Meeting Banner
Abstract #2342

Trial-By-Trial Global Modulation of BOLD Responses to Simple, Sensory Stimuli: Implications for Functional Brain Imaging and Understanding Positive and Negative BOLD Response Coupling

Stephen D. Mayhew1, Karen J. Mullinger2, Camillo Porcaro3, 4, Richard W. Bowtell2, Andrew P. Bagshaw1, Susan T. Francis2

1BUIC, School of Psychology, University of Birmingham, Birmingham, United Kingdom; 2SPMMRC, School of Physics and Astronomy, University of Nottingham, Nottingham, United Kingdom; 3Institute of Neuroscience, Newcastle University, Newcastle upon Tyne, United Kingdom; 4LETs-ISTC-CNR, Fatebenefratelli Hospital-Isola Tiberina, Rome, Italy

Conventional GLM analyses of fMRI data localise brain activity from the average stimulus response, ignoring trial-by-trial variability which is most relevant to the dynamics of brain function and behavioural outcomes. We combine data from visual, motor and somatosensory tasks to show that single-trial responses across the whole brain are concurrently modulated with activity in the stimulated primary sensory cortex. These modulations induce a positive correlation between single-trial positive and negative BOLD responses, despite a negative correlation between the average response magnitudes. These findings demonstrate that stimulus modulations extend over a far greater extent of the brain than previously suspected.

Abstract PDF Poster PDF

Keywords

acquisition activation active activity amplitude amplitudes analyses arise astronomy attention auditory averaging bilateral bilaterally bold brain brief cluster cohorts complexity concurrent condition conditions confound consistently contrast contrasts corrected correction correlated correlation correlations cortex cortices dabs datasets delay delayed delivered demand directly distribution double driven duration durations excess excluded experimental extent feedback filtering final finding findings fixed fluctuations force functional global handed highlight identify increasing individuals inform intensity inter interval investigate kingdom largest latency linear magnitude magnitudes mapping measured median mixed modalities modulation modulations motion motor negative network neurology neuronal noise observe parametric passive peak performance physiological positive positively post preprocessed previous primary providing pseudo pulses random recalculated recorded reduce reducing reflect regressors relevant represent respiration response responses school sense sensory session significance significantly simple smoothing sources stimulated stimulation stimuli stimulus strengths subject subjects subsequently substantial suggesting suggests task tasks temporal thought timings trial trials unattended understanding unilateral upon variability visual widespread