Working memory enhancement with transcranial direct current stimulation

Healthy volunteers

A total of 21 healthy volunteers between the ages of 18 and 45 years were recruited through convenience sampling. The sample was sufficient to identify any difference in n-back performance with an effect size of Cohen’s d of 0.8, at an alpha error of 0.025 (correcting for two outcome measures – reaction time and accuracy of 2-back) with a power of 90%. Subjects with a current or past history of any neurological or psychiatric disorder, pregnancy, developmental disorders, presence of implants, seizures, and family history of any psychiatric disorder (including dementia) in first-degree relatives and use of any psychotropic medications/drugs that are likely to interact with tDCS effects were excluded from the study. The study was conducted after a detailed description of the procedure was given to the subjects, along with a video demonstration, to obtain written informed consent. The study was approved by the Institute’s Ethics Committee (No. NIMHANS/EC [BEH.SC.DIV.] 12^th MEETING/2018, dated April 24, 2018).

Study protocol

N-Back task

This task comprised three components, zero-back, one-back, and two-back tasks, and was presented in e-prime^© stimulation presentation software (v3.0, Psychology Software Tools). The subjects would be presented with a pseudo-randomly arranged sequence of stimuli (one of the digits from 0 to 9 in each trial) in a block. In the zero-back task, the subject had to press the “Yes” button whenever “0” was presented as a stimulus and then had to indicate “No.” In one back, the target stimuli were the one which is the same as (repetition of) the one preceding it. In two-back, the patient had to press the “Yes” button if the stimulus (digit) was a repetition of the digit presented two trials earlier.^[24] Two separate versions of the task were used for every subject before and after intervention with tDCS in a counterbalanced order. In summary, the N-back task was administered immediately before the tDCS and repeated immediately after. The tDCS was administered for 20 min in a single session. Accuracy and reaction time were calculated for each type of trial and used for further analysis.

Tolerance to tDCS

A structured adverse-effect questionnaire with a severity rating from 1 to 5 and likely attribution to a tDCS rating of 1–4 was used to evaluate the tolerance of the device.^[25]

tDCS procedure

tDCS was delivered using an indigenously developed standard transcranial electric current device (WISER tES Neuromodulator). Rigorous guidelines were followed in the administration of tDCS to maintain safety and consistency in the procedure.^[26] The cathode was placed over the right orbitofrontal cortex (r-OFC) with electrode axes horizontally oriented. The anode was placed over the L-DLPFC with vertically oriented electrode axes. L-DLPFC was identified using the Beam F3 software (http://clinicalresearcher.org/F3/calculate.php). r-OFC was determined by locating the area inferior to FP2 and F8 using the universal 10/10 electroencephalogram (EEG) electrode placement system. Single session tDCS was administered to each participant for 20 mins, with 2 mA of current using 5 × 7 cm electrodes.

Statistical tests

The accuracy and reaction time values were calculated using MATLAB^© (ver r2012b The MathWorks, Inc). Wilcoxon signed-rank test was applied to compare the changes in N-Back performance with tDCS using IBM^® Statistical Package for the Social Sciences^® v23.

Tolerability

All patients completed a single session of tDCS without any major adverse effects. Among all the side effects, burning sensation, itching sensations, and skin redness were commonly noted, all at mild intensity and none requiring discontinuation due to intolerance. Neither of the subjects reported intra-session or post-session neck pain, scalp pain, sleepiness, nor any phosphenes [Table 1].

Most of the subjects reported the side effects in the mild intensity category. Exceptions to this were headache (n = 3) that was reported in moderate and severe and very severe intensity; trouble in the concentration (n = 2) after tDCS was noted as moderate and severe by two subjects; discomfort (n = 3) was majorly noted to be of moderate intensity [Table 1].

Effect on cognitive task

Wilcoxon signed-rank test showed a significant decrease in the reaction time, specifically in two-back performance across the subjects (z = 2.016, P = 0.04, median = 735.59, inter-quartile range = 686.56 – 936.02). No significant changes were observed in zero-back and one-back performance for both response time and accuracy [Table 2 and Figure 1].

Close

Figure 1:

(a) Bar graph depicting the percentage accuracy for zero back (ZB), one back (OB), and two back (TB) at baseline (Pre_PA) and post-transcranial direct current stimulation (tDCS) (Post_PA). (b) Bar graph depicting the mean reaction time (MRT) for ZB, OB, and TB at baseline (Pre_PA) and post tDCS (Post_PA). PA: Performance accuracy

Export to PPT

WM performance with N-back

The two-back component in the n-back assesses the WM of the individual, whereas zero-back and one-back primarily evaluate the visual processing and attentional components of executive functions. There was a significant improvement in the average reaction time in two back after a single session of tDCS. This finding is in alignment with other studies.^[27] that have reported that anodal DLPFC stimulation with tDCS plays a role in the enhancement of WM tasks. As the accuracy at baseline was more than 90% in almost all the participants, a ceiling effect would have set in, and further improvement could not be evaluated. Furthermore, there is a limitation to the effect of a single session of tDCS, and to examine accuracy-related effects, it may be better to study cognitive performance with multisession tDCS.

Our, results are in alignment with previous studies^[27,28] since there is a degree of similarity in the experimental design; however, a contradiction to some other studies^[21] where task accuracy has been found to improve but not the response time. However, it is worth mentioning here that there are several aspects related to other studies that differ remarkably from our study. Primarily, the study design of our experiment involved tDCS administration during resting state and n-back tested before and after the session, while few studies^[21] evaluated the online effects by evaluating the performance during the tDCS session. Since the immediate effect of tDCS is proposed to occur through mechanisms such as axonal modulation and synaptic doctrine.^[29] Our interest lies in the post-session effects rather than intrasession effects. These effects may have better translational applications.

The most noteworthy strength of the present study is the data in support of the utility of tDCS in the Indian population. In this study, tDCS has been put to evaluation under the variable functions in addition to providing sufficient data on its tolerability and safety. Although a small sample size limits the study, the preliminary findings are encouraging and underline the need for larger systematic studies. However, one major limitation of this experiment is that a cognitive task precedes and succeeds an intervention like tDCS within a short period, thereby introducing practice effects. To minimize such effects, the study implemented different versions of n-back before and after tDCS. Furthermore, perhaps conducting a study with a larger sample size, with a study design involving more tDCS sessions and exploring the same in a blinded controlled trial, would be better evidence in the application of tDCS for enhancing WM functions in the healthy Indian population.

A major limitation of this study is the small sample size. Nonetheless, it has added valuable evidence regarding the effects of tDCS on cognitive performance and its minimal side effects, which may encourage greater participation by healthy volunteers in the future. Another limitation is the absence of a controlled trial, but the findings from this study could be used to inform future controlled trials exploring the effects of tDCS on healthy volunteers. In addition, since most subjects were recruited using a convenience sampling technique within the campus, it was challenging to assess the influence of years of education on cognitive performance. However, as each subject was observed in a pre-post tDCS design, it allowed for a straightforward interpretation of tDCS effects on the N-back task. The study acknowledges the lack of a sham component but sets up a platform for exploration of the same in a controlled design in the future.

Overall, this pilot study has been able to successfully comment on the usefulness and safety of tDCS in the healthy population. This study has also given evidence of the investigative potential of this neuromodulation technique in examining the role of L-DLPFC in WM functions. Although the results of this study need to be tested in a more controlled design and the long-term effects of tDCS on WM need to be explored, it is reasonable to say that tDCS is a safe technique that has a promising role in addressing cognitive processes in the Indian population. In light of the evidence of enhancement of WM response time in healthy volunteers, and its tolerability and safety, tDCS may be studied more systematically.