Online Repetitive Transcranial Magnetic Stimulation of Dorsomedial and Dorsolateral Prefrontal Cortex in Cognition Decision Making, and Cognitive Dissonance

The overall goal of this procedure is to introduce a precise and effective method of repetitive transcranial magnetic stimulation using the neuron navigation system and the TMS robot to target the dorsolateral and dorsomedial prefrontal cortices during an experimental task. The main advantage of this protocol is the maximization of the RTMS effect as a result of more precise stimulation at a particular period of time during the experimental procedure, controlling its beginning and ending, relying on the stimulus onset and offset. This method can help to stimulate brain areas reducing the experimental assistance and influence on a participant's performance during an experimental task.

The protocol can be applied to different brain areas and experimental tasks. The MRI-assisted navigation system provides an accurate simulation of the target brain area and allows the TMS robot to adjust the position of its robotic arm according to the micro movements of the participants'head during the experiment. The further described protocol of the RTMS for inhibiting the DLPC and DMPC activation must be applied with the service of the MRI scan brain navigated system, magnetic stimulator, and robotic arm.

The MRI system navigation system provides an accurate stimulation of the target brain area, allowing the robotic arm to adjust its position according to the micro movements of the participant's head during the experiment. For this purpose, neuroanatomical MRI must be provided for all subjects of the study. The navigation system requires a specific set of equipment, such as the calibration plate for tracking a coil, the reference tracker for the participant's head localization, the pointer for setting a group of reference landmarks, and the infrared camera for tracking objects in space.

Coil calibration is provided by using the calibration plate, whereas the robotic arm does not need any additional equipment to be calibrated. All of them have silver reference balls, allowing the camera to recognize the objects in space. Align the calibration plate with the coil.

Set the scale on both the calibration plate and coil and calibrate using the software. Calibrate the robotic arm force sensor using the software. Press the robotic arm surface with different force levels in order to prevent hard pressure on the participant's head, but ensure a snug fit to the head during stimulation.

A major part of the navigation system preparation is to prepare individual MRI scans for the patient's registration procedure. It can be done even before a participant comes. The main goal lies in matching the individual brain coordinates and the coordinates in universal coordinate systems, such as MNI and Talairach.

Make a 3D model of the participant's head according to the software of the TMS system you use and then do the Talairach definition procedure. Set three benchmarks, namely the anterior commissure posterior commissure, and falx cerebri. And then adjust the Talairach space grid to the size of the 3D model of the head.

Set the targets of the feature stimulation and the required rotation angle according to the predefined MNI coordinates of your study. Calculate entries for the targets as the correspondent locations on the surface of the participant's head to position the robotic arm during the stimulation. We put entries for the vertex control point manually as the midpoint on the head surface.

However, you can also insert the standard coordinates from reliable studies as the target points and calculate entries. Set MRI landmarks on the 3D model of your participant's head for more accurate tracking of the participant's head location. For our study, we use landmarks on the left and right ears and the nasion, but you can use any other landmarks for your convenience.

The patient registration procedure is done when the participant arrives. Provide the consent forms regarding information about the experiment and ear blocks for protecting the participant's hearing. Place the reference tracker on the participant's head.

The reference tracker has to be attached to the participant's head at the point where the robotic arm will not occasionally touch the tracker during the stimulation. The tracker has three silver reflective reference balls, which are identified by the navigation system to localize the coordinates of the participant's head during the procedure and the experiment. Conventionally, the tracker is placed on the forehead area.

However, since we wanted to be sure that the entry targets of the DLPC and DMPC are available for the navigation system and can be reached by the robotic arm, we place the tracker between the forehead and the right ear. You can place it near the left ear as well, depending on the leading reference plate of the robotic arm. Fix the reference tracker on the participant's head by the elastic belt tightly enough, but quite comfortable for the participant.

The belt fixation is flexible depending on the participant's head size. We moved the belt down to the forehead area, bent it down, and fixed it to provide better access to the DMPC area by the robotic arm. We also tucked the middle of the belt edge to minimize the muscle contractions during stimulation.

The strong muscle contractions could prevent participants from seeing the screen with stimuli and might also cause unpleasant sensations. When the reference tracker is placed on the participant's head, set the real-world landmarks, ears, and the nasion in our case, according to the previously found MRI landmarks by using the pointer with a set of reference silver balls. The root mean square deviation value should not be higher than five millimeters.

If the value's higher, set the real-world landmarks again and recalculate the root mean square deviation value. Supplement these landmarks with a series of additional landmarks by the same pointer to define the participant's head surface. These points are added by creating continuous, horizontal and vertical lines uniformly distributed along the head surface.

A minimum of 300 additional surface landmarks are required. After the surface registration, the root mean square deviation value should be less than two millimeters to provide a good registration quality. While putting additional points, you should ensure that points are represented adequately on the head surface.

After the registration ends, check that the software correctly registers the position of the pointer on the 3D model by placing it at the head surface. You can choose different montages to attach the electrodes. We put the anode electrode on the projection of the first dorsal interosseus muscle of the right arm, which was chosen as the target to define the motor threshold.

The cathode was placed on the projection of the distal extremity of the proximal phalanx of the index finger, while the ground electrode was placed on the bone of the right arm near the elbow. Scrub participant's skin and remove remnants of the scrub with alcohol wipes. Take three clamp electrodes and insert them into the correspondent sockets of the amplifier.

Anode for the FDI muscle, cathode for the index finger bone, and the ground electrodes for the area near the elbow. Connect the amplifier to the charged battery. Stick three sticky electrodes to the participant's skin.

While sticking the electrode to the FDI muscle, ask the participant to test the FDI muscle by putting all the fingers together and stretching them. Whereas to stick the electrode to the index finger bone, ask the participant to bend their finger. Attach clamp electrodes to sticky electrodes.

Ensure that the signal from the FDI muscle is not noisy. The signal is registered with a special software. We use the BrainVision Recorder.

It's particularly necessary to ensure a good signal because otherwise, the motor threshold will be either overstated or understated. The noise amplitude should not exceed 50 microvolts peak-to-peak, otherwise, try some ways to make the participant's FDI muscle relaxed. You can place the participant's head on the pillow or ask the participant to strain his hand for a short time and then relax it.

Set the stimulator intensity to 50%and then adjusted depending on the reaction to pulses. Since the range of motor threshold is considerable, find the hand knob bearing at the precentral gyrus by orienting the goal according to the 3D individual head model. The goal should be put tangential to the head surface and oriented at 45 degrees with respect to the midline.

While finding the motor hotspot, make several pulses in the hand knob area and record the points on the head surface where the amplitude of the motor evoked potentials was large. Repeat the pulses to the points with the highest motor response to check their stability. You should increase or decrease the stimulation intensity to find the approximate density at which you can find the stable motor response only in the potential hotspot.

After choosing the hotspot, orient the goal in the chosen position and make 10 pulses with the intensity at which you choose the motor hotspot. You should count the number of times when the amplitude of the motor evoked potential exceeds 50 microvolts peak-to-peak. You should again adjust the stimulation intensity in order to find the one at which you can find that four out of 10 passes.

The motor evoked potential amplitude is equal to or higher than 50 microvolts. In order to start the experimental stimulation, switch the manual coil to the robotic arm coil and switch the cooling system to ensure the proper work of the robotic arm to provide better access to your region of interest, and to ensure that the coil does not prevent from seeing the screen. Adjust the back of the arm chair to make the reclining participant's position.

Due to the shift of the participant's head, you should adjust the location system, so that it can easily locate the reference tracker and all the required stimulation areas. If needed, the arm chair can be lifted or put down to ensure the robotic arm can access the stimulation points. Before launching the experiment, check the good position of the robotic arm on the participant's head surface, besides, the participants should gradually get accustomed to the stimulation frequency.

Since the stimulation will be triggered in the experimental software, connect the stimulator in the trigger station with the parallel port. Activate the trigger station software to synchronize the software with the trigger station and set the stimulation parameters. For the experiment itself, the number of pauses should be set to 20 for the 20 hertz stimulation, and the half period should be set to 25 milliseconds.

During the training session, start with one pulse and then increase the number of pulses to 5, 10, 15, and finally to 20. After each pulse or cranial passes, ask the participant whether the stimulation is comfortable. If the stimulation causes discomfort, stop the experiment and exclude the participant from the final sample.

During the coil alignment, you should ensure that the coil fits closely to the head surface, but does not push it hard. If the coil pushes hard or does not remain stable on the head surface, the coil sensitivity settings can be adjusted. Adjust the screen position to make it fully visible for the participant in the reclining position.

The mouse and the keyboard can be put on participant's knees or the pillow above. The stimulation task is organized in randomized blocks with the intervals between them when you should navigate the robotic arm to the required position on the head surface. During the interval, the participant can have some rest.

The main question of interest for the analysis was whether people changed their preferences for images, rating them for the second time compared to the first rating task differently for stimuli they chose or rejected between these two ratings. Namely, we were interested in whether the preference score and the second rating task increased or decreased compared to the first rating task, and most importantly, whether this positive or negative change is consistent with the choice made for different stimulation blocks, DLPC, DMPC, and two control blocks, vertex and sham. We can see that the difference in preference change is consistent with the choice.

Participants diminish their preferences more largely for rejected than chosen images. The difference in preference change is significant for DLPC, vertex, and sham stimulation blocks, which means that we could reproduce the behavioral effect in control conditions. However, as opposed to what we expected, the downregulation of DLPC did not affect the preference change process.

While according to the results, DMPC was suppressed, which lead to no consistent preference change for chosen versus rejected stimuli. Based on our results, several implications are crucial for future online rTMS research. Controlling for the place of the stimulation is very important.

Unlike in modern online experiments, when we can observe TMS effects using TMS, in cognitive experiments, such an observation is not available. One way to overcome this complication would be to refer to the place of the stimulation in the FMRI results. However, such interpretation often suffers from the reserved inference problem, or namely, the lack of sufficient selectivity of the area in question.

A special attention must be paid to the TMS intensity. Finding stimulation intensity by measuring MEP does not fully account for the distribution and intensity of the electric field induced in each trial. Recently, researchers have suggested modeling the electric field in TMS studies, which is a more accurate way to determine the stimulation intensity.

To target the neural substrate of cognitive functions, it is more beneficial to apply TMS and EEG, or TMS and FMRI concurrently, or before and after the TMS procedure if the neural response and interest cannot be directly measured, such as dorsomedial and dorsolateral prefrontal cortices activity. We conclude that this improvements to the protocol can make it applicable to other cognitive studies that presume the involvement and causal role of the DLPC and DMPC in cognitive processes of interest.