May 13, 2007

Harmless Hallucinations in the Elderly

SCR Feature — alice @ 9:07 pm Print This Post  AddThis Social Bookmark Button


It can be very upsetting, but it may have a fairly mild medical origin: It is called Charles Bonnet Syndrome, and it was first reported 250 years ago, in 1760, by the Swiss philosopher Charles Bonnet (Plummer et al, 2007).When damage occurs to visual nerves in elderly people, they may “see” all sorts of things. For example, a 73-year-old woman saw “inch-long black ants scurrying along the kitchen floor, walls and windows. In desperation, she began spraying insecticide throughout the house… ” Later, when she was brought to the Emergency Room, “floating seahorses and featherless chickens joined the colonies of ants… A Roman chariot, the rider dressed in gold, flashed across the curtain several times. On the ward, tropical vines would grow from the foot of her bed. A man stood with thick brown tree trunks for legs and green tree branches for arms. Nurses’ heads would shrink and then expand before melting into the floor. Brightly coloured fairies carrying wands invited her for walks around the hospital grounds.”A second patient was a 90 year old man, who “woke to find a bright pink handkerchief on the floor of his bedroom. Every time he tried to pick the cloth up it seemed to move out of reach. He followed it from room to room but was distracted by the appearance of rows of bearded human faces with bulging eyes jutting out from the lounge-room wall. Through the window he saw a large brown Ayrshire cow in the front yard and, beyond it, he made out a red sports car racing up the street towards his house. He opened the front door only to be greeted by a trio of policemen dressed like American highway patrol officers toting large guns. Fearing at this point that he was losing his faculties he summoned his wife and an ambulance was called. In the Emergency Department of the local hospital he saw a draught horse pulling a cart loaded with wooden logs.”

CBS is a limited problem, and patients who have it are not going mad, nor are they suffering from dementia.Scary stuff, but no, it isn’t a psychotic episode. It’s not even a difficult-to-manage disease. Charles Bonnet Syndrome (CBS) may be a fairly common side effect of macular degeneration, the age-related degradation of the retinal cells at the back of the eye. Patients with this condition lose their visual acuity in the damaged area, as if the normal “blind spot” (which all of us have in each eye) expands to cover the central area of vision. We can’t see the blind spot under normal circumstances – it is “filled in” with adjacent colors and textures – and CBS patients may fill in their growing “blind spot” (called a scotoma) with visual hallucinations of the most terrifying kind. Fortunately, there are now new treatments for age-related macular degeneration (AMD), if it can be caught early enough before the retinal cells die. Even without curing the underlying AMD, the distressing symptoms of Charles Bonnet Syndrome (CBS) may go away by themselves.

What do you do if an elderly relative develops these distressing visual imaginings? It is important to understand that CBS is not a mental condition. There isn’t anything wrong with the patients’ understanding of reality, and in the case of the 90-year-old man cited above, “He was soon convinced that the visions were imagined and his anxiety was replaced by a sense of amusement.” Five years later the visual hallucinations were gone. That makes CBS very different from mental disorders — people with those disorders don’t learn to laugh at their experiences. CBS victims know they are hallucinating.

So the first rule is the oldest treatment of all: reassurance. CBS is a limited problem, and patients who have it are not going mad, nor are they suffering from dementia. Rather, their visual hallucinations are similar to the visual images we tend to experience during dreams and drowsy “twilight states.”

That is in fact one of the significant features of CBS. It tends to occur during drowsy states. Plummer and colleagues suggest a specific mechanism in the outer visual system. The most plausible hypothesis, they argue, is that neurons in the visual pathway going from the retina to cortex become hyper-excitable, due to the loss of light receptors. This phenomenon has been observed directly in the brain. It is as if the pattern-recognition regions of the cortex are trying desperately to figure out what light patterns are coming into the eye, and are trying out quick, conscious hypotheses to explain the noisy input. The end result is a kind of projection of cortical guesses on the ambiguous stream of incoming stimulation.

Shakespeare had this figured out, in his own way, when Hamlet, Prince of Denmark, tried to make a fool of the old courtier Polonius:

“Ham. Do you see that cloud that’s almost in shape like a camel?

Pol. By the mass, and it’s like a camel, indeed.

Ham. Methinks it is like a weasel.

Pol. It is back’d like a weasel.

Ham. Or like a whale?

Pol. Very like a whale.”

Obviously Polonius was just following Hamlet’s suggestions. This is often called “top down processing.” CBS victim are living in a kind of visual cloud, and their brains can’t quite decide if their eyes are looking at a whale, a camel or a weasel.

Author Information

Bernard Baars is a Senior Fellow in Theoretical Neurobiology at the Neurosciences Institute in San Diego. He is widely known for his Global Workspace Theory and remains at the heart of the study of consciousness. He is also the founder of Science and Consciousness Review.


  1. Plummer, C., Kleinitz, A., Vroomen, P. & Watts, R. (2007) Of Roman chariots and goats in overcoats: The syndrome of Charles Bonnet, Journal of Clinical Neuroscience (in press).


  1. Hi Bernard,

    Thank you for this nice and well-written article about CBS!
    Apart from “treating” or “dealing with” CBS, one can also
    conjecture about binding CBS to sensory substitution, like
    I proposed at

    Sound-Induced Mental Imagery for the Blind

    After all, realistic visual hallucinations with the visual
    content tightly bound to live but crossmodally encoded camera
    views would be very much like… true vision for the blind?
    This would turn a “limited problem” into an real asset – though
    no one knows at present if it can be made to work as hoped for.

    Best regards,

    Peter Meijer

    Comment by Peter Meijer — May 14, 2007 @ 3:19 pm

  2. What an enjoyable article. A little help with one question would be appreciated though. Since light does enter the eye, would it be more accurate to describe this Charles Bonnett Syndrome as an illusion ( a misperception of external stimuli), rather than a hullucination (perception caused in the absence of stimuli). Thanks for the greatt read.

    Comment by Tony Brown — July 12, 2007 @ 1:36 am

  3. Hi Tony,

    Good question. The literature on CBS suggests that the eyes closed or eyes open condition stops the illusory percepts of CBS only in a minority of cases. So it is apparently not just a complex “modulation” of existing/remaining visual input from the eyes. See for instance

    Teunisse R.J., Craysberg J.R., Hoefugels WH., Verboek A.I., Zitman F.G. (1996), Visual hallucinations in psychologically normal people: Charles Bonnet syndrome, Lancet 347: 794-797.

    Comment by Peter Meijer — July 12, 2007 @ 5:49 am

  4. Your article was well done. However, it does not describe how to either counter the ongoing hallucinations or whether to cross stimulate with alternative or logical thinking, games etc. My mother is 97 yr with macular degeneration & cataracts. She recently went from very lucid to the CBS syndrome described above. I would appreciate further suggestions. Regards, Judith

    Comment by Judith Johnson — December 6, 2007 @ 8:48 pm

  5. [...] [...]

    Pingback by Seeing, not believing — February 1, 2011 @ 9:54 pm

RSS feed for comments on this post. TrackBack URI

Leave a comment

Line and paragraph breaks automatic, e-mail address never displayed, HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>