The Uncertain Outcome of Prefrontal tDCS

Highlights

• Studies using tDCS over prefrontal cortex were reviewed.

• Prefrontal tDCS modulates a wide variety of cognitive functions.

• It is difficult to predict the effects of tDCS due to variability in response.

• Cognitive functions can be modulated in parallel with the same stimulation parameters.

Abstract

Background

Transcranial direct current stimulation (tDCS) is increasingly used in research and clinical settings, and the dorsolateral prefrontal cortex (DLPFC) is often chosen as a target for stimulation. While numerous studies report modulation of cognitive abilities following DLPFC stimulation, the wide array of cognitive functions that can be modulated makes it difficult to predict its precise outcome.

Objective

The present review aims at identifying and characterizing the various cognitive domains affected by tDCS over DLPFC.

Methods

Articles using tDCS over DLPFC indexed in PubMed and published between January 2000 and January 2014 were included in the present review.

Results

tDCS over DLPFC affects a wide array of cognitive functions, with sometimes apparent conflicting results.

Conclusion

Prefrontal tDCS has the potential to modulate numerous cognitive functions simultaneously, but to properly interpret the results, a clear a priori hypothesis is necessary, careful technical consideration are mandatory, further insights into the neurobiological impact of tDCS are needed, and consideration should be given to the possibility that some behavioral effects may be partly explained by parallel modulation of related functions.

Introduction

In 1865, Broca introduced the idea of studying the neural basis of cognitive processes by the anatomical-correlative method [1]. While studying the effect of a brain lesion in his famous patient “Monsieur Tan”, who had a neurosyphilic lesion to the left hemisphere that impaired his language production, Broca concluded that it was possible to infer a causal relationship between a specific brain region and a cognitive function [2]. This discovery ultimately sparked the emergence of neuropsychology, which aims to better understand the link between brain and behavior, and led to a wide interest in the study of patients with various brain lesions. Subsequently, remarkable progress was made using this approach, for example during World War II, where researchers were able to study the effects of focal brain lesions induced by weapons in conjunction with cognitive testing [3].

Despite the numerous and significant insights derived from the “lesion method”, researchers were – and still are – confronted with methodological limitations when trying to ascertain brain–behavior relationships in patient populations. Firstly, lesions are usually large and often encompass multiple brain areas or networks, as they are most frequently acquired through stroke, ischemia, or traumatic brain injury. Secondly, and consequently, multiple functions are often altered simultaneously, inducing substantial variability in the nature and amplitude of the deficits observed in patients with relatively similar and overlapping lesions. Thirdly, patients often suffer from other medical conditions, either pre-existent or consequent to injury, further contributing to the heterogeneity of the studied population. Lastly, it is difficult to conduct a study with a large sample of patients with overlapping lesions, which has led to numerous case studies and findings that have been difficult to replicate [4].

The development of non-invasive neuromodulation methods in the early 1980's offered the promise to circumvent many of the methodological caveats associated with the “lesion method”, allowing causal inference in the study of brain–behavior relationship in healthy populations. While repetitive transcranial magnetic stimulation (rTMS) was increasingly used in the mid 1990's to study the influence of so-called “virtual lesions” in different regions of the brain, interest in transcranial direct current stimulation (tDCS) emerged more recently. tDCS involves the induction of a constant low-amperage electric current (usually 1–2 mA) applied to the cortex via surface electrodes positioned on the scalp of the subject that can be used to probe and modulate cortical plasticity in the human cortex [5]. In standard protocols, the “active” electrode is positioned over the region of interest while the “reference” electrode is placed contralaterally over the homologous region or supraorbital area. The current flows from the positively charged anode toward the negatively charged cathode. The effect of tDCS on a specific region is partly determined by the polarity of the stimulation: cortical excitability is thought to be enhanced under the anode, and decreased under the cathode [6].

As with TMS protocols, initial studies using tDCS [6], [7] investigated its effects on motor cortex, mainly because of the possibility to directly measure the increase or reduction of cortical excitability through TMS-induced motor evoked potentials (MEPs). Since tDCS was shown to be efficient in this regard, many studies began to report the impact of tDCS on other brain functions in healthy subjects, such as vision [8], language [9], and learning [10]. The investigation of the method's potential for the treatment of different neurological and psychiatric disorders, such as depression [11], stroke [12], and schizophrenia [13] has also recently arisen. In fact, over the past 16 years, over one thousand papers have been published on the use of tDCS on different brain functions. However, studies investigating the effect of tDCS on cognition have shown a lack of specificity and a relative inconsistency in both the modulatory effects and the choice of tDCS parameters, which has led to a large number of heterogenous results. For example, modulation of the dorsolateral prefrontal cortex (DLPFC), which is often chosen as target for tDCS because of its role in numerous high-order cognitive processes, has been associated with both an increase and a decrease in executive functions [14], [15], [16] and has been suggested to influence – among others – spatial memory [17], verbal fluency [18], risk taking [19] and craving [20].

Therefore, it remains to be determined to which extent tDCS can compensate for obvious limitations to the lesion method. For example, it is debatable whether tDCS can target specific behaviors associated with a given area when the physiologic impact of tDCS itself can vary considerably between subjects. Indeed, the effect of tDCS on a specific brain area will depend on a variety of factors including electrode montage and size, but also according to size and shape of the participant head and fat tissue amount, among others. As a result, the amount of current induced in a given brain area may vary considerably across individuals. Furthermore, the brain region and neuronal populations that underlie a specific cognitive function may also be subject to important variations. Finally, the effects of tDCS for a given brain region are state-dependent and the state of brain activity will differ for different cognitive functions (even if the same brain area is engaged in different functions).

Another, often overlooked issue arises from the fact that stimulation of a given area produces widespread modulation of brain activity, which in turn can affect multiple cognitive functions simultaneously. This can lead to an important problem of interpretation since the observed effect of stimulation could be due to the interaction of several parallel cognitive effects, which are sometimes in opposite directions. To better understand the challenges of interpretation of results of studies using tDCS to modulate dorsolateral prefrontal cortical functions, we undertook a systematic review of the literature. Care was taken to select and compare studies that target the same area and use similar electrode montages. The international 10-20 electrode system areas F3 and F4 were chosen, as they are the most commonly used in tDCS studies of the DLPFC.