Assessing the influence of scanner background noise on auditory processing. I. An fMRI study comparing three experimental designs with varying degrees of scanner noise

Abstract

We compared two experimental designs aimed at minimizing the influence of scanner background noise (SBN) on functional MRI (fMRI) of auditory processes with one conventional fMRI design. Ten subjects listened to a series of four one-syllable words and had to decide whether two of the words were identical. This was contrasted with a no-stimulus control condition. All three experimental designs had a duration of approximately 17 min: 1) a behavior interleaved gradients (BIG; Eden et al. [1999] J Magn Reson Imaging 41:13-20) design (repetition time, TR, = 6 s), where stimuli were presented during the SBN-free periods between clustered volume acquisitions (CVA); 2) a sparse temporal sampling technique (STsamp; e.g., Gaab et al., [2003] Neuroimage 19:1417-1426) acquiring only one set of slices following each of the stimulations with a 16-s TR and jittered delay times between stimulus offset and image acquisition; and 3) an event-related design with continuous scanning (ERcont) using the stimulation design of STsamp but with a 2-s TR. The results demonstrated increased signal within Heschl's gyrus for the STsamp and BIG-CVA design in comparison to ERcont as well as differences in the overall functional anatomy among the designs. The possibility to obtain a time course of activation as well as the full recovery of the stimulus- and SBN-induced hemodynamic response function signal and lack of signal suppression from SBN during the STsamp design makes this technique a powerful approach for conducting auditory experiments using fMRI. Practical strengths and limitations of the three auditory acquisition paradigms are discussed.

Figure 1

Experimental stimulation (A). Experimental condition (B). Control condition as well as image acquisition for STsamp. The delay between the end of the auditory stimulation and the beginning of the image acquisition was varied over 8 s. Each ITP corresponds to the volume acquired after the end of auditory stimulation, e.g., ITP0 corresponds to volumes acquired 0 s after the end of the auditory stimulation, and ITP5 corresponds to volumes acquired 5 s after the end of the auditory stimulation.

Figure 2

Experimental designs with timing of auditory stimulation and image acquisition. A: Sparse temporal sampling (STsamp) design. B: Event‐related with continuous scanning design (ERcont). C: Clustered volume acquisition design (BIG‐CVA). Lighter gray blocks reflect auditory/silence stimulation and darker gray (or yellow) blocks image acquisitions. D: ER65: Selecting the (in comparison to STsamp) time corresponding 65 scans (in gray/black‐yellow stripes) out of the ERcont design enables the direct comparison between the two designs and the direct assessment of the influence of SBN on signal intensities.

Figure 3

Imaging results for the three experimental designs and the subset ER65 (P < 0.025, corrected for multiple comparisons, false discovery rate (FDR)). Each row depicts one of the four designs in 3D rendering (left 2 columns) and in axial slices (right 3 columns) for activations in Table I. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]

Figure 4

Imaging results (random‐effects analysis) for the direct comparison between STsamp and ER65 (P < 0.05, corrected for multiple comparisons, FDR). [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]

Figure 5

Results for the ROI analysis for the three experimental designs and the subset ER65: (A) Heschl's gyrus (B) superior temporal gyrus.

Figure 6

Results for the ROI time point analysis for STsamp and ER65 for (A) left Heschl's gyrus (B) right Heschl's gyrus (C) left superior temporal gyrus (D) right superior temporal gyrus.

Figure 7

Imaging results for the cluster analysis for STsamp and ER65 (P < 0.05, corrected for multiple comparisons (FWE)) (A) cluster 1 (ITP0‐1), (B) cluster 2 (ITP2‐3), (C) cluster 3 (ITP4‐5). Cluster 4 did not reveal significant voxels for this threshold. 3D‐rendered images (left) and axial slices (right) are selected based on regions activated in experimental designs as shown in Table II. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]

Figure 8

Comparison of activation measured in auditory ROIs with CNR calculated for each design, based on response to auditory stimuli normalized to the STsamp design. ER65 shows much lower values than predicted and significantly lower values (see ROI Analysis) than STsamp due to masking and elevated baseline effects.