Let's face it, from trial to trial: Comparing procedures for N170 single-trial estimation

Similar sound intensity dependence of the N1 and P2 components of the auditory ERP: Averaged and single trial evidence

“…Also, analyses of the N1 and P2 intensity dependence have been typically performed on averages of multiple trials, precluding examination of possible habituation effects. The averaging process might mask important interactions between the components at the single trial level by eliminating trial-to-trial amplitude variability (De Vos et al, 2012;Rousselet et al, 2007). Given the inconsistencies and methodological limitations described above, the main goal of the present study was to examine the N1 and P2 response functions to sound intensity levels while controlling for habituation effects by examining single trial activity.…”

“…To examine single trial event-related activity, a denoising method other than averaging is necessary in order to isolate the activity of interest. Methods for conducting these analyses include frequency denoising (Quiroga and Garcia, 2003) and Independent Component Analysis (ICA)-based signal decomposition (De Vos et al, 2012). Of these methods, ICA-based analysis has been shown to be the most reliable for single-trial estimation (De Vos et al, 2012).…”

The quest for single trial correlations in multimodal EEG-fMRI data

“…It can be seen that the subjects show clear activation in both fMRI and EEG. In particular, the topographies are very similar to the topographies shown in [3], indicating that reliable components involved in faceprocessing were extracted. Figures 2 and 3 show the range of the ST estimations of the face and house trials in fMRI and EEG.…”

“…One possible motivation for the absence of correlation is that the dynamics of the process of interest is not reliably estimated. ICA is a method that separates different sources based on its statistical properties, and was recently shown to reliably estimate a face-sensitive component [3]. This implies that by studying the ST dynamics on the ICA source level, we largely remove contributions of other ongoing processes that could obscure the ST dynamics of the face process of interest.…”

“…Recently, [3] showed that Independent Component Analysis (ICA) is able to extract a face-sensitive component from the EEG, explaining most of the variance of the N170 component. By studying the dynamics of such a face-sensitive component, the SNR of ST fluctuations in EEG related to face processing can be assumed to be improved compared to dynamics observed on the sensor level.…”