Authors: Lauren Pulling
As part of our Spotlight on neuromodulation, we spoke to Eric Wasserman, head of the Behavioral Neurology Unit at the National Institute of Neurological Disorders and Stroke at the NIH. Dr Wasserman is a clinical neurologist with specialty training and extensive experience in neurobehavioral disorder, and his lab uses brain stimulation techniques and fMRI to investigate brain mechanisms.
In this interview, Dr Wasserman tells us more about his work, uses for noninvasive brain stimulation, and the potential for the field of neuromodulation in the research and treatment of neurological disorders.
Please could you tell us a little about your background and current role?
I am a neurologist with specialty training in clinical neurophysiology and movement disorders. I first got involved in neuromodulation during my fellowship with Mark Hallett in the National Institute of Neurological Disorders and Stroke, where we used transcranial magnetic stimulation (TMS) to explore plasticity in the motor system in healthy people and patients with various disorders. The introduction of repetitive TMS (rTMS) was very exciting because we now had a technique that could cause plastic changes, as well as measure them.
During that time, I also helped Mark George, who was a fellow in the National Institute of Mental Health at the time, design and run the first rTMS trials in depression. This was probably the first, or one of the first, times anyone set out to treat a behavioral disorder by artificially inducing a localized plastic change somewhere in the brain. The approach was crude and the mechanism vague, but it seemed to work and provided a new therapeutic model. Since then, while pursuing my scientific interests, I have maintained an interest in noninvasive brain stimulation, per se, and especially as a way of facilitating cognitive processes. I currently direct a research group focused on combining neuromodulatory techniques with functional imaging and a variety of behavioral interventions to explain and enhance learning, memory and other plastic processes in the human brain.
Your lab is using trTMS techniques to investigate brain mechanisms – can you tell us more about this work?
Let me start with some background to put what we are trying to do in context: in a classical rTMS treatment paradigm, for example treating depression by applying it to the prefrontal area, there is an assumption that the stimulation does something locally to the stimulated cortex, which, through some hypothetical chain of events, results in a behavioral change. We have done a lot of this, testing hypotheses about the functions of various cortical areas by inhibiting them with rTMS and seeing if resulting the behavioral effects are what we think they should be. There are two major problems with this paradigm. First, it involves some risky assumptions: for instance, generally there is no independent evidence of having actually done anything to the target and there is a tendency to confirm these assumptions with the outcome in a circular way. Second, behavioral outcomes – even ‘harder’ cognitive and perceptual ones – are quite variable across, and sometimes within, individuals. One way to solve these problems is to find direct, quantifiable and independent evidence of what’s going on in the brain when neuromodulation is applied and to use that evidence as a surrogate for the ultimate behavioral outcome.
The main thrust in our current TMS studies of learning and other behavioral adaptations is to use the functional connectivity between brain areas, detected with fMRI, to provide a quantitative measure of synaptic changes in targeted pathways. We and others have shown that neuromodulatory TMS doesn’t alter behavior simply by changing the activity of the targeted piece of cortex. It actually doesn’t do much to local activity, but causes significant changes at the circuit level, detectable as altered functional connectivity between distant brain areas. Moreover, these changes can correlate tightly with behavioral outcomes.
Functional connectivity is allowing us to use paradigms completely new to this field, such as a Bayesian, adaptive, dose-finding study to find the minimum number of sessions required to produce a clinically meaningful change in connectivity in a pathway important for visual associative memory. Whereas in the past the only hard readout on the effects of rTMS we had was the motor evoked potential (electrical muscle activity) from stimulation of the motor cortex, other physiological outcome measures will allow us and others to explore the multidimensional TMS dosing parameter space in other brain systems. Recording changes in connectivity in our studies also holds us to mechanistic hypotheses, naming the specific networks we intend to modulate. This is analogous to the markers of ‘target engagement’ now universally required in drug development. When we are wrong about the specifics, it will allow us to find new hypotheses.
We have two major projects involving TMS at the moment: one focuses on describing the brain and behavioral changes produced by visuo-motor adaptation to prism goggles, and then attempting to reproduce them by directing rTMS to the same network. Part of this project will be done in stroke patients with hemi-neglect. The other project is on using rTMS to increase connectivity in pathways and networks important for explicit and implicit learning and memory in healthy people and patients with traumatic brain injury.
What are the main uses of noninvasive brain stimulation techniques, such as TMS, at present?