Visualizing and plotting your data in BrainVision Analyzer

While Analyzer’s default displays are best suited for data from standard cognitive experiments, the default settings can be modified to suit other more particular requirements (e.g. mobile data, polysomnography). This can be achieved under the Preferences Menu (File > Configuration > Preferences) which includes the main presets for displaying the data. For example, while EEG and ERPs are traditionally plotted with the polarity positive down, an increasing number of researchers nowadays opt for having the positive side upwards. While Analyzer’s default settings follow the traditional convention, this can be changed under the “Scaling” tab of the Preferences Menu, by deactivating the option “Polarity Positive Down”. This change affects all displays in the time-domain across all workspaces.

Still within the Preferences Menu, features like the default scale and length of the display can be modified to fit specific requirements. In sleep EEG for example, a display time of 30 seconds is customary, in contrast to Analyzer’s default 10 seconds. This can be changed under the “Views” tab, by adjusting “Default Display Time” to 30 sec. Even though Analyzer’s Sleep Scoring Solution automatically adjusts to a 30 sec period, the settings under “Preferences” affect all views, such as Raw Data Inspection, or Edit Markers. Moreover, under the “Scaling” tab, the scale can be adjusted to fit the fluctuations seen in sleep EEG: for example, changing between a scale of 50 or 100 µV in the “Time Domain Scaling”. It is also possible to set individual scaling factors for non-EEG channels used in polysomnography that have different magnitudes, such as ECG or respiration.

Adjusting the default display is most helpful when regularly working with data whose scale does not fit Analyzer’s default settings, such as polysomnography data. This can be combined with data navigation and exploration tools to scroll through long recordings. Besides the default settings, adjustments can further be applied on individual displays should it be required, using the tools from the following sections.

In cognitive studies, it is common to use larger numbers of electrodes for recording and analysis. By default, Analyzer displays 32 channels on screen, grouped into separate pages. While this is helpful for inspecting channel features, some researchers prefer having more channels on display to get a better overview of all the data. To this end, the default number of channels can be changed under the “View” tab. Figure 4 shows the same data under 32 and 64-channel display configurations. While the 32-channel version enhances individual channel features, it requires the user to scroll to the next page to check the next channel group. Meanwhile, the adjusted display to 64 channels has a lower channel resolution but allows a broader overview of all channels. This reveals that the artifact seen over channels Fp1 and Fp2 next to the blink is also affecting the VEOG channel and likely reflects a head movement.

Channel numbers can be increased even further, as for recordings with 128 or more channels, such as hyperscanning paradigms where data from two or more participants is analyzed together. In this case, a display with more channels together with a vertical screen orientation can be helpful. Alternatively, channel numbers can also be reduced, which can be useful when working with smaller screens. In both cases, the channel numbers can be combined with functions like Edit Channels or Display Montages to select a specific channel order, or to assign colors to the channels.

Many more features can be customized under the “Preferences” menu. As a final example, the color of the nodes in the processing tree can be modified under the “Transformation Colors” tab. This feature can be helpful when managing large processing trees, as it facilitates the quick identification of different transformations or sections of the tree.

The “View Settings” menu allows the customization of each plot type, such as the standard view, mapping view, or grid view. View Settings can be specified as the default for all plots of the same type (via File > Configuration > View Settings), or for individual plots (via right-click on the plot).

An example for modifying the default grid view would be adding a baseline at the zero-amplitude level and placing the time scale at the time zero point. In the View Settings of Grid View, the “Show Baseline” and “Time 0” options can be used to this end. These settings can help assess the negative and positive inflections of ERPs more easily.

It is often necessary to make further adjustments, specifically to individual plots when preparing a graph for publication. For example, modifying the labels and the line thickness of the ERP waveforms above. In order to do this, the “View Settings” of a particular plot can be accessed by right-clicking on the display and selecting “Settings of the [X] view”, in this case, the grid view. Figure 7 shows the “View Settings” menu for this individual plot, so that the plot displays the channel and history node names to show the experimental conditions as figure labels. Furthermore, the line width was increased to a value of 2. All changes done on the individual “View Settings” are saved in the corresponding node in the history tree, remain after closing/re-opening the plot, and are replicated when using history templates.

Montages are channel arrangements that the user can create, and they are a helpful tool when evaluating data. They are defined by a certain channel number and order and/or a special reference scheme. Montages only affect the display and do not modify the underlying data. As a simple example, a montage can be created that excludes all non-EEG channels from display, and orders EEG channels from anterior to posterior. This reduces overlap of data from channels with large fluctuations to facilitate raw data inspection. The channels removed from display are not removed from the data.

New montages can be defined under Display > Montages > Create Montage. Bipolar montages like the double-banana commonly used in medical research can also be created. After choosing a reference scheme for the montage between the original reference, average, bipolar or Laplacian, the number and order of the channels can be defined in the menu. In Figure 9, the configurations are shown for a standard montage with channels from the 10-20 system and original reference, as well as for user-defined bipolar references. The effects on the data are seen on the right side.

Montages can be assigned to keyboard shortcuts under Display > Montages > Options. This enables to quickly shift between montages, and in this way evaluate the data from multiple perspectives. Toggling between the different montages using keyboard shortcuts can be used to identify relevant features on one reference scheme and check its effects on the others. This can be combined with Edit Markers to mark where the relevant features are found and locate them again during later processing stages. When searching for EEG features using montages with bipolar references, make sure that the positive polarity is facing down under the “Preferences” menu to follow convention.

All windows and panels in Analyzer can be flexibly resized and repositioned as needed. This is best demonstrated in Figure 10, where the history tree panel and the ribbon menu are hidden to increase space for EEG. The transient views generated from the data can also be moved at will. The panels can be returned to default settings under Windows > Reset Panel Layout. This is especially useful in case of closing a panel accidentally or just wanting to reset the display after experimenting with the layout.

Figure 10: Panels and windows are repositioned and resized flexibly to adjust the data display.¹

When working with multiple windows, Analyzer places them on different tabs by default, which can be selected with the pointer, or toggled using the CTRL + TAB keyboard shortcut. However, the tiling options from the toolbar re-arrange the windows to display them simultaneously. This feature is useful for comparison or exploration purposes.

In Figure 11, two windows were re-arranged using the “Tile Vertical” option from the toolbar to evaluate the frequency synchronization of two participants after a finger movement. With this tile arrangement, it is possible to visualize the plots simultaneously and interact with them such as when moving the crosshairs or adjusting the scale.

EEG and physiological data can have different characteristics depending on factors like the recording settings or the experimental paradigm. Besides customizing the display using View Settings, sometimes active adjustments are necessary during the exploration of the data, such as zooming, changing the scale, singling out a channel or aligning the data.

Some features in the data can only be visualized by zooming in closely, or on the contrary, zooming out. This can be achieved using the magnifier and interval tools from the toolbar. For example, in Figure 12, the electrodermal activity (EDA) channel was zoomed out to 2000 s, and the scale adjusted to visualize the fluctuation of arousal levels across various experimental blocks (seen as clustered markers).

Figure 12: The EDA channel was selected and zoomed out to visualize 2000 s of data, and the amplitude was increased to visualize skin conductivity fluctuations along an EEG experiment.¹

Conversely, some effects like TMS pulse artifacts are better visualized by zooming in closely into the data. To this end, an EEG channel was zoomed in to 10 ms using the “Zoom to selection” tool. After adjusting the amplitude to better visualize the artifact, this revealed a time window of 3.4 ms for the artifact, that can be later used for data interpolation. Once zoomed in, the “Marker Navigation” Add-In can be used to skip to other TMS pulse markers.

Data exploration tools are often used in combination. In Figure 14, a sleep dataset is shown where boundary lines are added at 75 µV per channel using “View Settings”. A channel is highlighted by clicking on it, before scrolling through the data. To prevent the data from shifting with each scroll due to its large fluctuations, the “Vertical Alignment” option is turned off, so that each channel is centered around zero instead of around the first data value on screen. Finally, the “Marker Navigation” add-in is used to skip to specific sleep stages that were previously marked. This saves having to scroll through the hours-long recording to reach the respective sleep stage. Besides these tools for checking the data, a specific Sleep Scoring Solution is also available, more information can be found in our Sleep Research webinar.

Figure 14: Diverse tools are used in combination to assess a sleep dataset: adding boundary lines, highlighting a channel, zero-centering the data, and marker navigation to skip to specific sleep stages.²

Besides using plots dynamically for data exploration, Analyzer can generate high resolution and vector images to use in publications. The image must first be prepared using the previously mentioned tools: preferences, view settings, montages, panel adjustments, navigation and exploration tools. Once the image is ready, export settings can be adjusted within the preferences menu under the “Graphics Export/Output” tab. Here the dimensions, margins and title of the image can be edited.

Figure 15 shows a plot as seen in Analyzer and after export. The plot represents a contrast of the LPP component between two experimental conditions during the presentation of emotion-inducing pictures. The original waveforms were overlaid through drag-and-drop, and the topography-over-time of the contrast was generated. Afterwards, the ERP and topographic plots were customized using “View Settings”, window repositioning and scale adjustment. To export both plots in one image, “Print to Graphics” was used.

Images can be exported in various formats including high resolution TIF and PNG, or vector-based EMF. Individual plots can be exported via Right-click > Save as File. This option may be preferred when the images will be edited in third-party software. Under Export > Generic Data, it is also possible to export the data to generate the plot in external software.

This article shows an overview of the possibilities for data visualization and plotting available in Analyzer. Hopefully, these examples illustrate how Analyzer can display your data according to your needs. Many further display customizations are possible, and they can be flexibly combined to best fit the requirements of your data. For any further questions you can always contact us.

Acknowledgements
¹ Data recorded using stimuli from the Nencki Affective Picture System (NAPS)^a,b,c
2 Sleep data courtesy of the Sleep Laboratory, University of Fribourg, Prof. Dr. Björn Rasch