Effective Connectivity in Cortical Networks During Deception: A Lie Detection Study Based on EEG

Area of Science:

• Neuroscience
• Cognitive Science
• Brain-Computer Interfaces

Background:

• Deception involves complex brain activity, but communication patterns between brain regions during deceit remain poorly understood.
• Investigating brain connectivity during deception is crucial for developing objective lie detection methods.

Purpose of the Study:

• To explore the most important information flows (MIIFs) between different brain cortices during deception.
• To identify unique brain connectivity patterns associated with deception using electroencephalogram (EEG) data.
• To assess the efficacy of effective connectivity (EC) analysis for lie detection.

Main Methods:

• Recorded 64-channel EEG signals from 30 participants (15 guilty, 15 innocent) during a guilty knowledge test.
• Estimated cortical current density waveforms and applied partial directed coherence (PDC) for EC analysis across delta, theta, alpha, and beta frequency bands.
• Utilized graph theoretical analysis to extract network parameters for distinguishing between guilty and innocent groups.

Main Results:

• High classification accuracy was achieved using network parameters from all four frequency bands, indicating suitability for lie detection.
• Identified specific brain 'hub' regions and significant differences in MIIFs between guilty and innocent groups.
• The fronto-parietal network emerged as the most prominent network for MIIFs across all analyzed frequency bands.

Conclusions:

• EEG-based EC analysis, particularly focusing on the fronto-parietal network, provides a viable method for lie detection.
• The identified MIIFs and brain hubs offer insights into the neural mechanisms underlying deception.
• Further analysis of MIIFs across frequency bands can elucidate the cognitive processes involved in deceptive behavior.