Functional magnetic resonance imaging (fMRI) measures the correlation between the fMRI response and stimulus properties. A linear relationship between neural activity and fMRI response is commonly assumed. However, the response to repetitive stimulation cannot be explained by a simple superposition of singleevent responses. This might be due to neural adaptation or the hemodynamic changes underlying the fMRI BOLD response. To assess the influence of adaptation, the BOLD responses and visual evoked potentials (VEPs) to identical stimuli were recorded. To achieve different adaptation levels, 2-s stimulus epochs alter- In most functional magnetic resonance imaging (fMRI) experiments the correlation between the fMRI response and the application of a stimulus is measured. Although the nature of the complex interactions among the neurons, hemodynamic changes, and MRI is still unclear, it has been hypothesized that the fMRI response is a linear transformation of the neural activity averaged over time and voxel volume (1,2). At least for simple visual stimulation the model allows predictions of the observed blood oxygenation level dependent (BOLD) fMRI signal. While avoiding issues surrounding habituation effects by using single trials of fixed duration, sufficiently separated in time, it was shown that the BOLD fMRI response to multiple trials behaved approximately linearly (3-6). However, there is substantial literature (2,7-11) on nonlinear responses, and some disagreement on this point. Although there is evidence for neural adaptation in single cells (12) and evoked potentials (13,14), it is not clear whether the adaptation in fMRI is neural, hemodynamic, or a combination of both.The goal of this study was to evaluate if neural adaptation is sufficient to predict the observed nonlinearity in fMRI experiments. Therefore, identical visual stimulation paradigms were presented during fMRI as well as VEP recording to assess the influence of neural adaptation on the fMRI response. METHODS SubjectsTen healthy volunteers (four males and six females, 22-40 years old) participated in the VEP experiments, and eight healthy volunteers (six males and two females, 22-40 years old) in the fMRI experiments. Three of the subjects participated in both the VEP and fMRI experiments. All subjects provided informed consent, and the study was approved by the Institutional Review Board. Visual StimulationIn all experiments a phase-reversing checkerboard pattern served as the visual stimulus. The checkerboard was confined to a circular area of 25°diameter, checksize was 1.3°, contrast 96%, and reversal rate was 4.16 Hz ϭ 8.3 reversals/s. As the baseline condition between the stimulation units (ϭ in the interstimulus intervals (ISI)), spatially homogenous gray stimuli were displayed with identical net luminance.In the fMRI session the stimulation pattern was generated on a PowerMac and projected by video projector onto a backprojection screen in the magnet bore. The stimulation timing was 120 s stimulation with a fixed ISI and 60-s pause. Th...

Localized proton spectroscopy using the PRESS sequence was performed to measure the free induction decay of the water signal in a 2 x 2 x 2 = ml voxel located in the visual cortex after a short light pulse. The experiment was run continuously with a repetition time of 400 ms; the onset of the light pulse was triggered by the spectrometer. The FlDs were measured as a function of the echo time (TE) and the length of the stimulus (TS). The time course showed three different phases: An initial response was observed 400 ms after the onset of the stimulus, which was demonstrated to be mainly a change in the apparent spin density and/or T , with a minor contribution of T;. No significant change in T, could be measured. Changes in the ionic cell environment and/or the subsequent increase in cell size upon stimulation are discussed as a potential mechanism for the observed signal response. 0 1995 John Wiley & Sons, Inc

No abstract

No abstract

No abstract