Brain Advance Access originally published online on May 21, 2008
Brain 2008 131(6):1411-1413; doi:10.1093/brain/awn097
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Visual capacity in the hemianopic field following a restricted occipital ablation. By L. Weiskrantz, Elizabeth K. Warrington, M. D. Sanders and J. Marshall (From the Department of Experimental Psychology, University of Oxford and The National Hospital, Queen Square, London, WC1N 3BG, UK) Brain 1974: 97; 709–728.
Professor Weiskrantz, from the University of Oxford, and his colleagues at Queen Square set the scene for their case report in detail of what—drawing on the linguistically economical 19th century terminologies of ‘word-blindness’ and ‘word-deafness’—they call ‘blind-sight’ by reference to the magisterial descriptions of visual field loss based on the analysis of penetrating injuries of the head by Gordon Holmes (1918) and George Riddoch (1917) after the Great War, and by Edward Tauber, William Battersby and Morris Bender (1960) following the Second World War. But these giants of descriptive neurology, restricted to knowledge gained from human, not monkey, neuroscience and limiting their gaze to the gross deficits that follow extensive occipital lesions, may have missed some subtleties of the visual experience that result from damage to the striate cortex. Proceeding with commendable caution, the authors prepare readers for the revelation that the cortex, with lesions restricted to Brodmann area 17, can still see: ‘... voluntary ocular fixation responses can be controlled (albeit rather weakly) by visual stimuli placed in regions of blindness ... even though the patients deny seeing the stimuli ... such an ability (if confirmed) [might] depend on the direct input to the mid-brain from the retina [and] it would be important to examine this ... in patients with striate cortex damage but with only minimal damage to the posterior association cortex, as the latter receives a projection ... from the superior colliculus’. A birth notice of this case had already been posted in the Lancet in an account led by Michael Sanders, clinical neuro-ophthalmologist to the team, who first noticed the unusual visual experiences of D.B. Michael Sanders recalls now that his interest and skills in cortical visual neurology had been honed whilst working with Bill Hoyt at Johns Hopkins Hospital, so he was alert to the possibility of unusual visual phenomena in patients losing Brodmann area 17: quite whether it was he or D.B. who sensed that there was something happening in the hemianopic field is not now so clear but the hint of blind-sight led Elizabeth Warrington, Larry Weiskrantz and Michael Sanders to spend the whole of the next weekend at the National Hospital examining D.B. and so providing the neuro-ophthalmological basis for this classic account. But not all experts were fully persuaded and David Cogan favoured the interpretation that an island of intact vision remained within the field defect in which D.B. accurately reported what was plainly there to see.
D.B. has experienced transient episodes in which headache and vomiting are associated with an enlarging left homonymous scotoma surrounded by a crescent of coloured lights, occasionally with associated left sided sensory symptoms. Their frequency has increased through childhood and, following a typical attack at the age of 25 years, D.B. is left with a fixed loss of field within the scotoma. After cerebral angiography has demonstrated an arteriovenous malformation at the right occipital pole and treatment with methysergide and cervical sympathectomy have not influenced the frequency of attacks, Prof. Valentine Logue removes the lesion in 1973, sacrificing the striate cortex on the medial surface of the hemisphere 
6 cm anterior to the occipital pole. Before too long, it is clear that surgery has relieved the headache at the expense of a dense right homonymous hemianopia that splits the macula, leaving a small crescent of preserved vision peripherally in the upper quadrant (Fig. 1).
|
Whilst looking ahead, a 2° bright spot of light is projected onto the ‘blind’ field at between 5° and 35° horizontal eccentricity and with 4° vertical variation, after which D.B. is asked to guess where it was placed by looking in that direction. Blank projections serve as control observations. Results are based on the steady state position adopted after a few corrective saccades, and the experiment is repeated several times. ‘There was a weak correspondence between target positions and eye position [but only] for stimuli placed between 5° and 25°’.
Then D.B. is asked to point, usually with the ipsilateral arm, at where a spot of bright light, of varying stimulus size, has been projected at between 15° and 90° horizontal eccentricity, whilst keeping his eyes fixed straight ahead. Repeated tests are used and his accuracy in reaching into the intact field is assessed. ‘The correspondence between target position and finger position was striking for stimuli larger than 23' in diameter and requires no statistical demonstration or justification ... [but] with the smallest size the correlation broke down’ (Fig. 2). D.B. does not do so well when guessing the position of blanks. Such errors as he does make in placing real spots of light err towards the point of fixation for stimuli with 45° eccentricity and away from it for those outside this displacement. His best shots are in spotting the location of lights placed at 60° from the midline. Exploration of D.B.'s ability to locate lines—horizontal, vertical or diagonal, and white-on-black or black-on-white—or ‘X’ and ‘O’ symbols show that he is equally adept at ‘seeing’ each so long as these are above a critical size. As the stimuli become smaller, he requires a longer exposure to get it right. Taken together, brief stimuli are more likely to reveal a horizontal than vertical or diagonal line, and each is more sensitive than symbols (Fig. 3).
|
|
Contrast sensitivity is now tested by asking D.B. to discriminate stimuli consisting of vertical bars at and above a spatial frequency that appears homogenous in the intact half-field. This demonstrates ‘acuity’ of 1.9' in the ‘blind’ field compared with 1.5' on the intact side. Whereas D.B. denies ‘seeing’ lines in the blind field that he can nevertheless accurately detect at a spatial frequency not much different from normal, his acuity in the right temporal crescent, where he faithfully reports seeing the stimuli, is only 8.2'. On ‘blind-colour’ the experiments are compromised by shortage of time and poor availability of appropriate filters on the perimeter but ‘ ... the scores are in the direction of indicating at least some residual capacity to differentiate red vs. green, [but] in the absence of more systematic and better stimulus control they cannot be taken as being more than suggestive’.
That D.B. can ‘see’ relatively large stimuli in a field that is absolutely defective to conventional assessments used in clinical practice begs the question of whether he might have sneaked a look at the bright light with his intact field. But his examiners do not observe any such ocular deviation; the exposures are too short for reflex movements or voluntary saccades to have confounded the results; his eye movements are monitored electrophysiologically during one trial; his successes correlate too precisely with size, duration and angular separation of the stimuli; D.B. is kept ‘in the dark’ as to how he is doing during individual trials and he himself alerts his examiners to a false test if the stimulus strays into his intact field. Shown a video recording of his experiments, D.B. is astonished by the success of his forced choices, and insists that he sees nothing. The best insight he can provide is that he may occasionally have experienced a feeling that something smooth or jagged, a stimulus pointing this way or that or no stimulus at all during the blank exposures has been going on. His relatively successful look at the stimulus, by comparison with accurate pointing, reflects elimination of the usual way of looking that involves a gross head movement followed by short-range homing saccades within the residual 30°. And whilst ‘ ... a whole host of questions about the possible properties of "blind-sight" remain to be explored in detail, among them colour, depth, acuity as a function of retinal locus, adaptation, number of stimulus alternatives available, the perceptual constancies – in short all of the attributes of normal "seeing" itself - ... the fundamental question is whether D.B.'s "blind-sight" is merely degraded normal vision or ... a qualitatively distinctive visual capacity’.
Can monkey studies resolve these questions and illuminate the nature of ‘blind-sight’? Total removal of the simian striate cortex with degeneration in the dorsal lateral geniculate nucleus leaves animals slow to learn but with preserved ability to discriminate patterns and reach with reasonable accuracy towards a visual stimulus placed in the blind field, albeit with some loss of the capacity to resolve fine detail—amblyopia. In this context, D.B. is also not normal with respect either to accuracy or acuity in his blind-sighted field. He shows similarities to ‘Helen’, a monkey with bilateral striate cortical lesions studied by Nicholas Humphrey and Larry Weiskrantz over many years who was unable to identify items with which she was familiar despite being able to locate objects throughout the visual fields: ‘in one sense she sees everything, in another sense, nothing’. Therefore, do two visual systems exist in parallel but with different structures and function: one that recognizes, identifies and examines; the other that merely detects and orientates? ‘The present results are a distinct embarrassment to ... the doctrine of "encephalization of visual function" [which] postulate[s] that visual capacity becomes increasingly dependent on cortical structures with ascending phylogenetic status’. In turn, the results obtained in D.B. challenge the anatomical doctrine that a striate lesion produces an absolute scotoma with amblyopic fringes reflecting the gradation of damage to radiations and surrounding cortical tissue; rather, the retention of vision in the blind field suggests that striate cortical damage causes a functional deficit—amblyopia—which increases towards complete loss of luminous flux discrimination as the extent of that damage encroaches upon prestriate and posterior association cortex. On this analysis, D.B. would have a lesion ‘restricted to area 17 in far anterior lower bank of the calcarine fissure, and more posteriorly to involve area 17 together with the surrounding prestriate and additional association cortex’. It seems unlikely as David Cogan had evidently proposed that ‘blind-sight’ can be explained by an island of intact vision in the hemianopic half-field. Rather, visual information can reach the mid-brain and from there access the posterior association cortex, normally also connected to the striate cortex, via the retinotopically faithful colliculo-thalamic projections that are likely also to have orientation sensitivities. Perhaps D.B. reaches the unseen target by accessing these intact collicular retinotopic maps but with restrictions on location and orientation that are exposed by smaller stimuli. And at a retinal level, the acuity displayed by D.B. in response to larger and transient stimuli suggests preferential function of the magnocellular or Y class of retinal ganglion cells that project both to the cortex via the lateral geniculate nucleus and to the colliculus.
D.B. has revealed that it is clearly not sufficient merely to ask a patient if he or she ‘sees’ a stimulus in order fully to explore the retro-chiasmal visual pathway; and—depending on the anatomy and extent of the striate lesion—knowledge of awareness must now be added to the list of potential defects that may dissociate from the crude losses charted by perimetry or confrontation. Significantly, for an era in which cortical plasticity is so much better characterized and understood, Prof. Weiskrantz and colleagues predict—based on observations in monkeys who appear to use their defective visual field more ably when discovering the lost world with feedback from tactile-kinaesthetic information—that patients might also be taught to improve their use of the altered visual world by exploiting blind-sight. But exactly which neuroanatomical pathways subserve this phenomenon remains unclear. Thirty four years on, Holly Bridge and Alan Cowey (to whom the authors pay tribute in their acknowledgements) provide a partial answer by mapping visual pathways in G.Y. to show various novel connections between the contralateral pathway connecting the right lateral geniculate nucleus to MT +/V5 on the affected (left) side, and between the human motion areas of each hemisphere (page 1433, and see commentary page 1414).
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||