Transcranial magnetic stimulation in cognitive neuroscience – virtual lesion, chronometry, and functional connectivity

Abstract

Fifteen years after its introduction by Anthony Barker, transcranial magnetic stimulation (TMS) appears to be ‘coming of age’ in cognitive neuroscience and promises to reshape the way we investigate brain–behavior relations. Among the many methods now available for imaging the activity of the human brain, magnetic stimulation is the only technique that allows us to interfere actively with brain function. As illustrated by several experiments over the past couple of years, this property of TMS allows us to investigate the relationship between focal cortical activity and behavior, to trace the timing at which activity in a particular cortical region contributes to a given task, and to map the functional connectivity between brain regions.

Introduction

The investigative tools used in science determine the kinds of empirical observations that can be made. Very often, the results produced by new tools in the neurosciences force us to re-evaluate models of brain–behavior relationships and even affect the kinds of questions that are asked. For example, over the past decade, the neuroimaging techniques of computerized tomography, magnetic resonance imaging, positron emission tomography (PET), magneto-encephalography and electro-encephalography (EEG) have shaped the way in which we model behavior. Anatomical neuroimaging techniques have produced ever more detailed descriptions of the extent of lesions produced by brain injury. Combining this knowledge with clinical examination of the affected patients should provide insights into the original function of the damaged brain areas. However such a ‘lesion study approach’ is hampered by the issue of compensatory plasticity, and by the possibility that the disturbance to function may be more, or less, widespread than the anatomical lesion [1]. Functional neuroimaging methods have overcome some of these problems and can demonstrate an association between behavior and patterns of activity in cortical and subcortical structures. Although careful design of experiments may allow us to conclude with reasonable certainty that the correlation of brain activity with behavior is attributable to a causal connection (i.e. that the brain activity causes the behavior), imaging alone will never be able to provide proof of that assertion.

Transcranial magnetic stimulation (TMS) is based on Faraday’s principles of electromagnetic induction. A pulse of current flowing through a coil of wire generates a magnetic field. If the magnitude of this magnetic field changes in time, then it will induce a secondary current in any nearby conductor. The rate of change of the field determines the size of the current induced. In TMS studies, the stimulating coil is held over a subject’s head and as a brief pulse of current is passed through it, a magnetic field is generated that passes through the subject’s scalp and skull with negligible attenuation (only decaying by the square of the distance). This time-varying magnetic field induces a current in the subject’s brain, and this stimulates the neural tissue. In many experiments, single pulses of stimulation are applied. Each of these lasts about 100 μs, so that the effect is similar to stimulating a peripheral nerve with a conventional electric stimulator. To date, the single-pulse technique appears to be completely safe when applied to healthy individuals. It is also possible to apply a series of pulses at rates of up to 50 Hz (this is known as repetitive TMS, or rTMS). This procedure is more dangerous and can cause seizures even in healthy subjects; because of this risk, safety and ethical guidelines must be followed [2^••].

Although studies in animal models and, particularly, in neurosurgical patients have provided considerable insight into the mechanisms of action of TMS 3, 4, 5•, 6•, 7, 8•, our knowledge is still limited 1, 9, 10: we are not yet able to ascertain precisely the depth of stimulation in the brain, nor its spatial resolution; we are not able to determine which neural elements are the most sensitive to stimulation in a particular area of brain; and we are not certain whether all the effects of stimulation are attributable to activity at the site of the stimulus or whether activity spreads through neural pathways to other more distant sites.

One might therefore be tempted to wait for greater insights into the neuronal effects of TMS before applying it widely to studies in cognitive science. We would argue that waiting may be desirable, but is not necessary. As we shall see below, the majority of TMS experiments in cognitive science rely on the fact that magnetic stimulation of an area of the brain disrupts any processing that is going on at the time. If that processing is contributing to behavior, then we would expect to observe deterioration in the performance of that behavior. From such an observation, we can conclude that there is a functional connection between the activity and the behavior. In this scenario, we need not know precisely which elements in the brain were activated by the stimulus. Any artificially induced synchronized activity in a population of neurones will interfere with their function — at this fundamental level, we can probably trust the technique. In fact, by analogy with the synchronised spike-wave discharges of an epileptic focus, it seems probable that a large magnetic stimulus will synchronously excite a population of neurones. These will fire a rapid series of impulses for a few milliseconds, and then the whole activity will be suppressed by a long-lasting period of (GABAergic) inhibition. The whole process may last between 20 and 200 ms, depending on the intensity of the stimulus.

In this review, we shall highlight two of the major potential contributions of TMS studies to our understanding of cognitive neuroscience: the transient disruption of focal cortical activity to establish the causal role and the timing of the contribution of a given cortical region in a behavior, and the application of TMS to the study of functional brain connectivity. What is critical and common to both of these contributions is that they allow us to further our knowledge beyond that which the study of patients can teach us — they allow us to empirically test specific neuropsychologic models and constructs.