The “thin man” phenomenon: imperfect filling in of visual space
Normal people do not perceive an “empty” region in the portion of their visual field corresponding to the physiological blind spot. Instead, the visual system perceptually fills in the blind spot with information surrounding it. Similar types of perceptual filling in also occur in patients with pathological monocular and binocular homonymous scotomas.1 2 Furthermore, stabilising the image of a spot on the retina leads to perceptual filling in of the spot with the surrounding background.3 Perceptual filling in of stabilised images can also be demonstrated for colour, brightness, and texture.4-10 Safranet al have previously shown that a spatial distortion of images occurs in the field surrounding a scotoma.11 They have now shown that the filling in phenomenon itself is not precise (see p 137, this issue). They report two patients with right inferior homonymous paracentral scotomas resulting from ischaemic brain insults in whom field defects were markedly smaller when tested with an Amsler grid than with a tangent (Bjerrum) screen. In addition, after both patients fixated another person’s face or neck for 5–10 seconds, the other person’s left shoulder (corresponding to the region of the field defect) appeared narrower than the right shoulder. Safran et al call this perceptual alteration “the thin man” phenomenon. Some patients with this alteration are aware of it; others are not, at least not until they are specifically questioned about it.
Filling in of a pathological homonymous scotoma is apparently related to expansion of receptive fields in the visual cortex,12 13 with the responsible neural mechanisms localised in retinotopic visual areas.3 The time required for this filling in to occur depends in part on the size, shape, and location of the scotoma. It appears to reflect the time required for figure ground segregation to fail rather than a slow spread of a surface feature from one region of the visual field into another,3 and it may be related to horizontal connections between pyramidal neurons in the extrastriate cortex.14
The findings reported here by Safran et alraise numerous questions with regard to the “thin man” phenomenon. Does it occur with homonymous field defects on both sides of visual space or only with right sided defects? Does it occur only for homonymous paracentral scotomas or for more peripheral scotomas as well? Does the size of the defect influence the severity of the phenomenon and in what way? We look forward to further work by Safran and his colleagues as well as others for the answers to these and other questions. In the meantime, we agree with Safran et al that patients who complain of blurred vision, particularly after a known or suspected stroke, should be carefully tested for an homonymous paracentral scotoma, the clue to which may be the patient’s spontaneous or queried observation of apparent asymmetry of a symmetric complex image in a particular region of visual space during the clinical examination.
