Investigating the source of BOLD nonlinearity in human visual cortex in response to paired visual stimuli

Nanyin Zhang, Xiao Hong Zhu, Wei Chen

Research output: Contribution to journalArticle

39 Scopus citations

Abstract

Several studies have demonstrated significant nonlinearity in the blood-oxygenation-level-dependent (BOLD) signal. Completely understanding the nature of this nonlinear behavior is important in the interpretation of the BOLD signal. However, this task is hindered by the uncertainty of the source of BOLD nonlinearity which could come from neuronal and/or vascular origin. The obscurity of this issue not only impedes accurate modeling of BOLD nonlinearity, but also limits generalization of the conclusions regarding BOLD nonlinearity. To examine this issue, we eliminated nonlinear contributions from the neuronal response and selectively study BOLD nonlinearity under only the vascular effect by employing a paired-stimulus paradigm composed of two ultra-short visual stimuli separated by a variable inter-stimulus interval (ISI). ISIs chosen were long enough (≥ 1s) to ensure invariant neuronal activity to all stimuli. Under this circumstance, we still observed significant nonlinearity in the BOLD signal reflected by a progressive recovery of BOLD response to the second stimuli as ISI gets longer and delayed BOLD onset latency. These nonlinear behaviors identified in the BOLD signal originate entirely from the vascular responses as the neuronal responses to all stimuli are identical. More importantly, we found that BOLD nonlinearity became much less significant after we removed activated pixels from large vessels. These finds reveal that the dominant component, if not all, of the source of BOLD nonlinearity comes from large-vessel hemodynamic response. They also suggest a possible mechanism to improve the spatial specificity of gradient-echo BOLD signal for fMRI mapping based on the characteristics of vascular refractoriness.

Original languageEnglish (US)
Pages (from-to)204-212
Number of pages9
JournalNeuroImage
Volume43
Issue number2
DOIs
StatePublished - Nov 1 2008

    Fingerprint

All Science Journal Classification (ASJC) codes

  • Neurology
  • Cognitive Neuroscience

Cite this