| Summary: | Transcranial Direct Current Stimulation (tDCS) is used increasingly often for testing cognitive hypotheses. It is, however, often ignored that many assumptions regarding how the neural tissue reacts to stimulation have only been verified in the motor domain. Extrapolating these assumptions to the cognitive domain has a set of unique issues which, if ignored, can lead to incorrect interpretations. In this talk I will review a number of common pitfalls in using tDCS for testing a cognitive hypothesis, and discuss some solutions for better-controlled designs. I will address the following issues:
1- Making an incorrect assumption about the nature of the effect: It is often assumed that anodal stimulation has “excitatory” and cathodal stimulation has “inhibitory” effects. Results are then interpreted in light of this assumption. Obviously, if the assumption is incorrect, the interpretation of the results too will be incorrect. I will discuss how the effects of polarity can change as a function of a number of design parameters, and the dangers of making a priori assumptions about the direction of stimulation effects, especially when employing a new design.
2- Choosing an inappropriate montage: By definition, tDCS requires two electrodes, although we are often only interested in stimulating one brain region. Where the second (reference) electrode is placed may not be of theoretical interest to us, but it can have serious consequences for our effects of interest. For one thing the path of the direct current changes as a function of where the reference electrode is placed. This affects the density of the current, as well as the regions that undergo stimulation. Moreover, the region directly under the reference electrode is very likely to be affected by stimulation. Therefore, sometimes the changes in behavior may be due to the unanticipated effects at the reference electrode site, as opposed to the hypothesized effects at the target electrode site.
3- Failing to correctly specify the mechanism that stimulation is targeting. In simple motor tasks such as finger tapping, exciting the motor cortex leads to “better” performance (e.g., faster/more forceful tapping). Linking the activation state of the neural tissue to overt performance in cognitive tasks is more complicated. Excitation of the neural tissue may in fact lead to worse performance, if the targeted neural tissue carries out an inhibitory function in the cognitive algorithm. This is particularly important when studying cognitive operations such as selective attention, which modulate a balance between increased processing of certain items, at the cost of decreasing processing of other items.
For each of the above I will discuss examples, and some possible ways to avoid the problem.
|
|---|