Long-term FISL improves fMRI-based classification of emotional faces but not emotional scenes¶

Claim · All claims
Evidence · 1 study (2 entries) · causal r1 · q3 · i0 negligible
Currency · newest study 2025 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q3 i? SVM classification accuracy for positive vs. neutral faces using IFG increased significantly post-FISL (AUC increase 0.134), and negative-face accuracy improved across ROIs. → Qi Liu 2025 q3 i0 FISL did not improve classification of emotional scene pictures (permutation ps > 0.05). → Qi Liu 2025 (2)

Evidence¶

Qi Liu 2025¶

Qi Liu, Siyu Zhu, Can Liu, Yanmiao Yang, Chunmei Lan, Xinwei Song, Benjamin Becker, Keith M. Kendrick, Xinqi Zhou & Weihua Zhao. (2025). Neural signals gradually shift towards inferior frontal gyrus and long-term effects following facial imitation-based social learning. npj Science of Learning. https://doi.org/10.1038/s41539-025-00372-9

q3 · i? · causal · r1

Pre/post fMRI multivariate pattern analysis (linear SVM, leave-one-subject-out cross-validation) on emotional face and scene perception tasks in the same 80 participants: post-FISL face classification improved, with the quoted IFG AUC increase of 0.134 for positive vs. neutral faces.

"10,000 times permutation tests showed that the accuracy of using the IFG as a restriction region for predicting positive and neutral expression faces was significantly higher in the post-FISL than inthe pre-FISL session (increase in AUC: 0.134, permutation test p = 0.047)"

Qi Liu 2025 (2)¶

Qi Liu, Siyu Zhu, Can Liu, Yanmiao Yang, Chunmei Lan, Xinwei Song, Benjamin Becker, Keith M. Kendrick, Xinqi Zhou & Weihua Zhao. (2025). Neural signals gradually shift towards inferior frontal gyrus and long-term effects following facial imitation-based social learning. npj Science of Learning. https://doi.org/10.1038/s41539-025-00372-9

q3 · i0 · causal · r1

Same SVM transfer analysis on the IAPS emotional scene task: the article reports no improvement in predicting emotional scenes after the FISL course (permutation ps > 0.05), establishing face-specificity of the transfer effect; equivalence was not tested.

"However, FISL course did not improve theaccuracy of predicting emotional scene pictures (permutation testps>0 . 0 5 ,F i g .4c, d)"

Discussion¶

Related Claims¶

• Long-term FISL increases informational connectivity between the MNS/IFG and social-cognition and reward regions during positive face perception — related
• Neural pattern similarity within the MNS increases across FISL sessions, more for positive emotions, and tracks behavioral improvement — related
• Imitative learning performance improves across repeated FISL sessions, with a positive-emotion advantage over negative emotions — related
• Early-phase FISL improves whole-process imitation performance while late-phase FISL selectively speeds successful imitation of positive emotions — related
• Information flow within the MNS progressively shifts toward IFG dominance over repeated FISL sessions — related
• Improved positive imitation performance after FISL is mediated by increased MNS neural pattern similarity and MNS–OFC informational connectivity — related