Task design and analysis methods.

(A) Movie structure. A 6.5-minute short film was created by editing fragments from Pippi on the Run into a coherent narrative. The movie consisted of 13 interleaved blocks of videos accompanied by speech or music. (B) Data analysis schematic. Standard analysis pipeline for extracting emotion features from the movie and constructing encoding model to predict iEEG responses while participants watching the short film.

Prediction performance of encoding models in DLPFC.

(A) Spatial distribution of electrodes in DLPFC. Electrodes in all participants from each group are projected onto MNI space and shown on the average brain. Red shaded areas indicate middle frontal cortex provided by the FreeSurfer Desikan-Killiany atlas25. Electrodes outside DLPFC are not shown. (B) The average prediction accuracy across participants for speech condition. The performance of encoding model is measured as Pearson correlation coefficient (r) between measured and predicted brain activities. (C) Prediction accuracy difference between speech condition and music condition for each group. Error bars are standard error of the mean. *P<0.05; **P<0.01.

Prediction performance of encoding models in pSTC.

(A) The electrode distribution of two children (s19 and s39). Electrodes in pSTC are green. (B) Prediction accuracy of encoding models in the two children. (C) Spatial distribution of recording contacts in post-childhood participants’ pSTC. The pSTC electrodes identified in individual space are projected onto MNI space and shown on the average brain. Contacts other than pSTC are not shown. Blue shaded areas indicate superior temporal cortex provided by the FreeSurfer Desikan-Killiany atlas25. (D) Average prediction accuracy across post-childhood participants. Error bars are standard error of the mean. **P<0.01.

Correlation between encoding weights and age.

(A) Left: Correlation between averaged encoding weights of five complex emotions and age. Right: Correlation between averaged encoding weights of six basic emotions and age. (B) Pearson correlation coefficient between encoding weights of 48 facial expression features and age. The results are ranked from largest to smallest. Significant correlations noted with *(P<0.05, uncorrected) or **(P<0.01, uncorrected). (C) Correlation between encoding weights of embarrassment, pride, guilt, interest and age (N=12).

Demographic information of childhood group.

Demographic information of post-childhood group.