may be disrupted in the visual agnosias is problematic because it has failed most, if not all, of these patients. But which cognitive to explain distinctions within the syndrome of visual function? Is the syndrome better explained by suggest- agnosia. Most importantly, a general sensory-deªcit ac- ing it is caused by perception through a peppery mask count does not explain the differences between apper- (Campion, 1987; Campion & Latto, 1985), or is it better ceptive and associative agnosia. Failing to account for the explained as a disruption of perceptual grouping proc- different types of agnosia is problematic because apper- esses (e.g., Farah, 1990; Gelb & Goldstein, 1918/1967)? ceptive and associative agnosics show very different To address this issue, we attempted to simulate apper- patterns of behavior, which would seem to preclude a ceptive agnosia in neurologically normal subjects to single explanation of both types of agnosia. Other expla- identify the affected cognitive functions and to distin- nations have been put forward speciªcally for appercep- guish the various accounts of the syndrome (see Farah, tive agnosia, thereby overcoming this difªculty with the Monheit, & Wallace, 1991; Miyake, Carpenter, & Just, earlier sensory-based accounts. 1994, for similar simulation approaches to neuropsy- chological deªcits). Because the study of apperceptive The Peppery Mask Account agnosic patients has led to multiple explanations of the syndrome, evidence from a converging methodology Given the failure of the sensory-deªcit account to ex- may be required to overcome the limitations of this plain differences between associative and apperceptive previous research. In what follows, we ªrst discuss the agnosia, more speciªc accounts of the agnosias were rival accounts of apperceptive agnosia; we then present needed. One such speciªc account is the peppery mask two experiments aimed at distinguishing the various account of apperceptive agnosia. Based on a patient with accounts. apperceptive agnosia, patient RC, Campion and col- leagues (Campion, 1987; Campion & Latto, 1985) argued that such patients may have multiple infarcts of early Sensory-Deªcit Account visual cortices, and these infarcts cause the visual ªeld One of the earliest explanations of the visual agnosias to be covered (“peppered”) with multiple scotomas of (both apperceptive and associative) was the sensory- varying size and spatial distribution. The result of these deªcit account. Several investigators argued that the ag- scotomas is that perception is degraded as if the patient nosias are nothing more than degraded low-level visual was looking