Open Access Highly Accessed Research article

Ultra-fast speech comprehension in blind subjects engages primary visual cortex, fusiform gyrus, and pulvinar – a functional magnetic resonance imaging (fMRI) study

Susanne Dietrich*, Ingo Hertrich and Hermann Ackermann

Author Affiliations

Center for Neurology/Department of General Neurology, Hertie Institute for Clinical Brain Research, University of Tübingen, Hoppe-Seyler-Str. 3, D-72076, Tübingen, Germany

For all author emails, please log on.

BMC Neuroscience 2013, 14:74  doi:10.1186/1471-2202-14-74

Published: 23 July 2013

Additional files

Additional file 1:

An example of ultra-fast speech: “Die Billigöfen aus dem Baumarkt scheiden bei einer Umweltbewertung deutlich schlechter ab als etwa Holzpelletöfen, die mit einer steuerbaren Verbrennungsluftregelung und anderen Mechanismen ausgestattet sind.“

Format: WAV Size: 162KB Download file

Open Data

Additional file 2:

An example of moderately fast speech: “Hinzu kommt eine mangelnde Kompetenz vieler Hausärzte und schlechte Versorgungsstrukturen.“

Format: WAV Size: 162KB Download file

Open Data

Additional file 3:

An example of reversed ultra-fast speech (see Additional file 1).

Format: WAV Size: 162KB Download file

Open Data

Additional file 4:

An example of reversed moderately fast speech (see Additional file 2).

Format: WAV Size: 162KB Download file

Open Data

Additional file 5:

An example from the repetition task concerning ultra-fast speech comprehension in a blind listener with more than 90% correctly reproduced words: “Öfen, die die Grenzwerte einhalten, kosten zwischen 500 und 700.”

Format: WAV Size: 612KB Download file

Open Data

Additional file 6:

Coordinates of the whole-head analysis on the impact of speaking rate on hemodynamic brain activation: SPM T-contrasts of each condition (ultra-fast, moderately fast/forward, reversed) versus baseline and vice versa. Displayed are the responses exceeding a threshold of p < .001 (uncorrected) at a voxel level and p < .05 (corrected) at a cluster level, including an extent threshold of k (contiguous voxels).

Format: DOCX Size: 35KB Download file

Open Data

Additional file 7:

Whole-head between-group analysis (blind versus sighted), including each of the various experimental conditions (versus baseline). Displayed are the responses exceeding a threshold of p < .001 (uncorrected) at a voxel level.

Format: TIFF Size: 82KB Download file

Open Data

Additional file 8:

Coordinates of the whole-head between-group analysis (blind versus sighted, experimental conditions versus baseline). Hemodynamic responses exceeding a threshold of p < .001 (uncorrected) at a voxel level and p < .05 (corrected, k = 68) at a cluster level are displayed, in addition, the activation of further interesting regions, though non-significant at the level of the corrected threshold.

Format: DOCX Size: 22KB Download file

Open Data

Additional file 9:

Coordinates of the whole-head between-group analysis (blind versus sighted), comparing late- and early-blind individuals versus sighted controls each (SPM T-contrasts of the condition “all versus baseline”), displayed are the hemodynamic responses exceeding a threshold of p < .005 (uncorrected) at a voxel level and p < .05 (corrected) at a cluster level as well as the activation of some further relevant regions, though non-significant at the level of the corrected threshold (k ≥ 15).

Format: DOCX Size: 23KB Download file

Open Data

Additional file 10:

Whole-head covariance analysis across (i) all subjects, (ii) early-blind individuals removed, and (iii) exclusively late-blind participants (see below). SPM T-contrasts identified the correlation between BOLD responses and ultra-fast speech comprehension capabilities, based upon the condition “ultra-fast versus baseline”. Upper row: 3 early-blind (EB) + 11 late-blind (LB) + 12 sighted subjects (SI), threshold p < .001 (uncorrected) at a voxel level; middle row: 11 LB + 12 SI, threshold p < .001 (uncorrected) at a voxel level; lower row: 11 LB, threshold p < .05 (uncorrected) at a voxel level.

Format: TIFF Size: 121KB Download file

Open Data

Additional file 11:

Coordinates of the whole-head covariance analysis across (i) all subjects, (ii) early-blind individuals removed, and (iii) exclusively late-blind participants (see Additional file 10). SPM T-contrasts identified the correlation between BOLD responses and ultra-fast speech comprehension capabilities, based upon the condition “ultra-fast versus baseline”. Displayed are the hemodynamic responses exceeding a threshold of p < .005 (uncorrected) at a voxel level and p < .05 (corrected) at a cluster level, in addition, activation of some further regions, though non-significant at the level of the corrected threshold (across all subjects: k ≥ 70; LB + SI: k ≥ 10), is shown.

Format: DOCX Size: 23KB Download file

Open Data

Additional file 12:

Region-of-interest (ROI) analyses exemplified for three areas, i.e., right-hemispheric primary visual cortex (V1), left-hemispheric fusiform gyrus (FG), and left-hemispheric inferior frontal gyrus (IFG). Displayed is the strength of hemodynamic responses (% signal change) is displayed within the respective ROIs during each of the following conditions (versus baseline): uf = ultra-fast speech, mf = moderately fast speech, rev-mf = reversed moderately fast speech, rev-uf = reversed ultra-fast speech (error bars = standard error of the mean across subjects; 14 blind, 12 sighted individuals; asterisk = significant (one-sample T-test) Bold responses).

Format: TIFF Size: 192KB Download file

Open Data

Additional file 13:

Correlations between signal change (%) and behavioral performance, onset and duration of blindness, taking into account moderately fast and ultra-fast speech materials. Values indicate correlations (two-tailed Pearson test) between the moderately fast (mf) or ultra-fast (uf) speech condition (versus baseline) and behavioral performance, onset and duration of vision loss. Upper values: correlation coefficient r; lower values in parentheses: significance p; bold numbers: significant results at the threshold p < .05

Format: DOCX Size: 18KB Download file

Open Data