Limits...
TopoToolbox: using sensor topography to calculate psychologically meaningful measures from event-related EEG/MEG.

Tian X, Poeppel D, Huber DE - Comput Intell Neurosci (2011)

Bottom Line: Because the topographic patterns are obtained separately for each individual, these patterns are used to produce reliable measures of response magnitude that can be compared across individuals using conventional statistics (Davelaar et al.TopoToolbox can be freely downloaded.It runs under MATLAB (The MathWorks, Inc.) and supports user-defined data structure as well as standard EEG/MEG data import using EEGLAB (Delorme and Makeig, 2004).

View Article: PubMed Central - PubMed

Affiliation: Department of Psychology, New York University, 6 Washington Place Suite 275, New York, NY 10003, USA. xing.tian@nyu.edu

ABSTRACT
The open-source toolbox "TopoToolbox" is a suite of functions that use sensor topography to calculate psychologically meaningful measures (similarity, magnitude, and timing) from multisensor event-related EEG and MEG data. Using a GUI and data visualization, TopoToolbox can be used to calculate and test the topographic similarity between different conditions (Tian and Huber, 2008). This topographic similarity indicates whether different conditions involve a different distribution of underlying neural sources. Furthermore, this similarity calculation can be applied at different time points to discover when a response pattern emerges (Tian and Poeppel, 2010). Because the topographic patterns are obtained separately for each individual, these patterns are used to produce reliable measures of response magnitude that can be compared across individuals using conventional statistics (Davelaar et al. Submitted and Huber et al., 2008). TopoToolbox can be freely downloaded. It runs under MATLAB (The MathWorks, Inc.) and supports user-defined data structure as well as standard EEG/MEG data import using EEGLAB (Delorme and Makeig, 2004).

Show MeSH
Results comparing the timing of motor response peaks as determined classically through the root mean square (RMS) across MEG sensors versus motor response peaks as determined by the angle dynamics test. The angle dynamics test used a template defined by the response-locked epoch, which was compared (angle test) at each time point along the cue-locked epoch. Because the cue-locked epoch has been adjusted according to each individual's RMS defined motor response peak, the zero point on the x-axis is the classically defined motor peak. Validating the angle dynamics test, the difference between the within and between angle measures becomes nonsignificant during a 101 ms window around the zero point. Shaded regions show the 95% confidence intervals for the between and within angle measures. The grand average of template and cue-locked responses at 3 different times are depicted on the bottom. As seen in these grand average responses, the cue-locked topography is similar to the response-locked template at the zero point but dissimilar 100 ms before and 100 ms after the zero point.
© Copyright Policy - open-access
Related In: Results  -  Collection


getmorefigures.php?uid=PMC3090718&req=5

fig2: Results comparing the timing of motor response peaks as determined classically through the root mean square (RMS) across MEG sensors versus motor response peaks as determined by the angle dynamics test. The angle dynamics test used a template defined by the response-locked epoch, which was compared (angle test) at each time point along the cue-locked epoch. Because the cue-locked epoch has been adjusted according to each individual's RMS defined motor response peak, the zero point on the x-axis is the classically defined motor peak. Validating the angle dynamics test, the difference between the within and between angle measures becomes nonsignificant during a 101 ms window around the zero point. Shaded regions show the 95% confidence intervals for the between and within angle measures. The grand average of template and cue-locked responses at 3 different times are depicted on the bottom. As seen in these grand average responses, the cue-locked topography is similar to the response-locked template at the zero point but dissimilar 100 ms before and 100 ms after the zero point.

Mentions: As seen in Figure 2, between and within angle measures were found at each sample point within the cue-locked epoch (using the response-locked template). This was done separately for each individual, and then these values were averaged and graphed with 95% confidence levels to produce the plots. The zero point of the x-axis is the time at which the motor response reached its peak value as classically defined by RMS. This was done separately for each individual, and time is shown relative to these individually determined peak times. As seen in Figure 2, the between angle measure approaches the within angle measure 50 ms before the RMS-defined peak latency (i.e., the zero point on the x-axis) and falls below the within angle measure 50 ms after the peak latency. Furthermore, beyond validating the timing of the peak time using the angle dynamics test, the angle test at the peak latency was not significantly dissimilar from the response-locked template, whereas they were significantly dissimilar 100 ms before and after the peak latency. The grand average topographies in Figure 2 further confirmed the results of the angle dynamics test: the cue-locked response at 0 ms shared the same distribution as the template, whereas the responses at −100 ms and 100 ms were apparently different from the template. This suggests that distribution of neural sources responsible for the motor response was different from the distribution of neural sources just before and just after the response. In contrast, during the peak time as defined by the angle dynamics test applied to the cue-locked epoch, the pattern across the sensors was similar to the template as defined by the response-locked epoch [2].


TopoToolbox: using sensor topography to calculate psychologically meaningful measures from event-related EEG/MEG.

Tian X, Poeppel D, Huber DE - Comput Intell Neurosci (2011)

Results comparing the timing of motor response peaks as determined classically through the root mean square (RMS) across MEG sensors versus motor response peaks as determined by the angle dynamics test. The angle dynamics test used a template defined by the response-locked epoch, which was compared (angle test) at each time point along the cue-locked epoch. Because the cue-locked epoch has been adjusted according to each individual's RMS defined motor response peak, the zero point on the x-axis is the classically defined motor peak. Validating the angle dynamics test, the difference between the within and between angle measures becomes nonsignificant during a 101 ms window around the zero point. Shaded regions show the 95% confidence intervals for the between and within angle measures. The grand average of template and cue-locked responses at 3 different times are depicted on the bottom. As seen in these grand average responses, the cue-locked topography is similar to the response-locked template at the zero point but dissimilar 100 ms before and 100 ms after the zero point.
© Copyright Policy - open-access
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC3090718&req=5

fig2: Results comparing the timing of motor response peaks as determined classically through the root mean square (RMS) across MEG sensors versus motor response peaks as determined by the angle dynamics test. The angle dynamics test used a template defined by the response-locked epoch, which was compared (angle test) at each time point along the cue-locked epoch. Because the cue-locked epoch has been adjusted according to each individual's RMS defined motor response peak, the zero point on the x-axis is the classically defined motor peak. Validating the angle dynamics test, the difference between the within and between angle measures becomes nonsignificant during a 101 ms window around the zero point. Shaded regions show the 95% confidence intervals for the between and within angle measures. The grand average of template and cue-locked responses at 3 different times are depicted on the bottom. As seen in these grand average responses, the cue-locked topography is similar to the response-locked template at the zero point but dissimilar 100 ms before and 100 ms after the zero point.
Mentions: As seen in Figure 2, between and within angle measures were found at each sample point within the cue-locked epoch (using the response-locked template). This was done separately for each individual, and then these values were averaged and graphed with 95% confidence levels to produce the plots. The zero point of the x-axis is the time at which the motor response reached its peak value as classically defined by RMS. This was done separately for each individual, and time is shown relative to these individually determined peak times. As seen in Figure 2, the between angle measure approaches the within angle measure 50 ms before the RMS-defined peak latency (i.e., the zero point on the x-axis) and falls below the within angle measure 50 ms after the peak latency. Furthermore, beyond validating the timing of the peak time using the angle dynamics test, the angle test at the peak latency was not significantly dissimilar from the response-locked template, whereas they were significantly dissimilar 100 ms before and after the peak latency. The grand average topographies in Figure 2 further confirmed the results of the angle dynamics test: the cue-locked response at 0 ms shared the same distribution as the template, whereas the responses at −100 ms and 100 ms were apparently different from the template. This suggests that distribution of neural sources responsible for the motor response was different from the distribution of neural sources just before and just after the response. In contrast, during the peak time as defined by the angle dynamics test applied to the cue-locked epoch, the pattern across the sensors was similar to the template as defined by the response-locked epoch [2].

Bottom Line: Because the topographic patterns are obtained separately for each individual, these patterns are used to produce reliable measures of response magnitude that can be compared across individuals using conventional statistics (Davelaar et al.TopoToolbox can be freely downloaded.It runs under MATLAB (The MathWorks, Inc.) and supports user-defined data structure as well as standard EEG/MEG data import using EEGLAB (Delorme and Makeig, 2004).

View Article: PubMed Central - PubMed

Affiliation: Department of Psychology, New York University, 6 Washington Place Suite 275, New York, NY 10003, USA. xing.tian@nyu.edu

ABSTRACT
The open-source toolbox "TopoToolbox" is a suite of functions that use sensor topography to calculate psychologically meaningful measures (similarity, magnitude, and timing) from multisensor event-related EEG and MEG data. Using a GUI and data visualization, TopoToolbox can be used to calculate and test the topographic similarity between different conditions (Tian and Huber, 2008). This topographic similarity indicates whether different conditions involve a different distribution of underlying neural sources. Furthermore, this similarity calculation can be applied at different time points to discover when a response pattern emerges (Tian and Poeppel, 2010). Because the topographic patterns are obtained separately for each individual, these patterns are used to produce reliable measures of response magnitude that can be compared across individuals using conventional statistics (Davelaar et al. Submitted and Huber et al., 2008). TopoToolbox can be freely downloaded. It runs under MATLAB (The MathWorks, Inc.) and supports user-defined data structure as well as standard EEG/MEG data import using EEGLAB (Delorme and Makeig, 2004).

Show MeSH