Association between carotid atherosclerosis and brain activation patterns during the Stroop task in older adults: An fNIRS investigation

Figures

• 

  Fig. 1. An example of the channel position of the Brite 24 cap obtained with the Pohemus digitization. Sources are shown in orange (long separation channels) and red (short separation channels). Detectors are shown in blue. Note that the sources and detectors are arranged in quadrants over the left and right somatosensory cortices and prefrontal cortices.

• 

  Fig. 2. An example of unprocessed (blue) and fully processed (red) fNIRS data obtained from a participant performing the verbal Stroop task. The trace shown is ΔO2Hb in units of mol/L. The stimulus periods (Stroop blocks 1, 2, and 3) are indicated by the yellow rectangles superimposed over the fNIRS traces.

• 

  Fig. 3. An example of how the t statistic map superimposed over a rendered brain was subdivided into the four anatomical regions used in the statistical analysis. Region 1 = left prefontal cortex (LPFC), Region 2 = right prefrontal cortex (RPFC), Region 3 = left sensorimotor cortices (LMC), and Region 4 = right sensorimotor cortices(RMC).

• 

  Fig. 4. Shows the mean t statistic maps for each group (Healthy and Plaque) when comparing the incongruous Stroop to the nominal Stroop for oxygenated haemoglobin (O2Hb) and deoxygenated haemoglobin (HHB). The higher the t statistic, the greater evidence of increased change in Hb (rise in O2Hb, fall in HHB), and thus functional brain activity. Note that the average t statistic is significantly lower overall in the Plaque group compared with the Healthy group in O2Hb as indicated by the pink asterisk. Also note the significant reduction in t statistic in the LPFC and LMC in the Plaque group compared with the Healthy group in HHB as indicated by the pink arrows.

• 

  Fig. 5. Individual t statistic maps showing the brain activation pattern for all 33 participants in the Healthy group performing the Stroop task (incongruous vs. null condition). The t statistic is indicated by the colorbar, with brighter colors indicating a high t value and thus greater evidence for NVC.

• 

  Fig. 6. Example channels from each of the four brain regions assessed for an example participant in the (left) Healthy group and the (right) Plaque group. The Hb signals shown here have been temporally pre-processed as described in Section 2.5.1, but otherwise reflect the raw data (i.e. before short separation channel regression with the GLM). Note that overall, the OHB signal (red lines) increases to a greater extent in response to the incongruous condition for the participant in the Healthy group compared to the participant in the Plaque group. Also note the increased drop in HHB (blue lines) in the LPFC in the Healthy participant compared with the Plaque participant (magenta arrows).

• 

  Fig. 7. Top row: Example preprocessed channels from the left prefrontal cortex (LPFC) of a participant in the Healthy group with O2Hb and HHB shown in red and blue respectively. Note that in response to the stimulus periods (yellow blocks), the O2Hb signal increases to a greater extent corresponding decrease in HHB. Bottom row: The same channels as above are duplicated below, this time with the corresponding O2Hb (magenta) and HHB (cyan) measures from the short separation channel superimposed. Note that the O2Hb short separation channel responds in a similar way to the O2Hb long separation channel throughout the Stroop task, whereas the HHB short separation channel maintains a relatively flat shape. This implies that O2Hb is more sensitive to extracerebral contamination than HHB.