Localizing P300 Generators in Visual Target and Distractor Processing: A Combined Event-Related Potential and Functional Magnetic Resonance Imaging Study

Christoph Bledowski, David Prvulovic, Karsten Hoechstetter, Michael Scherg, Michael Wibral, Rainer Goebel, David E.j. LindenJournal of Neuroscience 2004 · DOI

Abstract

Constraints from functional magnetic resonance imaging (fMRI) were used to identify the sources of the visual P300 event-related potential (ERP). Healthy subjects performed a visual three-stimulus oddball paradigm with a difficult discrimination task while fMRI and high-density ERP data were acquired in separate sessions. This paradigm allowed us to differentiate the P3b component of the P300, which has been implicated in the detection of rare events in general (target and distractor), from the P3a component, which is mainly evoked by distractor events. The fMRI-constrained source model explained >99% of the variance of the scalp ERP for both components. The P3b was mainly produced by parietal and inferior temporal areas, whereas frontal areas and the insula contributed mainly to the P3a. This source model reveals that both higher visual and supramodal association areas contribute to the visual P3b and that the P3a has a strong frontal contribution, which is compatible with its more anterior distribution on the scalp. The results point to the involvement of distinct attentional subsystems in target and distractor processing.

Introduction

The P300 is one of the most extensively studied signals in cognitive neuroscience, appearing as a positive deflection in brain activity roughly 300–600 milliseconds after a person encounters a significant or unexpected stimulus. Despite decades of research, its exact origin in the brain remained controversial due to the "inverse problem" in electroencephalography (EEG): a specific pattern of electrical activity on the scalp could theoretically be produced by an infinite number of different source configurations inside the brain.

Furthermore, the P300 is not a single wave but consists of at least two distinct subcomponents: the P3b, triggered by task-relevant targets, and the P3a, triggered by unexpected, task-irrelevant distractors. This study aimed to pinpoint the specific neural generators for both, testing whether they arise from the same or different neural networks by using fMRI activity to "seed" and constrain EEG source models.

Methods

The researchers employed a three-stimulus oddball paradigm where participants viewed a stream of blue shapes:

  1. Standards (90%): Frequent shapes requiring no response.
  2. Targets (5%): Rare shapes requiring a button press.
  3. Distractors (5%): Rare, task-irrelevant shapes to be ignored.

The core innovation was fMRI-constrained source modeling. Instead of estimating source locations from scalp data alone, the researchers used fMRI data (which identifies regions of increased blood flow) to identify candidate brain regions. These regions were then used as "seeds" for a mathematical model (BESA) to determine if electrical activity in those specific spots could explain the scalp EEG data. EEG was collected from 10 subjects using a 63-channel cap, while fMRI data was derived from a separate session with 13 subjects performing the identical task.

Figure 1: Grand-average ERP waveforms for target (black) and distractor (red) stimuli across the 63-channel scalp array.

Results

The fMRI-guided model explained an impressive 99.4% of the target condition variance and 99.5% of the distractor condition variance, suggesting that the regions identified by fMRI are indeed the primary generators of the electrical P300 signal.

Functional Dissociation of P3a and P3b

The study revealed that the P3a and P3b are spatially and temporally distinct:

  • P3b (The "Target" Network): Generated by the inferior parietal lobe (IPL), posterior parietal cortex (PPC), and inferior temporal cortex (IT). This suggests the P3b involves both high-level visual categorization and supramodal association.
  • P3a (The "Distractor" Network): Generated by the precentral sulcus (PrCS) and the insula (INS). This aligns with the "orienting response"—the brain's mechanism for quickly reallocating resources to an unexpected event.
Figure 5: Scalp maps (B) showing voltage and current source density (CSD) distributions, and representative waveforms (C) at Fz, Cz, and Pz electrodes.

Latency and Spatial Mapping

The P3a peaked significantly earlier (460 ms) than the P3b (513 ms). Spatially, the P3b was strongest in the parietal lobes, while the P3a showed a strong frontal contribution (PrCS), consistent with its more anterior distribution on the scalp. Interestingly, the PPC showed higher activity for targets than distractors, likely reflecting the goal-directed nature of the task.

Discussion

This study represents a significant step forward in human neuroimaging by combining the high temporal resolution of EEG with the high spatial resolution of fMRI. The results highlight the involvement of distinct attentional subsystems in target and distractor processing. Specifically, the inclusion of the inferior temporal (IT) cortex in the P3b network suggests that the brain's object-recognition machinery is directly involved in P300 generation. While the study was conducted in separate sessions, the high model fit provides robust evidence for these localized neural generators.

Reference: Bledowski, C., et al. (2004). Localizing P300 Generators in Visual Target and Distractor Processing. The Journal of Neuroscience, 24(42):9353–9360.

Summary updated

Abstract

Constraints from functional magnetic resonance imaging (fMRI) were used to identify the sources of the visual P300 event-related potential (ERP). Healthy subjects performed a visual three-stimulus oddball paradigm with a difficult discrimination task while fMRI and high-density ERP data were acquired in separate sessions. This paradigm allowed us to differentiate the P3b component of the P300, which has been implicated in the detection of rare events in general (target and distractor), from the P3a component, which is mainly evoked by distractor events. The fMRI-constrained source model explained >99% of the variance of the scalp ERP for both components. The P3b was mainly produced by parietal and inferior temporal areas, whereas frontal areas and the insula contributed mainly to the P3a. This source model reveals that both higher visual and supramodal association areas contribute to the visual P3b and that the P3a has a strong frontal contribution, which is compatible with its more anterior distribution on the scalp. The results point to the involvement of distinct attentional subsystems in target and distractor processing.
BibTeX
@article{bledowski2004localizing,
  author = {Christoph Bledowski and David Prvulovic and Karsten Hoechstetter and Michael Scherg and Michael Wibral and Rainer Goebel and David E.J. Linden},
  title = {{Localizing P300 Generators in Visual Target and Distractor Processing: A Combined Event-Related Potential and Functional Magnetic Resonance Imaging Study}},
  journal = {Journal of Neuroscience},
  year = {2004},
  doi = {10.1523/jneurosci.1897-04.2004},
}
Figure