Jochen Ditterich - US grants

Successful decision-making in a complex and ever-changing world involves multiple steps; First, identifying decision-relevant information: Second, identifying possibly sensible actions; and Third, deciding on an action that is most reasonable under the current circumstances. To achieve this, our brain needs to select decision-relevant sensory signals, and link the relevant sensory evidence to the action planning process. With funding from the National Science Foundation, Dr. Jochen Ditterich and his research team, are carrying out research on how our brains are able to make decisions on the basis of sensory information. His research is investigating the neural mechanisms that provide the necessary flexibility to deal with a changing environment. In his research, participants carry out a perceptual decision task that requires them to separate decision-relevant from irrelevant information and to switch stimulus-response associations. The investigators are analyzing human decision behavior and are using computational modeling to reveal the properties of the mechanisms that allow our brain to make flexible decisions. In addition, recordings of neural activity from animals performing the same decision task are providing insight into the neural mechanisms that can separate decision-relevant from irrelevant information, and that can switch stimulus-response associations based on the current task demands.

It is a hallmark of our cognitive abilities that our brain can easily adjust to very different situations in which we might find ourselves. From a mechanistic point of view, decision-making in complex and changing situations means that our brain must be able to change quickly how information flows between its different parts. While theories have been proposed how the brain might be able to achieve this, the actual neural mechanisms are surprisingly poorly understood. This study is expected to provide a major advance in our understanding of the necessary brain mechanisms, as well as their limits. Such knowledge will be a prerequisite for developing strategies to improve decision-making in complex situations and for understanding failures of these neural mechanisms in mental disorders, for example, in the case of schizophrenia.

Oscillatory activity in the brain has been known for a long time, but its functional significance is still being debated. It has recently been proposed that neural oscillations play an important role in controlling how information flows in the brain and which ensembles of neurons are able to exchange information, allowing the brain to flexibly adjust to varying task demands. While this view has been supported by recordings of brain activity from particular subsystems, a rigorous test of the ideas requires manipulation of brain activity to establish a causal link between synchronized brain activity and neural communication. While experimental techniques for generally suppressing or enhancing neural activity are readily available, addressing this scientific question requires new technology for manipulating neural activity with a precise timing relationship relative to ongoing neural activity. The goal of this research project is to develop a closed-loop stimulation system that can analyze ongoing brain oscillations in real time and that uses the resulting information to trigger neural stimulation. This allows manipulating neural activity time-locked to ongoing activity at another location in the brain. Sharing this tool with the scientific community is expected to provide novel, fundamental insights into the functional significance of synchronized brain activity in a variety of neural systems, which cannot be obtained with currently available technology. The technique might also find application in neural prostheses and brain stimulation systems for the treatment of neurologic and psychiatric disorders.

The project involves: 1) Developing an algorithm that reliably extracts instantaneous frequency and phase of a dominant oscillatory component from a local field potential and accurately predicts the next occurrence of particular phase angles. 2) Implementing the algorithm(s) in a combination of software and hardware that is fast enough for real-time control of a stimulation device. 3) Validating the closed-loop stimulation technique in vivo and developing applications for studying cortico-cortical and thalamo-cortical communication. Successful development of such a system would remove a major obstacle for testing theories about the functional significance of the timing of neural signals, in particular synchronized rhythmic activity. It would allow going beyond correlational measures and exploring the behavioral (and neural) consequences of artificial manipulation of neural signals contingent on the timing of currently ongoing neural activity. With the help of this technology, a major advance in understanding the role of the timing of neural signals in coding and transmitting information between ensembles of neurons is expected.

Project Summary An appropriate routing of information in the brain based on the current task demands is essential for successfully dealing with a complex, changing environment. Such routing mechanisms seem to be impaired in disorders like schizophrenia. Unfortunately, we currently do not understand how the brain is able to regulate the flow of information. A recent theory suggests that synchronous activity plays an important role. However, it is still fiercely debated in the neuroscience community whether synchronous activity and precise spike timing in cortex have a functional significance and, if so, what role they play. A major obstacle to answering these questions is the lack of suitable experimental techniques for artificially manipulating brain activity with a precise timing relationship to currently ongoing neural activity. Techniques that allow the experimental manipulation of neural activity play an instrumental role in establishing a causal link between brain activity and its functional significance. The goal of the proposed project is therefore to develop a closed-loop stimulation technique that allows the measurement of currently ongoing neural activity in the form of a local field potential, analyzes it in real time, and can trigger a stimulation device like, for example, electrical microstimulation or optogenetic stimulation in such a way that artificially injected neural activity is phase-locked to currently ongoing oscillations. This includes the development of an algorithm that reliably tracks the instantaneous phase of a dominant component of the local field potential as well as finding a suitable implementation of this algorithm for controlling a stimulator in real time. Making the proposed technology available to the scientific community is expected to provide major breakthroughs in understanding the functional significance of synchronous activity and precise spike timing in the brain.