Table S3.

Statistical tests, by figure

Fig. no.TestHypothesisSection/paragraph no.nResult
Exact valueDefined?Test statisticP value
S1APaired t testGroup mean FC difference before global signal regressionResults: Abnormal FC Patterns in Stroke, para. 152,326Edgesdf = 52325P = 0
t = 139.7
S1APaired t testGroup mean FC difference after global signal regressionResults: Abnormal FC Patterns in Stroke , para. 152,326Edgesdf = 52325P = 0.76
t = -0.31
ANOVAHomotopic FC by group, RSN, and head motionResults: Abnormal FC Patterns in Stroke , para. 1100, 27Patients, controls1124group: P = 3 × 10-22, RSN: P = 4 × 10−7, FD: P = 0.26
1Unpaired t testGroup homotopic FC differenceResults: Abnormal FC Patterns in Stroke , para. 2100, 27Patients, controlsdf = 124P = 0.000040
t = 4.26
1Unpaired t testGroup ipsilesional FC differenceResults: Abnormal FC Patterns in Stroke , para. 2100, 27Patients, controlsdf = 124P = 0.26
t = 1.13
1Unpaired t testGroup contralesional FC differenceResults: Abnormal FC Patterns in Stroke , para. 2100, 27Patients, controlsdf = 124P = 0.32
t = 1.00
1Unpaired t testGroup ipsi DAN-DMN FC differenceResults: Abnormal FC Patterns in Stroke , para. 2100, 27Patients, controlsdf = 124P = 0.0021
t = -3.15
S1BNonparametric permutation testing of t-statMultiple intranetwork FC differencesResults: Abnormal FC Patterns in Stroke , para. 227 tests - 10,000 permsPatients, controlsn/a7/27 P < 0.0017
S1BNonparametric permutation testing of t-statMultiple internetwork FC differencesResults: Abnormal FC Patterns in Stroke , para. 356 tests - 10,000 permsPatients, controlsn/a1/56 P < 0.0019
1Pearson correlationDAN homotopic FC vs. ipsi DAN-DMN FCResults: Abnormal FC Patterns in Stroke , para. 3100Stroke patientsdf = 98,P = 1 × 10−12
r = −0.61
1Pearson correlationDAN homotopic FC vs. ipsi DAN-DMN FCResults: Abnormal FC Patterns in Stroke, para. 327Controlsdf = 25,P = 0.19
r = 0.25
1Fisher r-to-z comparisonDAN homotopic FC vs. ipsi DAN-DMN FC, different in patientsResults: Abnormal FC Patterns in Stroke, para. 3100, 27Patients, controlsdf = 124,P = 0
z = −4.23
Pearson correlationLesion size vs. homotopic FCResults: Abnormal FC Patterns in Stroke, para. 4100Stroke patientsdf = 98,P = 6 × 10−7
r = 0.46
Pearson correlationLesion predicted vs. actual homotopic FCResults: Abnormal FC Patterns in Stroke, para. 4100Stroke patientsdf = 98,P = 7 × 10−7
r = 0.46
3Permutation testingAttention Lesion-deficit modelResults: Prediction of Behavioral Deficits Based on Lesion and FC, para. 280Stroke patientsr2 = 0.323P = 4 × 10−4
3Permutation testingAttention FC-deficit modelResults: Prediction of Behavioral Deficits Based on Lesion and FC, para. 280Stroke patientsr2 = 0.450P < 1 × 10−4
3Permutation testingVisual mem lesion-deficit modelResults Results: Prediction of Behavioral Deficits Based on Lesion and FC, para. 278Stroke patientsr2 = 0.109P = 9 × 10−4
3Permutation testingVisual mem FC-deficit modelResults: Prediction of Behavioral Deficits Based on Lesion and FC, para. 278Stroke patientsr2 = 0.364P < 1 × 10−4
3Permutation testingVerbal mem Lesion-deficit modelResults: Prediction of Behavioral Deficits Based on Lesion and FC, para. 279Stroke patientsr2 = 0.187P = 1.5 × 10−3
3Permutation testingVerbal mem FC-deficit modelResults: Prediction of Behavioral Deficits Based on Lesion and FC, para. 279Stroke patientsr2 = 0.416P < 1 × 10−4
3Permutation testingLanguage lesion-deficit modelResults: Prediction of Behavioral Deficits Based on Lesion and FC, para. 298Stroke patientsr2 = 0.646P < 1 × 10−4
3Permutation testingLanguage FC-deficit modelResults: Prediction of Behavioral Deficits Based on Lesion and FC, para. 298Stroke patientsr2 = 0.511P < 1 × 10−4
3Permutation testingL motor lesion-deficit modelResults: Prediction of Behavioral Deficits Based on Lesion and FC, para. 291Stroke patientsr2 = 0.469P < 1 × 10−4
S6Permutation testingL motor FC-deficit modelResults: Prediction of Behavioral Deficits Based on Lesion and FC, para. 291Stroke patientsr2 = 0.136P = 1.1 × 10−3
S6Permutation testingR motor lesion-deficit modelResults: Prediction of Behavioral Deficits Based on Lesion and FC, para. 291Stroke patientsr2 = 0.338P < 1 × 10−4
S6Permutation testingR motor FC-deficit modelResults: Prediction of Behavioral Deficits Based on Lesion and FC, para. 291Stroke patientsr2 = 0.325P < 1 × 10−4
S6Permutation testingL visual lesion-deficit modelResults: Prediction of Behavioral Deficits Based on Lesion and FC, para. 253Stroke patientsr2 = 0.331P = 4 × 10−4
S6Permutation testingL visual FC-deficit modelResults: Prediction of Behavioral Deficits Based on Lesion and FC, para. 253Stroke patientsr2 = 0.110P = 0.0104
S6Permutation testingR visual lesion-deficit modelResults: Prediction of Behavioral Deficits Based on Lesion and FC, para. 253Stroke patientsr2 = 0.603P = 1 × 10−4
S6Permutation testingR visual FC-deficit modelResults: Prediction of Behavioral Deficits Based on Lesion and FC, para. 253Stroke patientsr2 = 0.081P = 0.0902
3Wilcoxon paired sample signed rank testAttention FC-deficit model error vs. lesion-deficit modelResults: Prediction of Behavioral Deficits Based on Lesion and FC, para. 280Prediction errorsSR = 1343P = 0.074
3Wilcoxon paired sample signed rank testVisual mem FC-deficit model error vs. lesion-deficit model errorResults: Prediction of Behavioral Deficits Based on Lesion and FC, para. 378Prediction errorsSR = 1748P = 0.015
3Wilcoxon paired sample signed rank testVerbal mem FC-deficit model error vs. lesion-deficit model errorResults: Prediction of Behavioral Deficits Based on Lesion and FC, para. 379Prediction errorsSR = 2062P = 0.007
3Wilcoxon paired sample signed rank testLanguage FC-deficit model error vs. lesion-deficit model errorResults: Prediction of Behavioral Deficits Based on Lesion and FC, para. 398Prediction errorsSR = 2205P = 0.21
3Wilcoxon paired sample signed rank testMotor FC-deficit model error vs. lesion-deficit model errorResults: Prediction of Behavioral Deficits Based on Lesion and FC, para. 391Prediction errorsSR = 1501P = 0.009
3Wilcoxon paired sample signed rank testVisual FC-deficit model error vs. lesion-deficit model errorResults: Prediction of Behavioral Deficits Based on Lesion and FC, para. 353Prediction errorsSR = 466P = 0.013
4ANOVAConnection weights (+/− within/inter hem)Results: Topography of Behaviorally Predictive FC, para. 28FC-deficit modelsdf = 31P = 1.6 × 10−6
F = 38.03
6Pearson correlationMultidomain FC-deficit modelResults: Prediction of Common Behavioral Impairment, para. 1623Patients x domainsr2 = 0.287P < 1 × 10−30
S2Pearson correlationHomotopic FC vs. framewise displacementResults: Abnormal FC Patterns in Stroke, para. 5100Stroke patientsr = −0.15P = 0.13
S2Pearson correlationHomotopic FC vs. eyes openResults: Abnormal FC Patterns in Stroke, para. 5100Stroke patientsr = 0.08P = 0.5
S2Pearson correlationHomotopic FC vs. hemodynamic lagResults: Abnormal FC Patterns in Stroke, para. 5100Stroke patientsr = −0.15P = 0.13
  • Note that prediction of raw performance measures in Fig. S5 is not included here because prediction accuracy was assessed qualitatively and statistical significance tests are not reported. df, degrees of freedom; F, F-statistic; r, Pearson’s rho; SR, signed rank; t, t-statistic; z, z-statistic.