Research Summary: Can Brain Stimulation Modulate Speech Motor Control?

In the current study, we aimed to test whether tDCS could modulate speech motor control in a young, neurotypical population. The paper is published in the journal Neuropsychologia (open access).

Focussing this project on a young, neurotypical population means that we can collect lots of data in order to understand the best parametres for tDCS (i.e what is the optimal site for stimulation? What is the optimal duration and current density for stimulation?).

We therefore aimed to evaluate the effects of bi-hemispheric (i.e both sides of the brain) tDCS on performance of a complex speech motor learning task. The task involved repeating tongue twisters (e.g “Brad bravely broke Brooke’s brittle blades”). To further understand the effects of tDCS, we also investigated how tDCS affects the excitability of the area of the brain underneath the electrodes: The area of the brain that controls speech movements.

We took measurements of 1) how well participants performed the task and 2) how excitable their speech-motor cortex was, before and after participants received 13 minutes of 1 mili-amp tDCS. That way, we could look at the change in these measures from before the stimulation to after the stimulation.

Sixty participants were assigned to one of three groups: Anodal, Cathodal or Sham stimulation. Based on the results from previous studies, we predicted that:

  1. ANODAL tDCS (positive stimulation) would increase excitability (the amount of activity) of the speech motor cortex and lead to better performance on the speech task.

  2. CATHODAL tDCS (negative stimulation) would decrease excitability of the speech movement cortex and lead to worsened performance on the speech task. (Note that the evidence for this is less clear-cut compared with the affects of anodal stimulation).

  3. SHAM tDCS (fake stimulation) would not change the excitability of the brain nor performance of the task.

  4. There would be a relationship between the change in brain excitability and the change in performance on the task for each of the three stimulation conditions.

Against our expectations, we found that tDCS did not alter performance on the task, nor did it alter the excitability of the brain. In addition, there was no relationship between the change in brain excitability and change in task performance.

So, why not?

Did we not have enough participants to see an effect? Unlikely. Our study had a sample size double that of a previous study using a similar design to ours which did find an effect of tDCS. We conducted further analyses to check how confident we should be in our results: “Bayesian” analyses of our data confirmed substantial evidence in support of the null hypothesis (i.e no effect of tDCS) for both task performance and motor excitability.

Could participants tell whether they were receiving real or sham stimulation? No. After the experiment, participants were asked to guess which group they thought they were in and their guesses did not differ from chance.

Was the stimulation not strong enough? Maybe. We used 1mA, which is at the lower range of typical intensities (~1mA – 2mA). It is possible that we would have seen an effect of stimulation if we had used a higher intensity. However, previous studies show both behavioural and brain excitability changes after 1mA. In addition, previous work from our lab showed no difference in behavioural outcomes when 1mA was compared with 2mA. We wanted to avoid using 2mA tDCS as this intensity can lead to an increased ‘tingling’ sensation on the scalp, which would make it easier for participants to guess which condition they are in.

Was the duration of stimulation not long enough? Unlikely. Numerous reports of tDCS applied to the motor cortex in neurotypical humans show behavioural modulation of movement tasks with stimulation durations of between 10 and 16 min and brain excitability modulations after just 5 minutes.

Was the the behavioural task not sensitive enough? Maybe. A previous study that did find an effect of stimulation used well-known, long tongue twisters, whereas ours were novel and shorter. We can’t, therefore, rule out that these differences made our task less sensitive to any effects of tDCS compared to previous work, but we believe these differences are unlikely to explain our results.

So…. what then?

We believe the most likely explanation is that our participants were neurotypical, young adults (average age 23 years). A previous study that investigated tDCS with a tongue twister task studied older adults (average age 57 years). Another study that investigated language learning found age-dependent effects in that only the elderly group showed task improvement; performance in the group of young neurotypical adults was not modulated by either type of stimulation.

In a similar vein, it is possible that tDCS is most effective when the area of cortex being stimulated functions atypically. For example, left ventral premotor cortex is known to be less active during speaking in people who stutter compared with controls and anodal tDCS over this area lead to increased fluency in people who stutter compared with sham stimulation. Similarly, tDCS led to altered performance on a finger movement task in the non-dominant, but not the dominant hand of neurotypical adults. In our opinion, the negative results for both task and brain excitability in the current study are best explained by the fact that our healthy young adults function optimally, which renders modulation by tDCS ineffective. More research that compares young/older and neurotypical/neurodivergent populations is needed to understand how tDCS can be used optimally in each population.

Whilst we set out to use a neurotypical population in order to understand the parameters of tDCS, with the hope of informing therapy options in disordered or neurodivergent populations, it may be that this approach is appropriate as tDCS may act upon the brain in different ways for each population.