Analysis number | Figure | Description | Data structure | Type of test | Effect | p values | Power/CI |
---|---|---|---|---|---|---|---|

a | 1 | P1 component | Assumed normal | Repeated measures ANOVA | Hemisphere | 0.117 | 0.343 |

b | 1 | P1 component | Assumed normal | Repeated measures ANOVA | Identity | 0.39 | 0.447 |

c | 1 | P1 component | Assumed normal | Repeated measures ANOVA | Identity X hemisphere | 0.551 | 0.311 |

d | 1 | N170 component | Assumed normal | Repeated measures ANOVA | Hemisphere | 0.146 | 0.299 |

e | 1 | N170 component | Assumed normal | Repeated measures ANOVA | Identity | 0.513 | 0.373 |

f | 1 | N170 component | Assumed normal | Repeated measures ANOVA | Identity X hemisphere | 0.307 | 0.532 |

g | 1 | N250 component | Assumed normal | Repeated measures ANOVA | Hemisphere | 0.171 | 0.269 |

h | 1 | N250 component | Assumed normal | Repeated measures ANOVA | Identity | 0.001 | 0.980 |

i | 1 | N250 component | Assumed normal | Repeated measures ANOVA | Identity X hemisphere | 0.07 | 0.560 |

j | 1 | N250 component | Assumed normal | Repeated measures ANOVA | Identity in LH | 0.009 | 0.926 |

k | 1 | N250 component | Assumed normal | Repeated measures ANOVA | Identity in RH | 0.004 | 0.943 |

l | 2A | Group-based discrimination | Normality not assumed | Permutation test | Across expression | FDR-corrected p = 0.006 | |

m | 2B | Representative participant discrimination | Normality not assumed | Permutation test | Across-expression | FDR-corrected p = 0.004 | |

n | 3A | Group-based discrimination | Normality not assumed | Permutation test | Across expression/neutral | 0.001 | 95% CI: 47.1–51.3 |

o | 3A | Group-based discrimination | Normality not assumed | Permutation test | Across expression/happy | 0.001 | 95% CI: 47.2–50.9 |

p | 3A | Group-based discrimination | Normality not assumed | Permutation test | Within expression/neutral | 0.001 | 95% CI: 45.9–51.8 |

q | 3A | Group-based discrimination | Normality not assumed | Permutation test | Within expression/happy | 0.001 | 95% CI: 45.9–52.1 |

r | 3B | Single-participant-based discrimination | Assumed normal | Two-tailed t test | Across expression/neutral | 0.001 | 95% CI: 53.1–57 |

s | 3B | Single-participant-based discrimination | Assumed normal | Two-tailed t test | Across expression/happy | 0.001 | 95% CI: 53.2–57.2 |

t | 3B | Single-participant-based discrimination | Assumed normal | Two-tailed t test | Within expression/neutral | 0.001 | 95% CI: 0.559–0.602 |

u | 3B | Single-participant-based discrimination | Assumed normal | Two-tailed t test | Within expression/`happy | 0.001 | 95% CI: 54.8–60.4 |

v | n/a | Single-participant-based discrimination | Assumed normal | Repeated measures ANOVA | Discrimination type | <0.001 | >0.999 |

w | n/a | Single-participant-based discrimination | Assumed normal | Repeated measures ANOVA | Expression | 0.466 | 0.107 |

x | n/a | Single-participant-based discrimination | Assumed normal | Repeated measures ANOVA | Discrimination type X expression | 0.211 | 0.230 |

y | n/a | Single-participant-based discrimination within and across databases | Assumed normal | Repeated measures ANOVA | Discrimination type | <0.001 | >0.999 |

z | n/a | Single-participant-based discrimination within and across databases | Assumed normal | Repeated measures ANOVA | Pairs type | <0.001 | >0.999 |

aa | n/a | Single-participant-based discrimination within and across databases | Assumed normal | Repeated measures ANOVA | Discrimination type X pairs type | 0.033 | 0.565 |

ab | n/a | Single-participant-based discrimination within and across databases | Assumed normal | Post hoc (matched two-tailed t test) | Across-within database for within-expression versus across-within database for across-expression | 0.412 | 95% CI: –0.1–1.5 |

ac | n/a | Single-participant-based discrimination within and across databases | Assumed normal | Two-tailed t test against 0.5 | Within-expression, within-database discrimination | <0.001 | 95% CI: 54.6– 57.9 |

ad | n/a | Single-participant-based discrimination within and across databases | Assumed normal | Two-tailed t test against 0.5 | Within-expression, across-database discrimination | <0.001 | 95% CI: 56.4–61 |

ae | n/a | Single-participant-based discrimination within and across databases | Assumed normal | Two-tailed t test against 0.5 | Across-expression, within-database discrimination | <0.001 | 95% CI: 52.8–55 |

af | n/a | Single-participant-based discrimination within and across databases | Assumed normal | Two-tailed t test against 0.5 | Across-expression, across-database discrimination | <0.001 | 95% CI: 54–57.7 |

ag | n/a | temporally cumulative analysis | Assumed normal | Pearson’s correlation | EEG-based discrimination/behavioral discrimination | <0.001 | 0.436 |

ah | 4 | Temporal correlation | Assumed normal | Pearson’s correlation | Across-expression discrimination/behavioral discrimination | FDR-corrected p = 0.002 | |

ai | n/a | Temporally cumulative analysis | Assumed normal | Pearson’s correlation | Across-expression discrimination/CFMT | 0.219 | 0.237 |

aj | n/a | Temporally cumulative analysis | Assumed normal | Pearson’s correlation | Across-expression discrimination/VVIQ-2 | 0.676 | 0.05 |

ak | 5B | Face space fit | Normality not assumed | Permutation test | Happy-neutral similarity | <0.001 | 95% CI: 0.738–0.798 |

al | 6A | CIM | Normality not assumed | Permutation test | Neutral dimension 1; luminance | FDR-corrected p = 0.027 | |

am | 6A | CIM | Normality not assumed | Permutation test | Neutral dimension 1; red-green | FDR-corrected p = 0.018 | |

an | 6A | CIM | Normality not assumed | Permutation test | Neutral dimension 1; yellow-blue | FDR-corrected p = 0.024 | |

ao | 6A | CIM | Normality not assumed | Permutation test | Neutral dimension 2; luminance | FDR-corrected p = 0.012 | |

ap | 6A | CIM | Normality not assumed | Permutation test | Neutral dimension 2; red-green | N/A | |

aq | 6A | CIM | Normality not assumed | Permutation test | Neutral dimension 2; yellow-blue | FDR-corrected p = 0.009 | |

ar | 6B | CIM | Normality not assumed | Permutation test | Happy dimension 1; luminance | FDR-corrected p = 0.040 | |

as | 6B | CIM | Normality not assumed | Permutation test | Happy dimension 1; red-green | FDR-corrected p = 0.018 | |

at | 6B | CIM | Normality not assumed | Permutation test | Happy dimension 1; yellow-blue | FDR-corrected p = 0.024 | |

au | 6B | CIM | Normality not assumed | Permutation test | Happy dimension 2; luminance | N/A | |

av | 6B | CIM | Normality not assumed | Permutation test | Happy dimension 2; red-green | FDR-corrected p = 0.021 | |

aw | 6B | CIM | Normality not assumed | Permutation test | Happy dimension 2; yellow-blue | FDR-corrected p = 0.008 | |

ax | 7B | Temporal reconstruction accuracy | Normality not assumed | Permutation test | Neutral | FDR-corrected p = 0.002 | |

ay | 7B | Temporal reconstruction accuracy | Normality not assumed | Permutation test | Happy | FDR-corrected p = 0.006 | |

bz | 8B | Reconstruction accuracy (image-based) | Normality not assumed | Permutation test | Neutral | 0.001 | 95% CI: 42.3–57.8 |

ba | 8B | Reconstruction accuracy (image-based) | Normality not assumed | Permutation test | Happy | 0.001 | 95% CI: 42.1–58.1 |

bb | 8C | Reconstruction accuracy (experimental-based) | Assumed normal | Two-tailed t test | Neutral | 0.001 | 95% CI: 55.6–62.6 |

bc | 8C | Reconstruction accuracy (experimental-based) | Assumed normal | Two-tailed t test | Happy | 0.001 | 95% CI: 52.4–59.2 |

bd | n/a | Correlation between experimental and image-based accuracies | Assumed normal | Pearson’s correlation | Neutral | 0.001 | 0.912 |

be | n/a | Correlation between experimental and image-based accuracies | Assumed normal | Pearson’s correlation | Happy | 0.002 | 0.898 |

bf | n/a | Correlation between reconstruction and discrimination | Assumed normal | Pearson’s correlation | Averaged across expressions | <0.001 | >0.999 |

bg | n/a | Single-participant-based reconstruction accuracy (image-based) | Assumed normal | Two-tailed t test | Neutral | 0.027 | 0.676 |

bh | n/a | Single-participant-based reconstruction accuracy (image-based) | Assumed normal | Two-tailed t test | Happy | 0.045 | 0.580 |

bi | n/a | Correlation between single-participant-based reconstruction and discrimination | Assumed normal | Pearson’s correlation | Neutral | <0.001 | 0.974 |

bj | n/a | Correlation between single-participant-based reconstruction and discrimination | Assumed normal | Pearson’s correlation | Happy | <0.001 | 0.980 |