The overall goal of this procedure is to map cortical dynamics underlying different human cognitive states. This is accomplished by first capturing magneto and electroencephalographic data or MEG and EEG in short, while a subject is performing a behavioral task. The second step is to obtain anatomical data using relevant MRI sequences.
Next co-registration is performed to establish spatial correspondence between sensor locations of MEG and EEG with anatomical information. The final step is to use an inverse imaging approach to map individual's brain activities onto the cortical space. Ultimately, statistical inference based on a common surface based coordinate system is used to discover significant spatial temporal patterns that distinguish one cognitive state from another.
Demonstrating a procedure will be Eric Lawson and Ross Maddox. For postdocs in my laboratory, Begin this protocol by obtaining structural MR images of the subject. First, acquire a structural MR scan using a magnetization prepared rapid gradient echo or MP rage or similar sequence.
This sequence may take five to 10 minutes depending on the specific scanning resolution and imaging protocol used if EEG data will be used for inverse imaging analysis. Also acquire two fast low angle shot or flash MRI scans. These flash sequences provide different tissue contrast from the standard MP rage sequences.
Once imaging is complete, use MNE and free surfer software to reconstruct the skin outer skull and inner skull surfaces from the MP rage and flash images. Then use these surfaces to generate a three layer boundary element model or bem prior to the MEG experiment. First test the auditory and visual latencies to ensure timing integrity.
Use a microphone and photo DDE attached to the screen, and subsequently ensure there is no observable jitter. This may necessitate setting the presentation projector to its native resolution. Next, prepare the subject for recording, referring to the previous video article by luital for details of electrogram and reference electrode preparation, as well as digitization of the subject's fiduciary landmarks, head position indicator coils and EEG electrodes.
Once the subject is comfortably seated in the MEG measure head position, using the head position indicator or HPI coils, then start recording and begin presentation of auditory and visual stimuli. Note HPI Measurements can also be taken continuously. The subject should respond to the auditory and visual stimuli via an optical button box while performing an audiovisual behavioral task.
Here the subject reports the spoke and digit originating from the hemifield as queued by the visual cue. Occasionally, subjects are visually prompted to switch attention to the contralateral hemifield mid trial. To study switching of auditory attention, many hardware and software solutions are available to perform stimulus presentation.
Here, Tucker Davis Technologies RZ six is used for auditory stimulus presentation and trigger stamping with psych toolbox for visual stimulus presentation, both controlled by matlab. To begin data processing coregister the EEG data to the structural MR using MNE software as seen here, first load the digitizer data into the subject's reconstructed MRI head model. Next, pick fiduciary landmarks to initiate the co-registration process and then proceed to use the automatic alignment procedure to complete the coordinate transformation.
Then to relate the location of each dipole in the source space with the location of each sensor. Combine the recorded head position indicator data to compute a forward solution with the three layer boundary element model to further increase the data's signal to noise ratio. Apply time domain artifact removal such as removing epics containing abnormally high amplitude signals due to spiking of a channel.
Also apply frequency domain artifact removal such as band notch filtering at 50 or 60 hertz line frequency use signal space projection or other noise reduction techniques such as signal space separation to project or separate out spatial field patterns from ambient environmental field contamination or other undesirable physiological signals such as those associated with eye blinks and cardiac artifacts. Now generate a brain movie of the distributed dipole estimate being the current estimate at each dipole location in the source space in time for each experimental condition. Depending on the temporal characteristics of the experimental design, data can be bend in time by averaging current estimates using non-overlapping temporal windows to continue analysis morph.
The previously created brain movies for each subject onto a common cortical space based on a surface based coordinate system that optimally aligns individual sical jarral patterns. This allows for comparison or averaging of cortical activities across subjects. To use a region of interest approach, ROIs can be defined anatomically, for example, by an automatic parcelization algorithm and or functionally by recording a functional localizing task such as a go no-go secon task to identify the oculomotor regions analysis can be further constrained to a specific time of interest that is appropriate to the experimental paradigm that was used, for example, constrained to a time period immediately before and after the onset of the sound stimuli.
Other statistical inference associated with time series analysis can also be used using the behavioral paradigm outlined above. Here we see representative results using the non-parametric spatiotemporal clustering procedure. The right frontal eye field is found to be significant when an individual subject is performing a reorientation task compared to a standard task.
Using the ROI approach, the time course of the right frontal eye field is shown along with the time period that these two conditions are significantly different. After watching this video, you should have a good idea of how to use M-E-G-E-E-G and MRI to map cortical dynamics in different behavioral tasks. By employing appropriate statistical approaches, you can discover different spatial temporal patterns that distinguish between cognitive states.
Thank you for watching and good luck with the experiments.
