Fig. 3: Target-specific changes in perceptual outcomes during human thalamic LIFU application.
Perceptual outcomes that demonstrated significant omnibus effects, measured at (a, b) 70% and c 5% DC, including (a) sensitivity d’, b categorization accuracy for real images, and c criterion c, are shown. Statistical significance was first evaluated using an omnibus F-test from the LME models, followed by two-sided post-hoc pairwise comparisons using coefficient tests. P-values were adjusted for multiple comparisons using FDR method (α = 0.05). Asterisks and pounds denote significance of post-hoc effects (asterisk: pFDR < 0.05; pound: uncorrected p < 0.05). FDR-corrected p-values of significant pairwise comparisons (pFDR < 0.05) are as follows; a Baseline vs. VA: 0.0246, VA vs. VP: 0.0246, VA vs. DA: 0.0034, VA vs. DP: 0.0119; b Baseline vs. VA: 0.0119, Baseline vs. VP: 0.0049, Baseline vs. DA: 0.0012, Baseline vs. DP: 0.0007; c VA vs. DP: 0.0182, VP vs. DP: 0.0125. Boxes indicate interquartile ranges. Median values are marked by white circles. Dashed lines are reference values for each measure: a no discrimination of d’ = 0, b chance accuracy of 0.25, and c no bias of c = 0. For (a), Baseline: n = 25; VA: n = 23; VP: n = 23; DA: n = 21; DP: n = 20; for panel b, Baseline: n = 27; VA: n = 26; VP: n = 26; DA: n = 27; DP: n = 27; for panel c, Baseline: n = 24; VA: n = 21; VP: n = 19; DA: n = 18; DP: n = 20. DC duty cycle, FA false alarm, VA ventroanterior thalamus, VP ventroposterior thalamus, DA dorsoanterior thalamus, DP dorsoposterior thalamus. Statistics and source data are provided as a Source Data file.
