Enhancing Visual Working Memory Anodal Transcranial Direct Current Stimulation
of the Right Dorsolateral Prefrontal Cortex Stefan L. Jongejan
Student 10182241 University of Amsterdam
Under supervision of Ilja G. Sligte and Alexander Laufer Number of words: 4149
2 Abstract
Among human cognitive abilities resides the ability to retain visual information which is no longer in view for later use, this is called visual working memory (VWM). Previous studies have suggested that VWM has maximum capacity of three to four items,
depending on the item complexity, and that visual working memory is to be attributed to the right dorsolateral prefrontal cortex (rDLPFC). Using change detection tasks the performance of VWM van be tested. Recent studies suggest that impairing the rDLPFC by administering transcranial magnetic stimulation (TMS) lowers the performance of VWM. VWM performance might be enhanced by administering anodal transcranial direct current stimulation (aTDCS) to the rDLPFC, as previous studies suggest that atDCS enhances working memory when administered to brain locations attributed to memory. In the present study atDCS was administered to the rDLPFC to explore the possibility of
enhanced functioning of VWM, determined by change detection task scores. These scores will be compared to those resulting from sham stimulation. No effect was shown that administering atDCS to the rDLPFC enhances performance on change detection.
Therefore the outcomes do not support the hypothesis that atDCS might enhance VWM. Whether the lack of effect is due to the impossibility of VWM enhancement through atDCS to the rDLPFC, the misunderstanding of the role of the rDLPFC or procedural deficiencies is to be researched in follow-up studies.
3 Introduction
Among human cognitive abilities resides the ability to retain visual information which is no longer in view for later use, this is called visual working memory (VWM) (Luck & Vogel, 1997). VWM has been shown to have a maximum capacity of three to four items, depending on complexity of these items (Luck & Vogel, 1997). VWM and its capacity can be studied using a change detection task, in which participants are asked to detect and report the possible change of a certain aspect of items which were previously shown (Luck & Vogel, 1997). Previous research by Floel et al. (2004) and Epstein, Sekino, Yamaguchi, Kamiya and Ueno (2002) has shown that the prefrontal cortex (PFC) is involved in VWM. More specific the right dorsolateral prefrontal cortex (rDLPFC) has been shown to be involved in nonverbal memory encoding (Floel et al., 2004) and the encoding of visual item locations (Epstein et al., 2002).
Previous research by Sligte, Wokke, Tesselaar, Scholte and Lamme (2011) involving transcranial magnetic stimulation (TMS) to the rDLPFC has shown to disrupt VWM. TMS to the rDLPFC decreased neural activity of the rDLPFC (Sligte et al., 2011). Sligte et al. (2011) further showed that this disruption of VWM is best shown in participants who feature a high VWM capacity compared to those with a lower VWM capacity.
As decreasing neural activity of the rDLPFC by means of TMS shows a disruption of VWM (Sligte et al., 2011), it might be possible to enhance VWM by means of electrical stimulation. Anodal transcranial direct current stimulation (atDCS) is a non-invasive brain stimulation technique (Rampersad et al., 2014) that has been shown to boost neural activity by enhancing excitability of brain locations to which it is administered (Coffman, Clark & Parasuraman, 2014; Nitsche et al., 2003; Rampersad et al., 2014). Boosting neural activity by means of atDCS might enhance cognitive functioning.
Previous studies have shown that atDCS of the dorsolateral prefrontal cortex (DLPFC) enhances working memory in healthy participants (Fregni et al., 2005) as well as patients with Parkinson’s disease (Boggio et al., 2006), improves speed of accuracy of working memory (WM) performance (Mulquiney, Hoy, Daskalakis & Fitzgerald, 2011),
4 improves performance on digit span forward tasks (Andrews, Hoy, Enticott, Daskalakis & Fitzgerald, 2011) and generally enhances memory when administered to brain locations attributed to memory (Coffman, Clark & Parasuraman, 2014).
Following findings from previous studies suggesting that atDCS increases aspects of WM when administered to the DLPFC it might be that administering atDCS to the rDLPFC enhances VWM, to be shown in change detection task scores. Presently it is unclear whether atDCS administered to the rDLPFC will enhance VWM functioning. Therefore in the present study atDCS was administered to the rDLPFC to explore the possibility of enhanced functioning of VWM, determined by change detection task scores. These scores will be compared to those resulting from sham stimulation. It is expected that change detection task scores during atDCS administered to the rDLPFC will be higher than scores during sham stimulation.
Method Subjects
18 right-handed students from the Netherlands (13 females) in the ages of 18 to 23 years old (average 20, SD 1.64) participated in this experiment. Participants had no current mental and neurological health problems, did not take psychiatric medication or had a family history of epilepsy. Every participant gave their written informed consent for participating in the experiment. Due to personal reasons two female participants
terminated their participation early. Therefore 16 participants (11 females) participated for the entire experiment. Half of these participants were rewarded with academic points, the other participants participated voluntarily.
Materials
atDCS to the rDLPFC was administered by a neuroConn DC-Stimulator with a 3 by 3 cm stimulation electrode positioned at F4, directly superior to the rDLPFC. And a 5 by 7 cm reference electrode positioned on the left side of the forehead directly above the eyebrow. Both electrodes were placed in cushions, which were soaked in a saline solution
5 of 12 grams of salt per liter. A 64 channels EEG cap was used for determining the
position of F4 on the scalp.
The administered current during atDCS was 1 mA. This current was gradually build up from 0 to 1 mA during 60 seconds, followed by 1200 seconds of 1 mA
stimulation and subsequently gradually turned down from 1 to 0 mA during 60 seconds. During the sham stimulation a current was gradually build up from 0 to 1 ms during 60 seconds, followed by 1200 seconds of no stimulation and subsequently gradually turned down from 1 to 0 mA during 60 seconds. For each participant two different keys were programmed in advanced for the atDCS device. One key for the atDCS condition and a different key for the sham stimulation condition. These keys were programmed by an affiliate researcher to ensure double blind administering of stimulation as well as a counterbalanced design in order of conditions. The use of atDCS was approved by the Ethics Comity of the University of Amsterdam.
Figure 1 provides the five subsequent parts of a single trial of change detection task with a changed rectangle orientation during test phase, figure 2 provides a non-changed rectangle orientation during test phase. Each change detection task trial consisted out of a (A) fixation cross on a grey background, during 1000 ms. Followed by a (B) attention cue green fixation cross, during 500 ms. Followed by a (C) memory array of eight black rectangles in per trial varying orientations grouped in a circle around a black fixation cross, during 250 ms. Followed by a (D) retention period, during 1000 ms. Followed by a (E) test array of one black rectangle placed in one of the eight spots around a black fixation cross, during 3000 ms. The rectangle displayed during the test array (E) would either have the same orientation as during the memory array (C), or have chanced orientation 90 degrees. The possibilities for change and no change were equally divided. The position of the rectangle in the test array (E) was evenly randomized to make the probability of being in one of the eight locations equally possible.
A HP Elite 7300 Series MT desktop computer with operating system Microsoft Windows 7 Professional 64-bit Service Pack 1 on an Intel Core i5-24000 CPU at 3.10 GHz
6 with 4 GB of RAM was used for running the change detection task in Presentation version 17.2 with build 08.11.14.
A Dell P2412H, 24 inch, LED monitor with a resolution of 1920 x 1080 x 16 with a 60 Hz refresh rate was used for displaying the change detection task.
A Gigabyte GK-Osmium RED mechanical keyboard was used for registering answers during the change detection task. The ‘z’-key was labeled with a red round sticker. The ‘m’-key was labeled with a green round sticker.
The experiment consisted out of three sessions, with a duration of one and a half hour per session. The change detection task during each session consisted out of three blocks. Each trial lasted 5.75 seconds. Each block consisted out of 160 trials, with 95 seconds break following the first 53 and the second 54 trials. The total duration of each block was 160 x 5.75 s (trial time) + 2 x 95 s (break time) = 1110 seconds.
Figure 1 Change Detection Task Trial with Changed Rectangle Orientation during Test Phase.
7 Figure 2 Change Detection Task Trial with Non-changed Rectangle Orientation during Test Phase.
Procedure
Provided informed consent was filled out and signed by all participants. A document with research specifications with information about atDCS use was read and understood by all participants. Participants could exclude themselves from the
experiment for any reasons at any given time.
During the first session before commencing with the change detection task the electrodes were attached to the participants and atDCS was administered during 60 seconds, building up from 0 to 1 mA during 5 seconds and in the last 5 seconds turning down from 1 to 0 mA. The administering of atDCS during 60 seconds formed a sensitivity test. In case a participant would feel uncomfortable with atDCS the experiment could be terminated. During all three sessions a researcher trained in first aid was present in case of stimulation side effects.
During the first session visual and verbal instructions for the change detection task were provided to the participants before the first testing block to ensure their understanding of the change detection task used in this study. After ensuring the participants understood the change detection task, the first session of three blocks commenced. Participants were asked during each trial to decide whether the rectangle displayed in the test array (E) had changed orientation compared to the orientation during the memory array (C). If no orientation change was perceived by the participants, the participants were asked to press the red labeled key with the left hand. If an
orientation change had taken place the participants were asked to press the green labeled key with the right hand. The original orientation, position, response and reaction time were stored. When participants responded correctly to the test array a sound was played, no sound was played in case of an incorrect response. The first session was a training session to ensure that the participants understood the change detection task and formed an optimal strategy for the task.
8 After the first session participants agreed to an appointment for the second session at least 72 hours later. During the second session participants participated in another three blocks of change detection task following the same procedure as first session. Before the beginning of the second block of change detection task, the atDCS device was activated with the key that was assigned beforehand in a counterbalanced design, with this in half of the participants during the second session atDCS was administered, the other half was administered sham stimulation. Participants were advised to stay for one hour after the experiment in case of possible side effects of atDCS.
The same procedure as the second session followed at least 72 hours later for the third and final session. Resulting in using the remaining key for administering atDCS or sham stimulation during the second block. Participants were once again advised to stay for one hour after the experiment in case of possible side effects of atDCS.
Change detection task correctness scores where required by the sum of correct responses to change and no change during each individual block per participant. The sum was divided by the amount of trials. During each block there were 160 trials. Due to no response or response recorded before the test screen some of the trials were excluded from the analysis.
Results
The present study explored the possibility of enhancing visual working memory (VWM) by applying anodal transcranial direct current stimulation (atDCS) to the right dorsolateral prefrontal cortex (rDLPFC). To determine the effect of atDCS to the rDLPFC on VWM, change detection task correctness scores where acquired.
Overall means of change detection task correctness scores were determined for each of the nine blocks.
9 Mean Change Detection Task Correctness Scores and Standard Deviations (between Brackets) for Three Blocks of Training, Three Blocks of Sham Stimulation and Three Blocks of Transcranial Direct Current Stimulation to the Right Dorsolateral Prefrontal Cortex Block Mean Training 1 0.61 (0.05) 2 0.62 (0.06) 3 0.65 (0.18) Sham Stimulation 1 0.64 (0.08) 2 0.69 (0.09) 3 0.68 (0.07) Anodal Transcranial Direct
Current Stimulation 1 0.64 (0.05)
2 0.67 (0.09)
3 0.68 (0.09)
Note. Correctness scores where determined by the sum of correct responses to change and no change divided by the amount of trials.
10 Figure 3 Overall Mean Change Detection Task Correctness Scores with Error Bars Per Session Block.
A Two-way repeated-measures ANOVA was applied to the compiled mean correctness scores of each corresponding block of the conditions sham stimulation and atDCS. There was a main effect for the condition corresponding blocks (F(2,30) = 6.405,p = .005). No main effect for the conditions (F(1,15) = .804,p = .384). And no interaction effect for the condition corresponding blocks and the conditions (F(2,30) = .521,p =.599). No difference was found between conditions, therefore no effect of atDCS on VWM was found. Further a difference was found between condition corresponding blocks. Therefore a difference was found of change detection task correctness scores between blocks.
Comment [A1]: Kijk uit , hier is het niet duidelijk waarover je het hebt
11 Paired-samples T-tests where applied to the means of each corresponding block of atDCS and sham stimulation. There was no difference between the overall mean change detection task correctness score of sham stimulation block 1 and atDCS block 1 (t(15)=-.146,p=.886), no difference between sham stimulation block 2 and atDCS block 2 (t(15)=1.288,p=.217) (see Figure 4) and no difference between sham stimulation block 3 and atDCS block 3 (t(15)=.037,p=.971). As there was no difference in change detection task correctness scores between sham stimulation block 2 and atDCS block 2 there again no effect of atDCS on VWM performance was found.
The difference between blocks was further explored with paired-samples T-tests applied to the means of each consecutive block within the atDCS condition and the sham condition. There was a difference between the overall mean change detection task correctness score of sham stimulation block 1 and block 2 (t(15) = -2.609,p = .020) (se Figure 5), a difference between sham stimulation block 1 and sham block 3 (t(15) = -2.590,p = .021) (see Figure 6) and a difference between sham atDCS block 1 and atDCS block 3 (t(15) = -2.258,p = .039) (see Figure 7). There were no differences between sham stimulation block 2 and block 3, atDCS block 1 and block 2, and atDCS block 2 and block 3. Participants performed better at the change detection task during sham
stimulation block 2 compared to block 1, even though no stimulation was administered during either blocks. Participants also performed better in both conditions during the third block compared to the first block, again also even if no stimulation was
12 Figure 4 Overall Mean Change Detection Task Correctness Score of Sham Stimulation Block 2 and atDCS Block 2 with Error Bars.
Figure 5 Overall Mean Change Detection Task Correctness Score of Sham Stimulation Block 1 and Sham Stimulation Block 2 with Error Bars.
13 Figure 6 Overall Mean Change Detection Task Correctness Score of Sham Stimulation Block 1 and Sham Stimulation Block 3 with Error Bars.
Figure 7 Overall Mean Change Detection Task Correctness Score of atDCS Block 1 and atDCS Block 3 with Error Bars.
14 Discussion
The present study explored the possibility of enhancing visual working memory (VWM) by applying anodal transcranial direct current stimulation (atDCS) to the right dorsolateral prefrontal cortex (rDLPFC). No effect of atDCS to the rDLPDC on VWM was found: there was no difference between change detection task correctness scores between sham stimulation and atDCS stimulation. These findings are not in line with the expected effect based on previous atDCS studies which suggest that atDCS enhances different aspects of WM (Fregni et al., 2005; Boggio et al., 2006; Mulquiney et al., 2011; Andrews et al., 2011). These findings are also not in line with Sligte et al. (2011) as they suggest that TMS decreases VWM, the opposite therefore could be expected to be apparent when administering atDCS in an equal manner as Fregni et al. (2005), Boggio et al. (2006) Mulquiney et al. (2011) and Andrews et al. (2011).
Furthermore a difference was found between sham stimulation block 1 and sham stimulation block 2. As there was no actual stimulation during either of these blocks this effect is unexpected. Two possible explanations for this effect arise: either subject-expectancy effect occurred, where participants believed to receive stimulation and therefore as a placebo effect performed better on the change detection task. Or participants performed better because of the sequential order in which the sham
stimulation took place, as the sham stimulation was always during the second block, the first block might have lower scores as participants were recovering there trained ability for the change detection task and therefore did not commence at their optimal level of performance. Both explanations are plausible as a similar effect, between atDCS block 1 and aTDCS block 2, was almost found. A similar difference, of block 3 receiving a better performance as block 1 occurred in both conditions. As participants were aware of the stimulation, being sham or atDCS, had ended before commencing block 3 a subject-expectancy effect might be discredited, as there was no difference found between block 2 and block 3 of both conditions. Therefore a subject-expectancy effect in the form of a placebo might be possible, but more likely would be that participants needed the first block to regain their training experience and their optimal performance. Follow-up studies
15 are suggested to differ in block stimulation order by perhaps adding a fourth block per session and thereby vary between block 2 and block 3 as the sham stimulation or atDCS block.
The lack of atDCS effect on VWM could be due to procedural deficiencies, the general lack of effect of atDCS to the rDLPFC or a misunderstanding of the role, if any, of the rDLPFC in visual cognitive abilities.
The latter possible explanation, being a misunderstanding of the role, if any, of the rDLPFC in visual cognitive abilities, is most likely insufficient or even wrong. As previous studies have shown that the rDLPFC is involved in nonverbal memory encoding (Floel et al., 2004) and the encoding of locations of visual objects (Epstein et al., 2002). Both forms of encoding involved in visual cognitive abilities.
Furthermore, the second explanation, being the general lack of effect of atDCS to the rDLPFC is somewhat likely. As it has been shown by Fregni et al. (2005) that atDCS to the left DLPFC enhances working memory. Meaning that the DLPFC, at least the left DLPFC, is prone to effects of atDCS. But this provides no information about the rDLFPC ans its proneness to effect of atDCS. Moreover, Sligte et al. (2011), have shown that transcranial magnetic stimulation (TMS) administered to the rDLPFC disturbs VWM. But it has not been shown in previous studies whether VWM through the rDLPFC can be
enhanced. Due to lack of previous studies and the outcomes of the present study it might be likely that atDCS has no effect on VWM. Follow-up studies are suggested to explore this possibility further by acquiring participants with a below average VWM capacity. Perhaps suboptimal performers on VWM task could be optimized by atDCS.
Further procedural deficiencies might be a cause for the lack of an effect of atDCS to the rDLPFC on VWM. In the present study atDCS current of 1 mA was administered, this current might have been an ineffective current, either too low or too high to enhance VWM. Possibly a different current might cause an effect. Follow-up studies are suggested to explore the effects of varying currents of atDCS to the rDLPFC on VWM.
The duration of a stabile atDCS, being 1200 seconds, might be too short of time for an effect. Meinzer et al. (2014) found an enhanced effect of atDCS on learning and
16 maintaining novel vocabulary during multiple sessions of atDCS. Therefore follow-up studies are suggested to explore the effects of prolonged or multiple sessions of atDCS of the rDLPFC on VWM.
Locating the rDLPFC in the present study was done by utilizing 64 channels EEG caps. Hereby the position of F4, being the position superior to the rDLPFC, was roughly determined. This correct use and placement of the EEG cap was determined by asking participants to point to and touch the middle of the top of their head. The correct placement of the EEG cap and therefore positioning of F4 and therefore the rDLPFC was assumed once participants could touch the CZ location of the EEG cap spot on. This method of locating the rDLPFC might be imprecise, causing the current not to reach the rDLPFC sufficiently and therefore lacking an effect. Quite possibly there are more exact methods of determining the position of the rDLPFC and it is recommended for follow-up studies to apply more precise methods.
Further, the present study took place in an office room, in which at least one instructor was present in the vicinity of the participant. This might have caused some distraction to the participant causing less visual cognitive ability. Follow-up studies are recommended to utilize individual experimentation chambers with a reduced possibility of external distraction.
The preliminary conclusion based on the present study is that there is no effect of atDCS to the rDLPFC on VWM. Follow-up studies will determine whether this preliminary conclusion holds even when controlled for the various procedural deficiencies presented.
17 References
Andrews, S. C., Hoy, K. E., Enticott, P. G., Daskalakis, Z. J., & Fitzgerald, P. B. (2011). Improving working memory: the effect of combining cognitive activity and anodal transcranial direct current stimulation to the left dorsolateral prefrontal cortex. Brain Stimul, 4(2), 84-89. doi: 10.1016/j.brs.2010.06.004
Boggio, P. S., Ferrucci, R., Rigonatti, S. P., Covre, P., Nitsche, M., Pascual-Leone, A., & Fregni, F. (2006). Effects of transcranial direct current stimulation on working memory in patients with Parkinson's disease. J Neurol Sci, 249(1), 31-38. doi: 10.1016/j.jns.2006.05.062
Coffman, B. A., Clark, V. P., & Parasuraman, R. (2014). Battery powered thought: enhancement of attention, learning, and memory in healthy adults using transcranial direct current stimulation. Neuroimage, 85 Pt 3, 895-908. doi: 10.1016/j.neuroimage.2013.07.083
Floel, A. (2004). Prefrontal Cortex Asymmetry for Memory Encoding of Words and Abstract Shapes. Cerebral Cortex, 14(4), 404-409. doi: 10.1093/cercor/bhh002 Fregni, F., Boggio, P. S., Nitsche, M., Bermpohl, F., Antal, A., Feredoes, E., . . .
Pascual-Leone, A. (2005). Anodal transcranial direct current stimulation of prefrontal cortex enhances working memory. Exp Brain Res, 166(1), 23-30. doi: 10.1007/s00221-005-2334-6
Luck, S. J., & Vogel, E. K. (2013). Visual working memory capacity: from psychophysics and neurobiology to individual differences. Trends Cogn Sci, 17(8), 391-400. doi: 10.1016/j.tics.2013.06.006
Meinzer, M., Jahnigen, S., Copland, D. A., Darkow, R., Grittner, U., Avirame, K., . . . Floel, A. (2014). Transcranial direct current stimulation over multiple days improves learning and maintenance of a novel vocabulary. Cortex, 50, 137-147. doi: 10.1016/j.cortex.2013.07.013
Mulquiney, P. G., Hoy, K. E., Daskalakis, Z. J., & Fitzgerald, P. B. (2011). Improving working memory: exploring the effect of transcranial random noise stimulation and transcranial direct current stimulation on the dorsolateral prefrontal cortex.
18 Clin Neurophysiol, 122(12), 2384-2389. doi: 10.1016/j.clinph.2011.05.009 Nelson, J. T., McKinley, R. A., Golob, E. J., Warm, J. S., & Parasuraman, R. (2014).
Enhancing vigilance in operators with prefrontal cortex transcranial direct current stimulation (tDCS). Neuroimage, 85 Pt 3, 909-917. doi:
10.1016/j.neuroimage.2012.11.061
Nitsche, M. A., Fricke, K., Henschke, U., Schlitterlau, A., Liebetanz, D., Lang, N., . . . Paulus, W. (2003). Pharmacological modulation of cortical excitability shifts induced by transcranial direct current stimulation in humans. J Physiol, 553(Pt 1), 293-301. doi: 10.1113/jphysiol.2003.049916
Rampersad, S. M., Janssen, A. M., Lucka, F., Aydin, U., Lanfer, B., Lew, S., Wolters, C. H., Stegeman, D. F., & Oostendorp, T. F. (2014). Simulating Transcranial Direct Current Stimulation With a Detailed Anisotropic Human Head Model. Neural Systems and Rehabilitation Engineering, IEEE Transactions, 22(3), 441-452. doi: 10.1109/TNSRE.2014.2308997
Sligte, I. G., Wokke, M. E., Tesselaar, J. P., Scholte, H. S., & Lamme, V. A. (2011). Magnetic stimulation of the dorsolateral prefrontal cortex dissociates fragile visual short-term memory from visual working memory. Neuropsychologia, 49(6), 1578-1588. doi: 10.1016/j.neuropsychologia.2010.12.010
