One very late night during my flirtation with law school, I sat at my desk scanning habeas corpus petitions and wiping sleep from my eyes. I’d been up for at least 36 hours and, while my face broadcasted my exhaustion, my mental lucidity had remained impressively intact, or so I thought. At some point, I looked across the room and noticed that my jewelry stand had moved from one side of my dresser to the other. After a few seconds of confused staring, the wrought iron stand took a turn for the anthropomorphic, and moved back across the dresser. I realized that what I saw couldn’t have been happening and yet, the object crawled on. So I crawled, too, right into bed, to escape my madness and the roving tchotchke.
Unsurprisingly, I woke up the next morning to find a standstill jewelry stand. My brush with surreality was simply a hallucination, a perception of something that doesn’t exist outside the mind. In terms of objective reality, perceiving and hallucinating differ in unambiguous ways. But the relationship between the brain functions underlying the two acts of awareness is less cut-and-dry.
When we correctly perceive objects and phenomena in our surrounding environments, the brain draws on two sources of information: expectations of the environment based on prior knowledge and visual stimuli, or incoming sensory information. We use our prior knowledge to interpret the stimuli, which is otherwise too ambiguous to make sense of. A lot of objects and people resemble one another; a lot of voices sound alike. So, we use knowledge to differentiate among the similar and create coherent images from abstract visual signals.
Mental disorders are often explained as a product of some neural chink, but one group of UK psychologists is instead framing hallucinations as a product of normal brain function that’s out of balance. When we see things that don’t exist, they argue, the brain is relying too heavily on prior knowledge to interpret incoming sensory info. These psychologists began testing their abstract concept in a recent study, published in the journal PNAS, that explored how psychotic and non-psychotic people see images before and after gaining knowledge that could skew their perceptions.
The study involved one “clinical” group of mental health patients with low-level but measurable psychosis and a second group of “healthy” volunteers evaluated on their proneness towards psychosis. Hallucinations are one of two main psychosis symptoms; the other are those altered and irrational states of belief known as delusions. Psychosis isn’t itself a condition, but rather a feature common to psychiatric diseases including psychotic and affective disorders, and paranoid states.
Volunteers viewed a series of two-tone images and had to identify which were 3D. The images were generated from full-color photos but appeared as abstract black-and-white “patches” to anyone who hadn’t seen the original photo templates. Exposure to the templates, however, let viewers make out coherent 3-D objects embedded inside the two-tone images, supplying the knowledge needed to construct meaningful pictures. Volunteers repeated the task twice but only saw the 3D-ifiying templates before the second viewing.
Across the board, all volunteers did a better job distinguishing 3D images after seeing the templates. But the psychotic group, as a whole, improved more significantly. Additionally, within the non-psychotic group, those who'd exhibited greater proneness towards psychosis also showed greater improvement between the first and second image viewings than the less-psychotic volunteers.
What does this mean? Well, going back to the hypothesis, visual hallucinations are marked by over-reliance on prior knowledge to shape incoming sensory information. In this case, the photo templates were the “prior knowledge” used to interpret ambiguous shapes. Peeking at them enabled volunteers to construct coherent 3D images. But, not all volunteers made use of them with equal success.
In this context, heightened sensitivity to the impact of relevant knowledge translated to more success on a task dependent solely on applying that knowledge. In the real world, the same tendency, the researchers wrote, might result in “anomalous perceptual experiences,” otherwise known as hallucinations.
Combined, the experiments “provide evidence to suggest that early psychosis and psychosis proneness is associated with a shift in visual processing that favors prior knowledge over incoming sensory evidence,” wrote study authors
In general, people with psychiatric disorders fare worse than non-disordered people on these sorts of perceptual tasks. So, it’s notable, researchers wrote, that psychosis alone would predict high scores. Furthermore, the psychosis-performance link showed up not only in clinical patients, but also in people who’d been evaluated for psychosis proneness, leading researchers to suggest that increased reliance on prior knowledge is a “fundamental trait that contributes to the emergence of psychosis rather than a reflection or consequence of the psychotic state.”
To eliminate the possible influence of delusions on volunteers’ performance, researchers separately compared evaluated performance with respect to both types of psychotic symptoms. The psychosis-performance link remained when they isolated hallucinatory tendencies, but disappeared when did the same for delusional proneness.
Does flipping through a book of well-hidden optical illusions constitute a psychosis test? That’s probably a stretch, but I still know where my afternoon is headed.