Elsevier

Cognitive Brain Research

Volume 8, Issue 2, 16 July 1999, Pages 157-164
Cognitive Brain Research

Research report
Event-related neural activity associated with the Stroop task

https://doi.org/10.1016/S0926-6410(99)00017-8Get rights and content

Abstract

The time course of neural activity supporting performance during the Stroop task was investigated using event-related brain potentials (ERPs). Four spatially and temporally distinct modulations were observed differentiating the ERPs elicited by incongruent trials from the ERPs elicited by congruent, neutral, or word identification trials. Two of these modulations reflected increased negativity over the fronto-central region and positivity over the fronto-polar region for incongruent trials and may reflect conflict detection and resolution processes. The other modulations, distributed over the left parietal and temporo-parietal regions, may reflect the activity of a meaning-based conceptual level system active during congruent, neutral, and word identification trials; and the activity of a perceptual level system supporting task performance when only color information can guide an efficient response on incongruent trials.

Introduction

The Stroop interference effect [19]has been one of the most extensively studied phenomena in cognitive science. The effect refers to the increase in response latency observed when an individual is required to identify the color of a color-word when these aspects of the stimulus are incongruent (e.g., the word RED presented in the color blue) compared to the time required to name the color of a neutral (e.g., XXX in blue), or congruent (e.g., the word RED presented in the color red) stimulus. The Stroop effect has been utilized by researchers to explore the nature of automatic and controlled cognitive processes 9, 16, disturbances in cognition resulting from various psychiatric and/or neurological disorders [13], the neuro-cognitive architecture of selective attention [14], and age-related declines in inhibitory processing [22].

Current computational [5]and mathematical 11, 12models of the Stroop task propose that color and word information are processed in independent pathways that converge on a shared or common response system. For congruent and neutral trials responding is relatively fluent, as the color and word pathways activate the same response level representation. Interference, or the Stroop effect, arises on incongruent trials when information from the color and word pathways activate competing response level representations. The activation of more than one response level representation requires additional time for input by the color pathway to accrue enough activation to become differentiated from input by the word pathway and guide a response.

Neuro-imaging studies incorporating the positron emission tomography (PET) technique have begun to identify the neural architecture supporting performance of the Stroop task. Increased regional cerebral blood flow (rCBF) has been observed in the anterior cingulate in a number of studies 4, 15and has been proposed to reflect cognitive processes supporting efficient selection between competing color and word information. Increased rCBF has also been observed in the frontal polar region [1], the inferior frontal gyrus [20], and the inferior parietal region [1]. These findings indicate that the anterior cingulate is part of a distributed network supporting task performance. While these studies have identified a number of brain regions active during performance of the Stroop task, little is known about the time course of these neural events.

Scalp recorded event-related brain potentials (ERPs) offer real-time temporal resolution of neural processes, permitting a precise analysis of the time course of neural events supporting task performance. ERPs reflect event locked electrical activity generated by neural ensembles and consist of a series of positive (P) and negative (N) deflections above some pre-stimulus baseline level of activity that peak at particular intervals following stimulus onset. For instance, the third positive deflection in the waveform is generally labeled as P3, and a negative deflection occurring at 400 ms would be labeled N400.

In an early study, Duncan-Johnson and Kopell [7]reported that there were no differences in the latency or amplitude of the P3 elicited by congruent, neutral, or incongruent trials in the Stroop task. In comparison, there were large differences in response latency between the congruent and incongruent trials. Together these findings led them to suggest that interference was related to conflict in response selection rather than stimulus evaluation processes, consistent with dual-pathway models of the Stroop task. However, the conclusions of this study [7]were based upon findings of null results in the P3 modulation and a fairly conservative low-pass filter was used obscuring modulations other than the P3. Rebai et al. [14]also found no differences in the ERPs elicited by congruent and incongruent trials in the amplitude or latency of the P3. In comparison, significant differences between the ERPs for these two conditions did emerge over the central region in the range of 350 to 450 ms that were observed when the task required selection between competing color and word information. However, the enhanced N400 observed in this work for incongruent trials was greatest in a silent naming task making it difficult to determine exactly what individuals were doing in this condition. Also, both of these studies employed a very limited number of electrodes placed primarily over the midline of the scalp making it impossible to examine the topographic distribution of these ERP modulations.

In the current study, we recorded ERPs while individuals performed the Stroop task using a large array of electrodes in an effort to characterize the temporal and spatial patterns of neural activity associated with task performance. Based upon dual-pathway models and the findings of previous ERP studies we expected the ERPs reflecting early stages of perceptual processing and stimulus evaluation (i.e., P3) observed over the occipito-parietal regions to be similar for congruent, neutral, and incongruent trials. Given the findings of Rebai et al. [14], greater negativity for incongruent trials, relative to congruent trials, was anticipated over the fronto-central region possibly reflecting the activity of response selection or conflict resolution processes active when incompatible color and word information is encountered [3]. In addition to congruent, neutral, and incongruent trials we also included word identification trials in the task. These trials permitted us to examine patterns of neural activity supporting task performance when only color information could serve to guide a response on incongruent trials and neutral trials; and when word information could guide a response on congruent and word identification trials.

Section snippets

Subjects

Twelve volunteers (six females) 24–31 years of age participated in the study. All participants reported normal or corrected to normal visual acuity, 10 reported a right hand preference and two characterized themselves as being ambidextrous.

Materials and procedure

In the Stroop task, participants were asked to identify one of four colors (red, blue, green, or yellow) or the names of these colors by pressing one of four keys (V, B, N, or M) on a computer keyboard using the index and middle fingers of the right and left

Behavioral data

Group average response latency and proportion of errors for the congruent, neutral, incongruent, and word identification trials for the 25% and 50% word identification conditions are presented in Table 1. A 4 (Stroop task condition)×2 (proportion of word identification trials) ANOVA performed on the response latency data revealed significant main effects of Stroop task condition (F(3,33)=87.24, p<0.001) and proportion of word identification trials (F(1,11)=21.15, p<0.001) and a significant

Discussion

In the study, a robust Stroop effect was observed in both the response latency and response accuracy data. As anticipated increasing the proportion of word identification trials served to enhance the fluency of word identification processes demonstrated by the faster word identification latencies in the 50% than 25% condition. Complementing the behavioral data, four temporally and spatial distinct ERP modulations were observed reflecting differential neural processing for incongruent trials

Acknowledgements

This research was supported by a post-doctoral fellowship sponsored by grant RO1 AG13845-01 awarded to the first author and a grant from the National Alliance for Research on Schizophrenia and Depression awarded to the second author.

References (22)

  • S.A Hillyard et al.

    Event-related brain potentials in the study of visual selective attention

    Proc. Natl. Acad. Sci. U.S.A.

    (1998)
  • Cited by (0)

    1

    After July 1: Department of Psychology, University of Notre Dame, Notre Dame, IN 46556-0399, USA.

    View full text