MEG-to-MEG Transfer Learning and Cross-Task Speech/Silence Detection with Limited Data

• Per-subject MEG speech-decoding data are typically limited to minutes, constraining practical BCI deployment.Source: §1 · p. 1
• Transfer learning has not previously been demonstrated for MEG-based speech decoding despite progress in other neuroimaging modalities.Source: §1 · p. 1
• The study pre-trains a Conformer-based model on 50 hours of single-subject listening data and fine-tunes it using 5 minutes per subject across 18 participants.Source: §1 · p. 1
• Transfer learning improves in-task performance and cross-task generalization across speech perception and production, with gains of 1-6% across metrics.Source: §1 · p. 1
• Production-trained models decode passive listening above chance, indicating representations shared across speech tasks rather than task-specific motor activity alone.Source: §1 · p. 1

• MEG speech decoding has progressed from closed-vocabulary word classification to phone-level analysis and large-scale competitions.Source: §2 · p. 1
• EEG studies have used self-supervised and contrastive pre-training to improve cross-subject generalization, unlike prior MEG speech-decoding work.Source: §2 · p. 1
• Transfer learning had not been demonstrated for MEG-based speech decoding, leaving its value for limited per-subject clinical BCI data unresolved.Source: §2 · pp. 1–2
• Cross-task speech decoding between perception and production remains underexplored, including the roles of motor-related activity and shared speech representations.Source: §2 · p. 2

Datasets

• The evaluation combines over 50 hours of single-subject English listening data with approximately 5 minutes per task from 18 native Spanish-speaking participants.Source: §3.1 · p. 2
• The contrasting datasets test whether single-subject pre-training improves decoding for new subjects performing different speech tasks with limited data.Source: §3.1 · p. 2

Model Architecture

• MEGConformer is a compact Conformer encoder operating directly on 306-channel windowed raw MEG segments sampled at 250 Hz.Source: §3.2 · p. 2
• Fine-tuning adds validation-loss checkpoint selection, RollAugment temporal shifts, soft speech-fraction targets, and shorter early-stopping patience.Source: §3.2 · p. 2

Experimental Protocol

• Models are evaluated in-task and across all six train-test pairings among Listen, Playback, and Production, comparing transfer learning with training from scratch.Source: §3.3 · p. 2
• Performance is assessed with F1-macro, balanced accuracy, and AUC-macro, using paired Wilcoxon tests with Holm-Bonferroni correction across subjects.Source: §3.3 · p. 2

In-Task Transfer Learning

• +3.7% accuracy, +2.6% F1, and +7.3% AUC were achieved for listening with transfer learning.These gains were statistically significant (W = 17.0, p = 0.005).Source: §4.1 · p. 3

Cross-Task Decoding Baseline

• 65.0%-73.4% cross-task accuracy was achieved without transfer learning, with all cross-task decodings significantly above chance.Perception-to-perception transfer was stronger than production-to-perception transfer.Source: §4.2 · p. 3, Table 2 · p. 3

Cross-Task with Transfer Learning

• +6.1% accuracy and +4.2% F1 improved listen-to-playback transfer, while playback-to-listen improved by +6.3% accuracy and +4.1% F1.Both directions also showed significant AUC gains, and the overall transfer benefit was highly significant (p < 0.001).Source: §4.3 · p. 3

Transfer Effect Size

• 1.7-3.5% cross-task gains exceeded the 0.5-2.2% gains observed for in-task decoding.Cross-task decoding involving production showed greater variability across subjects.Source: §4.3 · p. 3

Cross-Task with Transfer Learning

• 85.3% F1 for listen-to-production exceeded 80.1% F1 for production-to-listen, revealing directional asymmetry in cross-task transfer.Listening and playback transferred bidirectionally with similar F1 performance, whereas production-involving pairs showed clearer directional preferences.Source: §4.3 · pp. 3–4, Figure 2 · p. 4

Subject-Level Effects

• 15 of 18 perception subjects and 16 of 18 production subjects improved with transfer learning, although subject-level effects varied substantially.Subject 16 showed a marked negative production effect of -13.3% F1.Source: §4.3 · p. 4

Discussion and Conclusion

• Transfer learning enhanced MEG speech detection across new subjects and tasks despite only a few minutes of subject-specific data.The findings indicate that pre-training improved performance beyond task- or subject-specific training patterns.Source: §5 · p. 4
• Models trained on perception and production data showed asymmetric transfer, with production-to-perception decoding consistently weaker than the reverse direction.Perception tasks transferred bidirectionally with comparable performance, whereas production-to-perception transfer was weaker.Source: §5 · p. 4
• Figure 3 compares each subject’s in-task F1 improvement from transfer learning across the three tasks.Each bar represents transfer learning minus training from scratch; colors distinguish tasks.Source: Figure 3 · p. 4
• Cross-task gains were larger than in-task gains, with statistically reliable improvements emerging primarily for generalization between speech perception and production.The authors frame cross-task improvement as the more important outcome for transfer learning.Source: §5 · p. 4
• The study is limited to speech detection, uses English pre-training and Spanish fine-tuning, relies on one pre-training subject, and finds modest, variable improvements.The authors characterize transfer learning as complementing rather than replacing subject-specific adaptation.Source: §5 · p. 4