Internal vision

Katie Haylor spoke to Avital Hahamy from UCL, whose recent paper sheds light on some blind people's experience of visual hallucinations...

Avital - We have found that in blind people who experience visual hallucinations, it's what we call the Charles Bonnet Syndrome, the visual areas of the brain are working healthily just like in any other brain of a sighted individual. And this healthy and normal activity gives rise to visual hallucinations. These brain areas are not activated by some external image that they see because these individuals are blind. Instead they're activated by spontaneous brain activity and this internal activity causes hallucinations.

Katie - So why are you looking at this in people who can't see, but are hallucinating?

Avital - The problem is if I ask you to behave spontaneously, I'm pretty sure you won't be able to be able to do so because the mere instruction is a spontaneity killer. So spontaneous behavior has to be something that is induced by you by some internal desire you have. The second reason is that when you behave some signals in your brain would be evoked by the actual behaviour and by the environment and we can't separate those from this spontaneous panic. However, in blind individuals, the visual centres in the brain don't respond to the environment. So what happens in their visual centres is just spontaneous activations. On the other hand, they experience vision and this vision is completely spontaneous. It's completely uncontrolled. They cannot control the content or the onset of these hallucinations. So this is why we chose these individuals to study whether spontaneous brain activity could underlie spontaneous behaviour.

Katie - What experiments did you do?

Avital - We asked these participants to report whenever they experienced hallucinations in the FMRI scanner, which allows us to look at their brain activity noninvasively. And we recorded what they say when they see hallucinations. And after being scanned, we played back these recordings to them and ask them to give as many details as they can about the content of their hallucinations. So for example, there's one participant who hallucinates a face. It's a face she has never seen before when she was able to see, but she repeatedly hallucinates this space. She can describe it in many details. Another participant sees something that he describes as a kaleidoscope, just rapidly altering shapes and colours all across his visual field.

Katie - Is it quite important for this experiment that these people have had sight previously?

Avital - It is very important because when a participant says, I see face, then I can understand what she means. She has seen many faces before. Now, a person who was never sighted before, what we call congenitally blind, if he says he sees a face, then I can't really know what he means by that.

Katie - Okay. So you've got your, in this case, five participants right in this group, and you put them in an FMRI scanner, so you understand which bits of the brain are active during these hallucinations. And you've got quite a detailed account of these hallucinations. What do you do next?

Avital - So now, if you think about it, I have a scenario. I know which images appear, and when. And based on these scenarios, I can build a movie describing the hallucinations. And I take these movies that are made by recording actors or taking images, and I show these movies to sighted individuals while I scan their brains in the FMRI scanner. And I asked them to report what they see, just like the blind individuals did.

Katie - You're trying to get it as close as possible that they're having the same experience, right?

Avital - Exactly. So now I have two groups of participants who experience a very similar stream of visual events, but one group is blind and the other one is sighted. And I want to understand what's different in the way their brain is activated in this experience. Another difference is that in the Charles Bonnet Syndrome group, they're seeing internal content while the sighted controls are seeing content that we show them. So we introduced another condition in which we see internal content, but it is not spontaneous. We asked blind individuals who do not hallucinate to imagine. We tell them when to imagine and they do it. So what they see is not out there in the environment, it's internal, but it is not spontaneous.

Katie - So you've got these three versions of vision, as it were. They're all seeing the same thing or as close as possible. And then you've looked at what their brains are doing. What did you find?

Avital - The first thing we found was that across these three groups and conditions, the same brain areas are working - the same brain areas that are used in all our brains in order to see, this is what we call the visual system. However, the timing in which the visual system is activated is different. In all these cases of non-spontaneous vision, in the sighted controls who see our movies and in the blind controls who are engaged in imagery: first, the instruction is given or the image is shown, and only then the brain reacts to it and we see activation in the visual system. However, in those individuals with Charles Bonnet Syndrome who hallucinate: first, the visual system is activated. And once we see enough activation, once it reaches enough amplitude, only then the participants report to hallucinate.

Katie - You did only have five people in that group. So is that enough to say something meaningful? What can you say from this?

Avital - It is true. Our sample is very small. So for example, we can't say anything based on five participants about the entire populations of people suffering from Charles Bonnet syndrome. We can't say that visual hallucinations are always the result of spontaneous brain activity in the brain. We can say that in those individuals, this is the case. So although there could be other brain mechanisms that give rise to visual hallucinations in the blind, we can identify one of them in our subgroup of participants. We can also use this example as a test case. In these five individuals, we can say that spontaneous brain activity can evoke behavior. So it is indeed possible that spontaneous brain activity could induce behaviour in typical individuals who see.

Katie - What would you say these findings, albeit in five people, might mean for someone who's got Charles Bonnet Syndrome?

Avital - These individuals know they are blind. And they know it doesn't make sense that they see. And for this reason some of them are afraid of being tagged as insane. So they wouldn't necessarily tell their family members or their doctors and it causes a lot of stress. What we found is that the reason they can see is that their visual system is intact. It is not deteriorated. It is not processing other information. It is intact. It does what it knows to do in a sighted brain. It knows to take signals and to interpret them as vision.