How do transcranial direct current stimulation effects on cognition in Parkinson's disease vary by electrode placement and cognitive domain?

Why the electrode position should matter, and what has actually been compared

Transcranial direct current stimulation (tDCS) passes a weak electrical current between two electrodes on the scalp. The electrode placed over the region you want to excite is called the anode; the return electrode, the cathode, sits elsewhere. Because current flows through the brain along a path set by where those two pads are, moving them changes which tissue receives the strongest dose. In principle, then, the same device settings could help one mental skill and do little for another.

Direct comparisons are scarce. One study simulated the current in brain models built from MRI scans of four people with Parkinson's disease and compared an anode over the primary motor cortex (M1) with an anode over the dorsolateral prefrontal cortex (DLPFC) [2]. The motor-cortex placement produced stronger stimulation of the globus pallidus and subthalamic nucleus, deep structures involved in movement. That is a modelling result about current distribution and predicted membrane excitation, not a measured cognitive outcome, and the study did not test cognition at all [2]. It is the clearest available hint that placement changes where the current lands, but it says nothing about which placement suits which cognitive domain.

The cognitive trials that exist mostly use a single arrangement. In one pilot randomized trial, the anode sat over the left DLPFC and the cathode over the right supraorbital region, above the eye, while patients did cognitive training [1]. No second placement was tested, so the design cannot separate the effect of that montage from the effect of the training or of tDCS in general.

Which cognitive skills have been tested, and what the results look like

The pilot trial measured four skills with computer tasks: attention, working memory, reaction time, and response inhibition (the ability to stop an initiated response). Thirty-two patients were split into real stimulation plus cognitive training or sham stimulation plus the same training, with ten sessions of each [1]. Scores improved over time in both groups across all four skills and in daily living activities, and the authors concluded that active tDCS with training could facilitate improvement compared with training alone [1]. Because the trial was small and single-blinded, that conclusion is best read as promising rather than settled.

Pooled evidence paints a flatter picture. A meta-analysis of 12 randomized sham-controlled trials covering 263 patients found no significant difference between active and sham tDCS for attention and working memory, executive function, or memory and learning [3]. The authors also reported substantial variation between studies, and their exploratory analysis suggested small positive effects mainly in younger patients with milder symptoms [3]. A separate systematic review reports improvements in memory and reaction time, but its available summary does not specify which placements produced them [5].

Put together, the domain-specific claims rest on small trials using one frontal placement, while the average across trials is close to zero. That pattern is consistent with benefits concentrated in particular patients or particular setups rather than a uniform cognitive boost, though the evidence cannot yet confirm that explanation.

Does pairing stimulation with cognitive training change the picture?

Several of these studies deliver tDCS during cognitive training rather than at rest, on the reasoning that stimulating a brain region while it is engaged in a task may strengthen that activity [1][4]. A review of this combined approach describes tDCS as a potentially safe and effective strategy for cognitive impairment in Parkinson's disease, while noting that more research is needed before clinical use [4]. The pilot trial is an example of the combination in practice, with stimulation given during 45-minute training sessions [1].

This matters for interpreting placement. If the benefit depends on stimulating a region while it works, then the useful placement is the one matched to the trained skill, not simply the one that reaches the deepest structure. The modelling comparison favoured the motor-cortex placement for reaching deep motor targets [2], but that logic does not transfer automatically to a prefrontal placement aimed at attention or memory. No study in this set tests that match directly.

What a reader should conclude about choosing a placement

For now, the left DLPFC anode with a right supraorbital cathode is the arrangement with the most direct cognitive support in Parkinson's disease, and it has been tested mainly alongside cognitive training and mainly in small trials [1]. The motor-cortex placement has modelling support for reaching movement-related deep brain structures, but no cognitive data [2]. The pooled null results mean that even the better-studied frontal arrangement has not reliably beaten sham across trials [3].

The practical implication is that placement is a plausible lever, not a proven one. A patient or clinician should treat claims that one montage reliably improves a specific cognitive domain as untested, because the head-to-head studies needed to support such claims have not been done. The most useful next evidence would be trials that randomize patients to different electrode positions and measure the same cognitive tasks in each arm.

Sources used in this answer

Related research questions