Neural Decoding of Multi-Modal Imagery Behavior Focusing on Temporal Complexity

Area of Science:

• Neuroscience
• Cognitive Science
• Biophysics

Background:

• Mental imagery involves visual, auditory, and motor modalities, with alterations linked to psychiatric disorders.
• Previous research correlates imagery with modality-specific and supramodal brain region activity.
• Complexity analysis, though available, remains underutilized in neural decoding for imagery.

Purpose of the Study:

• To characterize neural oscillations in multimodal imagery using complexity-based neural decoding.
• To investigate the time evolution of temporal complexity for neural decoding applications.
• To explore modality-independent mechanisms underlying mental imagery.

Main Methods:

• Modified complexity measures to analyze temporal complexity evolution.
• Utilized magnetoencephalography (MEG) data from healthy subjects performing imagery and non-imagery tasks.
• Applied Hilbert-Huang transform for data decomposition and calculated complexity values for decoding.

Main Results:

• Intra-subject decoding using complexity achieved high accuracy, generating characteristic sensitivity maps for visual perception (VP), visual imagery (VI), motor execution (ME), and motor imagery (MI).
• Sensitivity maps showed inverted patterns in occipital regions (VP vs. VI) and central regions (ME vs. MI).
• Two-class (imagery vs. non-imagery) and four-class classifications performed better with complexity than raw data or band power.

Conclusions:

• Complexity measures are effective for neural decoding of mental imagery.
• Findings suggest the existence of modality-independent mechanisms for imagery.
• Time evolution of temporal complexity analysis can enhance understanding of hierarchical brain functions.