Visual agnosia and imagery after Lissauer

Abstract

The year 2021 marks the 130th anniversary of the untimely death of Heinrich Lissauer (1861-1891). In his thirty years of life, Lissauer managed to put together an impressive number of contributions to neurology and neuroanatomy. Most influential is his famous distinction between apperceptive and associative forms of visual agnosia. It is perhaps less well-known that in the same article, Lissauer outlined a model of possible dissociations between visual perception and visual mental imagery. Drawing on Hermann Munk’s animal experiments, Lissauer proposed that complete destruction of occipital visual cortex would provoke both perception and imagery deficits, whereas its deafferentation resulting from subcortical white matter damage would only affect visual perception, and leave visual memories unimpaired. This proposal resonates with the present debate on the neural bases of visual mental imagery.

Full text

Visual agnosia and imagery after Lissauer
Paolo Bartolomeo1
Author affiliations:
1 Sorbonne Université, Institut du Cerveau - Paris Brain Institute - ICM, Inserm, CNRS, AP-
HP, Hôpital de la Pitié-Salpêtrière, F-75013 Paris, France
Correspondence to: Paolo Bartolomeo
INSTITUT DU CERVEAU
HOPITAL PITIE-SALPETRIERE
47 BOULEVARD DE L’HOPITAL
CS 21414
75646 PARIS CEDEX 13
E-mail paolo.bartolomeo@icm-institute.org
Running title: Moving Lissauer forward
Keywords: Visual perception; visual mental imagery; white matter disconnection; ventral
temporal cortex

2
2
The year 2021 marks the 130th anniversary of the untimely death of Heinrich Lissauer (1861-
1891). In his thirty years of life, Lissauer managed to put together an impressive number of
contributions to neurology and neuroanatomy. Most influential is his famous distinction
between apperceptive and associative forms of visual agnosia. It is perhaps less well-known
that in the same article, Lissauer outlined a model of possible dissociations between visual
perception and visual mental imagery. Drawing on Hermann Munk’s animal experiments,
Lissauer proposed that complete destruction of occipital visual cortex would provoke both
perception and imagery deficits, whereas its deafferentation resulting from subcortical white
matter damage would only affect visual perception, and leave visual memories unimpaired.
This proposal resonates with the present debate on the neural bases of visual mental imagery.
In 1890, Heinrich Lissauer, then an assistant in Carl Wernicke’s Breslau clinic,
provided a detailed description of a patient with “mind blindness” (Seelenblindheit) (English
translation in Ref. 1). Gottlieb L., an 80-year-old salesman, had suddenly become “quite
incapable of visually recognising the most common objects, although he could recognise
everything by touch or hearing” (Ref. 1, p. 160). The following year, Sigmund Freud would
introduce the modern label “agnosia” for these disorders of recognition. Lissauer’s patient had
a dense right-sided homonymous hemianopia with macular sparing, and reasonably preserved
visual acuity in his intact field. Reading was impossible, but he could write fluently and without
errors.
On the basis of extensive behavioural testing, Lissauer concluded that his patient
“perceived many things without comprehending them” (p. 160). In Lissauer’s theoretical
framework, the patient had normal conscious awareness of sensory impressions
(“apperception”), but could not associate the content of apperception with stored knowledge.
Lissauer thus outlined here the famous dichotomy between apperceptive and associative

3
3
agnosias (although he did not see this distinction as an absolute one, and did not expect to find
pure examples of either form of agnosia).
Can patients with visual agnosia use their “mind’s eye” to conjure up mental images of
the objects they cannot recognise? Visual mental imagery designates our ability to mentally
visualize objects that are not in our direct line of sight (Fig. 1). For example, most of us can
state from memory whether the (typical) red of cherries is darker or lighter than the red of
strawberries, or whether or not Monna Lisa is staring at us from Leonardo da Vinci’s painting.
In visual mental imagery, information from semantic memory (such as the typical appearance
of cherries and strawberries) is modeled to simulate a specific perceptual problem (compare
the hues). Some patients with visual agnosia lose this ability in parallel with their perceptual
impairment; others, however, show dissociations of performance, with impaired perception but
preserved mental imagery2. Concerning Gottlieb L., Lissauer noted that “[f]urther discussion
is required to decide whether the patient really had a clear visual memory or visual imagery or
whether his perception of common objects was related to concise mental pictures. As is well
known, visual imagery is a variable characteristic of which few really definitive personal
reports, such as Charcot’s famous patient, are available. Only gross deficits in visual imagery
lend themselves to clinical investigation. Although confirmation of such deficits would be of
great theoretical importance, in practice they are very difficult to study. A patient can be asked
to describe from memory the shape and the colour of a given object, or of an animal or a plant.
In many instances our patient was able to give satisfactory descriptions. For example, he
described both in words and by indicating the relative sizes what an eel looked like, a plaice, a
swan, an apple, a pear, a plum, and a cherry. There was therefore no good evidence of a deficit
in his ability to visualise mentally” (p. 166).
Concluding his case report, Lissauer discusses his patient’s pattern of performance in
relation to contemporary models of visual processing. He states that “it is accepted that visual

4
4
agnosia is caused by moderately extensive damage to the occipital cortex. It is a fact that during
the process of recognising a sensory impression that part of the cortex that has to effect
recognition is activated first. There then follow a series of associative processes which elicit
the various determining memory images relating to the object concerned. To activate these
images it is necessary for the excitation to spread from the perceiving part of the cortex to the
whole of the cortex in order to elicit throughout a specific and finely tuned reaction. This is the
process which in our case must have been mediated by the transcortical tracts of the visual
cortex. A blockage of transmission along these tracts [...] would prevent a linkage between
perception and those associations which are normally necessary for the process of recognition
and this would result in visual agnosia” (p. 186).
Lissauer’s discussion here is grounded on the theoretical models and empirical evidence
of his time. His boss Wernicke maintained that modality-specific cortical areas, such as the
visual cortex, contain ‘memory images’ related to sensory experiences. On the other hand,
Hermann Munk (1839-1912) had coined the term ‘mind blindness’ (Seeleblindheit) to describe
the deficits of visual recognition that occurred in dogs by partially damaging the posterior part
of their occipital cortex3. Munk considered that the operated dogs had forgotten the meaning
of previously familiar visual objects. However, they remained able to avoid obstacles and
navigate, hence the term mind blindness. Larger occipital lesions provoked instead total
(cortical) blindness.
Lissauer makes then an important distinction between cortical and subcortical (white
matter) damage. “We have so far considered only one visual field as if the hemianopic field
had been totally eliminated. This elimination would be complete if the existing loss of the right
visual field was the result of the destruction of the visual field itself, i.e. if this was a case of
visual agnosia and cortical hemianopia. In that case there would be no replacement in the
hemianopic hemisphere for the loss of memory images reported in other cases of visual

5
5
agnosia. It is different, however, if the hemianopia is of subcortical origin. In that case memory
images of that hemisphere would be available to the conscious mind, unimpaired, despite an
existing visual agnosia for all visual perceptions” (p. 187, emphasis added).
In other words, direct damage to the visual cortex would also eliminate visual mental
images. However, if the visual cortex is only deafferented by white matter damage, then
perception would be affected, but the deafferented cortex could still sustain visual mental
imagery. Lissauer believed that this last case applied to Gottlieb L.’s pattern of performance.
Importantly, here Lissauer offers a principled explanation for associations and dissociations of
disorders of perception (agnosia) and of visual mental imagery. Specifically, Lissauer’s words
imply that visual mental imagery results from top-down activation of the primary visual cortex.
These considerations resonate with the current debate on the neural bases of visual
mental imagery. The currently dominant neural model of visual mental imagery stresses the
importance of early visual cortex for this ability4, in line with Lissauer’s proposal. Contrary to
this view, however, we know today that brain-damaged patients with damage restricted to the
occipital cortex typically have perfectly normal visual mental imagery abilities and
phenomenology2. On the other hand, consistent with Lissauer’s ideas, white matter
disconnections can determine dissociations between perception and imagery deficits. For
example, Madame D., a patient with bilateral, predominantly white matter damage at the
borders between occipital and temporal cortex5, had severe deficits of visual recognition for
forms, colours, letters and faces. Nevertheless, she could conjure up high-definition mental
images of all these items6. White matter damage severely perturbed her visual perceptual
processing, while completely sparing her visual mental imagery, possibly sustained by top-
down activity in the intact temporal cortex. In sharp contrast, patients with more anterior
lesions in the temporal lobe, especially in the left hemisphere7, may show deficits of visual
mental imagery (today often labelled as acquired aphantasia8). Consistent with this abundant

6
6
neuropsychological evidence, a recent meta-analysis of functional MRI studies in normal
participants9 demonstrated the engagement of the left ventral temporal cortex in visual mental
imagery; Bayesian analysis excluded a role for occipital cortex.
Lissauer’s model of visual cognition was based on Munk’s early hypothesis that all
visual processing occurred in the occipital cortex. It can benefit today from a more refined
knowledge on the large-scale circuits subserving visual identification, spanning from the
occipital cortex to the temporal pole, and receiving important contributions from more dorsal,
fronto-parietal networks10. More articulated models of visual mental imagery are now
available, based on large-scale circuits including anterior, ventral and mesial temporal cortex,
as well as fronto-parietal networks of attention and working memory9. The theoretical
landscape emerging from these considerations is certainly reminiscent of Lissauer’s distinction
between cortical and subcortical lesions in the determinism of perceptual and imagery deficits.
However, recent evidence suggests that the crucial cortical region here is not the primary visual
cortex, but high-level visual areas in the ventral temporal cortex. Thus, Lissauer’s 1890 model
of visual knowledge still makes a lot of sense today, provided that its neural nodes are pushed
forward along the occipito-temporal visual pathway.
Funding
The work of the author is supported by the Agence Nationale de la Recherche through ANR-
16-CE37-0005 and ANR-10-IAIHU-06, and by the Fondation pour la Recherche sur les AVC
through FR-AVC-017.
Competing interests
The author reports no competing interests.

7
7
References
1. Lissauer H, Jackson M. A case of visual agnosia with a contribution to theory. Cognitive
Neuropsychology. 1988;5(2):157-192. doi:10.1080/02643298808252932
2. Bartolomeo P, Hajhajate D, Liu J, Spagna A. Assessing the causal role of early visual
areas in visual mental imagery. Nat Rev Neurosci. Sep 2020;21(9):517.
doi:10.1038/s41583-020-0348-5
3. Finger S. Origins of neuroscience: a history of explorations into brain function. Oxford
University Press; 1994:1 vol. (XVIII-462 p.).
4. Pearson J. The human imagination: the cognitive neuroscience of visual mental imagery.
Nat Rev Neurosci. 2019;20(10):624-634. doi:10.1038/s41583-019-0202-9 PMID -
31384033
5. Bartolomeo P, Bachoud-Levi AC, Thiebaut de Schotten M. The anatomy of cerebral
achromatopsia: a reappraisal and comparison of two case reports. Cortex. Jul
2014;56:138-44. doi:10.1016/j.cortex.2013.01.013
6. Bartolomeo P, Bachoud-Lévi AC, de Gelder B, et al. Multiple-domain dissociation
between impaired visual perception and preserved mental imagery in a patient with
bilateral extrastriate lesions. Neuropsychologia. 1998;36(3):239-249.
7. Bartolomeo P, Bachoud-Lévi AC, Chokron S, Degos JD. Visually- and motor-based
knowledge of letters: evidence from a pure alexic patient. Neuropsychologia.
2002;40(8):1363-1371.
8. Zeman A. Aphantasia. In: Abraham A, ed. The Cambridge Handbook of the Imagination.
Cambridge University Press; 2020.
9. Spagna A, Hajhajate D, Liu J, Bartolomeo P. Visual mental imagery engages the left
fusiform gyrus, but not the early visual cortex: A meta-analysis of neuroimaging
evidence. Neurosci Biobehav Rev. Mar 2021;122:201-217.
doi:10.1016/j.neubiorev.2020.12.029
10. Bartolomeo P, Vuilleumier P, Behrmann M. The whole is greater than the sum of the
parts: Distributed circuits in visual cognition. Cortex. Nov 2015;72:1-4.
doi:10.1016/j.cortex.2015.09.001

8
8
Figure 1 Pablo Picasso using his remarkably vivid visual mental imagery to paint without
any external model. Photograph taken during the shooting of “Le Mystère Picasso” (1955),
by Henri-Georges Clouzot. Photo Edward Quinn, © edwardquinn.com