fMRI evidence that the neural basis of response inhibition is task-dependent

Stewart H. Mostofsky, Joanna G.b. Schafer, Michael T. Abrams, Melissa C. Goldberg, Abigail Flower, Avery Boyce, Susan Courtney, Vince D. Calhoun, Michael A. Kraut, Martha B. Denckla, James J. PekarCognitive 2003 · DOI

TL;DR

This study uses fMRI to investigate how the brain inhibits unwanted actions. They found that the specific brain regions involved in stopping a response depend on the complexity of the task, suggesting that response inhibition isn't a single, monolithic process but rather a flexible system that recruits different neural resources based on cognitive demands.

What Problem Does This Paper Address?

Response inhibition, the ability to stop an action, is crucial for everyday behavior and is often impaired in various neuropsychiatric disorders like ADHD. While previous research has pointed to frontal lobe involvement, the exact neural mechanisms have been inconsistent, leading to the hypothesis that the brain regions responsible for inhibition might vary depending on the task's nature. This paper aims to clarify this by comparing brain activity during two different Go/No-go tasks: one simple and one that requires significant working memory.

Methods

The researchers employed event-related functional Magnetic Resonance Imaging (fMRI) to observe brain activity. They designed two Go/No-go tasks:

  • Simple Go/No-go: Participants had to press a button for green spaceships (Go) but withhold a response for red spaceships (No-go). This task was designed to isolate motor response inhibition with minimal cognitive load.
  • Counting Go/No-go: This task increased the working memory load. Participants had to respond to green spaceships and red spaceships only if they were preceded by an even number of green spaceships. This required participants to keep track of a sequence of stimuli.

fMRI data was analyzed using random effects models to identify significant signal changes associated with Go and No-go events. They specifically looked for differences in activation patterns between the two tasks.

Datasets

Not discussed in the paper. The study used human participants.

Key Insights and Ideas

  • Task-dependent neural basis: The core finding is that the brain regions involved in response inhibition are not fixed but depend on the task's cognitive demands.
  • Pre-SMA as a general inhibitor: The study suggests that the pre-supplementary motor area (pre-SMA) is consistently involved in inhibiting motor responses, regardless of task complexity. This points to its role as a fundamental component of motor inhibition.
  • DLPFC for working memory-guided inhibition: When a task requires holding information in working memory to guide inhibition (like the counting task), the dorsolateral prefrontal cortex (DLPFC) becomes crucially involved. This highlights the integration of executive functions with inhibitory control.
  • Simple vs. Complex Inhibition: The simple task primarily activated the pre-SMA for No-go responses, while the more complex counting task recruited the right DLPFC in addition to the pre-SMA. This supports a hierarchical model where basic inhibition relies on a core network, and more complex inhibition recruits higher-level cognitive control regions.

Key Results

  • Simple Go/No-go:
    • Go responses were associated with activation in the left primary sensorimotor cortex, supplementary motor area (SMA) proper, and anterior cerebellum.
  • No-go responses were primarily associated with activation in the pre-SMA.
  • Counting Go/No-go:
    • Go responses showed similar activations to the simple task, with additional involvement of the left thalamus and right middle occipital gyrus.
  • No-go responses showed activation in the pre-SMA, but crucially, also in the right DLPFC.
  • Between-task comparison: The counting task elicited significantly greater No-go activation in the right DLPFC compared to the simple task, confirming its role in working memory-dependent inhibition.

Limitations and Future Work

  • Single Subject Variability: The authors acknowledge that single-subject replication of No-go effects was less consistent than for Go effects, suggesting that inhibitory processes might be more subtle or variable across individuals.
  • Temporal Resolution: While event-related fMRI offers better temporal resolution than block designs, the authors note that limitations in fMRI's temporal resolution might obscure the precise timing of pre-SMA's involvement in preparation versus inhibition.
  • Future Directions: The paper suggests further research to explore the domain-specific roles of other frontal regions, such as the orbitofrontal cortex (OFC), in response inhibition governed by motivational information or reward costs. They also propose using combined EEG and fMRI to better disentangle the temporal dynamics of motor preparation and inhibition.

Summary updated

BibTeX
@inproceedings{mostofsky2003fmri,
  author = {Stewart H. Mostofsky and Joanna G.B. Schafer and Michael T. Abrams and Melissa C. Goldberg and Abigail Flower and Avery Boyce and Susan Courtney and Vince D. Calhoun and Michael A. Kraut and Martha B. Denckla and James J. Pekar},
  title = {{fMRI evidence that the neural basis of response inhibition is task-dependent}},
  booktitle = {Cognitive},
  year = {2003},
  doi = {10.1016/s0926-6410(03)00144-7},
}
Figure