A team from the Parkinson’s Disease and Movement Disorders Group at the Institut de Recerca Sant Pau (IR Sant Pau) and the Neurology Department of Hospital de Sant Pau has described in detail, for the first time, how visual perception is altered in people with Parkinson’s disease who experience hallucinations. The study, led by researcher Laura Pérez-Carasol and Drs. Javier Pagonabarraga and Saül Martínez-Horta, has just been published in npj Parkinson’s Disease, part of the Nature group, and provides one of the most comprehensive characterizations to date of the neurocognitive mechanisms underlying this phenomenon.
Visual hallucinations are one of the most frequent and clinically significant non-motor symptoms in Parkinson’s disease, even at early stages. They usually manifest as brief sensations of presence, figures that disappear when looked at directly, or shadows crossing the periphery of the visual field. They may also include very subtle illusions, such as attributing facial features to everyday objects. Although many individuals recognize that these perceptions are not real, their appearance indicates that the systems responsible for interpreting the environment are beginning to function with reduced precision. Numerous studies have shown that these early episodes increase the risk of developing cognitive impairment and of progressing toward more complex and persistent hallucinations.
According to Dr. Javier Pagonabarraga, “These early manifestations show us that the brain is already struggling to harmonize what it receives from the senses with the internal mechanisms that give meaning to experience. Even if the patient is fully aware that these perceptions are not real, they reflect an imbalance that may intensify over time.”
To understand which processes become dysregulated leading to hallucinations, the researchers studied 93 patients with Parkinson’s disease without dementia. Participants performed a visual categorization task in which they had to quickly decide whether an image corresponded to a face, an object, or an object that merely resembled a face, while their brain activity was recorded using electroencephalography. This methodology made it possible to observe, with high temporal precision, how the brain constructs a percept from its most basic stages through to interpretation and cognitive review.
The analyses showed that patients with hallucinations do not present alterations in vision or in the earliest stages of visual processing. The P100 signal, which reflects this initial phase, is normal across all groups, indicating that they perceive sharpness, contrast, or motion in the same way as individuals without hallucinations. Alterations emerge at later stages, when the brain must organize visual information and assign it meaning.
Detailed analysis of brain activity indicates that the first disruption appears during the structural encoding phase, when the brain interprets the shape of the stimulus and generates a recognizable representation. This activity, reflected in the N170 signal, is markedly reduced in patients with hallucinations, even in those without cognitive impairment. This reduction indicates that initial visual representations are built with less precision, lower internal clarity, and weaker structural robustness, forcing the brain to operate on a more ambiguous and vulnerable perceptual foundation.
In the words of Laura Pérez-Carasol, “This first alteration is essential to understanding everything that follows. If the initial visual representation is already weak or incomplete, the brain loses the stable reference on which it normally builds meaning. It is as if the image arrives ‘half-formed,’ opening the door for other internal signals to take on a greater role than they should.”
On this weakened perceptual base, a second mismatch occurs. Under normal conditions, the brain draws on prior knowledge—memories, learned associations, expectations—only when visual information is ambiguous. However, in patients with hallucinations, this mechanism is activated too early and with exaggerated intensity. The increase in the N300 signal indicates that the system attempts to complete perception prematurely, imposing internal meanings before visual information is fully available. This tendency is even more pronounced in patients with mild cognitive impairment, who rely more heavily on internal processes to interpret external stimuli.
The study also identifies a third alteration in the phase responsible for reviewing and correcting the generated interpretation. This stage, associated with the P600 signal, is weakened in patients with hallucinations, particularly in those who also present mild cognitive impairment. The reduction in P600 indicates a decreased capacity for cognitive monitoring, making it harder to correct interpretations that do not fit reality and facilitating the persistence and conviction of erroneous perceptions.
This chain of alterations—weakened structural encoding, premature semantic activation, and insufficient monitoring—explains why perception loses stability and becomes more prone to generating hallucinatory experiences. As Dr. Pagonabarraga summarizes, “What we see is a progressive loss of coordination between systems that should operate in a precise and sequential manner. When this harmony breaks down, perception ceases to be a stable process and becomes overly dependent on internal signals. That combination ultimately creates fertile ground for hallucinations to emerge.”
Beyond describing how hallucinations originate, the authors emphasize that the pattern of desynchronization between the different stages of visual processing provides valuable insight into the clinical course of the disease. The results prove that these perceptual alterations reflect a progressive shift in cognitive dynamics, particularly in individuals who are beginning to show mild cognitive impairment. This combination—perceptual fragility and cognitive fragility—defines a profile of heightened vulnerability, relevant both for understanding current symptoms and for anticipating future progression.
As Pérez-Carasol explains, “These early signals allow us to see that the brain begins to lose precision and coordination long before more obvious symptoms appear. If we understand this vulnerability from the outset, we can identify patients who will require closer follow-up and adapt interventions before symptoms become more disabling.”
Dr. Saül Martínez-Horta highlights that a clinical reading of these results could help transform care practice. “If we can detect these changes in perceptual and cognitive dynamics early on, we will be able to intervene sooner and tailor treatments and support strategies according to each patient’s profile.”
Overall, the study strongly suggests that visual hallucinations are part of a broader process of neurocognitive dysregulation, in which coordination between perception, meaning, and control becomes increasingly fragile. This pattern, more pronounced in patients with mild cognitive impairment, emerges as a clinically relevant indicator for defining risk groups and guiding personalized assessment and follow-up strategies.
The authors note that having a detailed model of how different phases of perception are altered could transform the way non-motor symptoms of Parkinson’s disease are assessed. Identifying characteristic neurophysiological patterns—such as reduced N170, early enhancement of N300, or decreased P600—opens the door to the development of objective markers capable of detecting subtle changes before hallucinations become clinically evident or before more marked cognitive decline occurs.
For Dr. Javier Pagonabarraga, this approach represents a significant advance. “Understanding this sequence helps us explain hallucinations and allows us to identify early risk signals and propose more precise intervention strategies. We can begin to envision more refined assessments that combine clinical and neurophysiological measures to anticipate symptom onset and provide truly personalized care.”
In addition, the study reinforces the role of IR Sant Pau as an international reference center in research on non-motor symptoms of Parkinson’s disease. The work demonstrates that detailed analysis of perceptual mechanisms can yield key insights into disease progression and opportunities for intervention at early stages. This type of research, funded by the Fundació La Marató de TV3 and the Instituto de Salud Carlos III, defines a line of work aimed at developing advanced diagnostic tools and new therapeutic strategies. This is based on a fine-grained understanding of the brain processes underlying perception and cognition.
Pérez-Carasol L, Martinez-Horta S, Horta-Barba A, Bejr-Kasem H, Marín-Lahoz J, Perez-Perez J, Aracil-Bolaños I, Pagonabarraga J, Kulisevsky J. Disrupted visual-to-semantic dynamics promote visual hallucinations in Parkinson’s disease. NPJ Parkinsons Dis 2025:1–15. https://doi.org/10.1038/s41531-025-01235-1.
