de Gelder et al. discuss a patient — referred to as TN to protect his privacy — who had two sequential strokes that damaged his brain. The parts of the brain that were damaged included the primary visual cortex in both hemispheres — rendering the patient blind. However, the patient could still respond to some visual stimuli through a phenomenon called blindsight.
Even more interesting, the patient could still navigate around visual objects, while reporting being unable to see them and having no memory for what they were.
I have discussed the organization of the early visual system at length before, so I won’t repeat myself. Basically, after visual information is gathered in the retina, it enters the optic nerve and goes to two different sets of places in the brain. The first place is the thalamus after which it goes to the primary visual cortex in the occipital lobes. This cortical stream of information contains most of what we think of when we think of vision: line, contour, color, position, etc. However, there is an additional subcortical of information that goes to a variety of places at the base of the brain — places like the superior colliculus. We think that this stream of information is used to perform the housekeeping functions of vision like orienting the eyes to particular stimuli. We also think this stream may be involved in reacting to emotionally-relevant stimuli — like a snake that you didn’t expect — prior to the recognition of what those stimuli are. You jump back before the visual cortex identifies the object as a snake.
The fact that there are multiple visual streams allows for the curious phenomenon of blindsightedness. Blindsight results from lesions to the visual cortex leaving the retina intact. Because the visual cortex has been destroyed the patient has no conscious awareness of what they see. However, if we were to test them on the location of a blinking light, they may be able to identify its location at greater than chance — while at the same time having no conscious awareness it was there. They can do this is because visual information does not go solely to the visual cortex.
Patients with blindsight are interesting because some of their abilities associated with vision are intact. For example, you can train someone to startle every time they see a light by administering a small electrical shock in conjunction with that light. What is interesting is that this startle memory is maintained in some patients with blindsight. Likewise, in some cases blindsighted patients are not blind to the affect of faces. They can determine whether a face is angry or sad at greater than chance even though they don’t know they are looking at it.
All of these observations are intriguing, but I have never heard about the one presented in this paper. de Gelder et al. document patient TN who despite have bilateral damage to his visual cortices can navigate visually down a hallway with randomly placed impediments without hitting any of them. Don’t believe me? Watch for yourself. (Video from the supplementary info.)
Crazy, right? I have to say that I am just at an absolute loss to explain how it is possible. Given my current understanding of where the information for spatial navigation goes, it shouldn’t be possible. As we understand it, that information went through parts of the brain TN doesn’t have.
The authors speculate:
There is evidence suggesting that blind or blindfolded sighted subjects are able, to some extent, to use the natural auditory obstacle sense to locate a travel path, although auditory guidance is notably inferior to visual guidance and deteriorates markedly when small targets are used to define the travel path . Therefore, the theoretical possibility that TN was guided by echolocation abilities from reflection of sound waves, rather than by unacknowledged visual inputs, cannot be totally ruled out. Nevertheless, this appears as an extremely remote possibility in the present case, as neither TN nor the experimenter following behind him emitted any detected sound during navigation that might have generated sound waves reverberation from the objects laying on the floor. Moreover, the spatial resolution that can be obtained through any echolocation capacity in humans is significantly below that necessary to explain TN’s accurate navigation performance through small objects, as can be observed in the video  and .
Similar findings to these were reported by Humphrey  for a monkey, ‘Helen’, with bilateral striate cortical lesions, which at present remain the only antecedents of our results. Helen successfully avoided various obstacles in an open-field test (as illustrated in a video clip downloadable from the web link: http://viperlib.york.ac.uk/login_pop.asp?filename=Helen3.mpg&thumb_id=2582). The lesion was found not to be absolutely complete in one hemisphere, leading to the surmise that there was a small region of intact vision in the far periphery of the right visual field. This could not have accounted for all the varieties of residual visual functions in Helen, although there perhaps remains some ambiguity in this regard for her obstacle course performance. Given the consonance with the animal research background and the extreme rarity of cases with complete cortical blindness in humans, this striking observation will serve as a take-off point for future studies, when and if other patients will come to light.
If I had to speculate, I would guess that the patient may be co-opting information from their eye motions. Patients with blindsight may have intact eye orientation to stimuli — due to information going through the superior colliculus. Perhaps the patient can feel the eyes orienting to objects and infer their location from that. You have to wonder, though, how that would have enough resolution to do the job. You wouldn’t know whether the object was big or small.
This is often how science — and particularly brain science — goes. You think something goes one way until a finding comes along that says that you were dead wrong. I look forward to further experiments that this lab has planned with patient TN.
Hat-tip: Faculty of 1000
de Gelder, B., Tamietto, M., van Boxtel, G., Goebel, R., Sahraie, A., van den Stock, J., Stienen, B., Weiskrantz, L., & Pegna, A. (2008). Intact navigation skills after bilateral loss of striate cortex Current Biology, 18 (24) DOI: 10.1016/j.cub.2008.11.002