Hemodynamic effects of sex and handedness on the Wisconsin Card Sorting Test: the contradiction between neuroimaging and behavioural results

Figures

• 

  Figure 1. Estimated marginal means of hemodynamic response parameters (A, HbO2; B, HbR; C, HbT; D, OX) regarding sex and handedness. Box and the star indicate significant differences between groups, (p < 0.05). Results are from 70 participants’ two-way RM-MANOVA tests (female/male, n = 35/35; right-handed/left-handed, n = 40/30; subsequently right-handed female/male, n = 20/20, and left-handed female/male, n = 15/15).

• 

  Figure 2. Spatial differences of HbO2 concentration between groups of participants during the WCST. (A) Error plot of HbO2 mean (±SD) concentrations of HbO2 between participants’ groups in each of 16 optodes, and the mean (±SD) concentrations of HbO2 across optodes per group. Participants groups are: female right-handed, n = 20, F-R; female left-handed, n = 15, F-L; male right-handed, n = 20, M-R; male left-handed, n = 15, M-L). (B) Statistical activation maps (t-values, p < 0.05) display where the obtained HbO2 concentrations were significantly higher for left-handed males than right-handed females (t (33) = 3.039, p < 0.05) and left-handed females (t (28) = 3.230, p < 0.05). Significant optodes are outlined. Statistical values were spatially visualised using the BSpline18 interpolations method. The measured areas are marked with ellipses.

• 

  Figure 3. The distribution of oxygenation for participants’ groups (A, right-handed women; B, left-handed women; C, right-handed men; D, left-handed men) during the performance of WCST. Visualisation of time-series oxygenation changes, together with mean spatial signal distribution on frontal brain view. All 70 participants were decomposed into four groups by gender and handedness. Time-series of OXY concentrations is normalised by the max value. The mean signal is shown below. Recordings were also normalised in time (from the start of WCST being 0, and to the end of WCST being 1) individually and later averaged per group. Note that time series presentation does not correspond directly to the frontal brain view distribution below due to double line probe geometry. Values were spatially visualised with BSpline18 interpolations method.