The ability to predict upcoming events is a core feature of human cognition. Even though predictions are widespread, we do not fully understand how they are computed in the brain.
What are predictive processes, and why are they important to study?
In a broad sense, predictive processes incorporate knowledge from the past to predict future states of the body and the environment, shaping how we perceive the world.
For example, in a noisy cocktail party, we can use the conversation context to comprehend unclear spoken words. At the same time, we are constantly monitoring our environment. This allows us to detect unexpected events, such as a glass breaking on the floor.
To put it simply, reality as we perceive it is built from the interaction between real-world events and predictions based on our past experiences. Altogether, predictive processes are helping us to function efficiently in a rapidly changing environment. Furthermore, neuropsychiatric conditions, like schizophrenia, are associated with disturbed predictive processes.
The AudioPred project will contribute to understanding how auditory predictive processes are computed in the human brain. We will record electrical signals directly from individual neurons as well as groups of thousands of neurons in the brains of patients with epilepsy (implanted with electrodes for clinical reasons) while they listen to streams of expected and unexpected sounds. We aim to delineate how individual neurons and distinct brain areas play different roles in encoding auditory regularities, predicting future sounds, and detecting unexpecting sounds. In contrast to other methods, our methodological approach allows us to find out more precisely where in the brain this processing occurs and how it evolves over time.
The research team encompasses researchers at Front Neurolab - RITMO, UiO, collaborating with colleagues at Oslo University Hospital, University of California at Berkeley, USA, and El Cruce Hospital, Argentina.
Predictions about upcoming stimuli and events are critical for information processing, enhancing perception, motor- and cognitive control needed for fluent conversation, decision-making, and other higher-order cognitive functions. Failure to make accurate predictions is seen in multiple neurocognitive disorders, including schizophrenia, ADHD, and autism. It is currently unknown how predictions are encoded by populations of neurons in different areas in the human brain, and how these regions communicate. Using state-of-the-art intracranial EEG and single-unit recordings in patients with drug-resistant epilepsy performing cognitive tasks, this project aims at providing new insights into the neural mechanisms that support predictions in auditory perception.
These methods, which are rarely available in human neuroscience, will enable a detailed examination of the dynamic spatial (mm) and temporal (msec) interplay within local- and across distant brain regions at a level not possible with the coarser-grained resolution of noninvasive techniques. The use of advanced data processing and modeling techniques will allow us to delineate how hierarchically organized brain areas integrate predictive information about the content of stimuli, how predictive information is encoded by oscillatory mechanisms supporting inter-area communication, the dynamical changes of predictions, the complex interaction between attention and predictions, and the functional role of single-units for predictive processes.
By this multifaceted approach, we aim to elucidate key aspects of predictions, and the neurophysiological mechanisms underlying their formation and updating, that in turn is expected to impact our understanding of aberrant cognition in neurological and neuropsychiatric disorders.