Detecting spontaneous deception in the brain.

Area of Science:

• Neuroscience
• Cognitive Psychology
• Forensic Science

Background:

• Deception detection is crucial for legal investigations.
• Prior research on neural deception signatures is limited by difficulty confounds, potentially conflating deception with cognitive effort.
• The generalizability of deception detection across instructed versus spontaneous lies and between participants remains underexplored.

Purpose of the Study:

• To investigate neural correlates of deception while controlling for task difficulty.
• To examine the generalizability of deception detection across instructed and spontaneous responses.
• To assess the feasibility of cross-participant deception detection.

Main Methods:

• A dual-task paradigm was employed to balance task difficulty between truth-telling and lying.
• Instructed and spontaneous truthful and deceptive responses were collected independently.
• Multivoxel pattern analysis (MVPA) was used to decode neural patterns associated with deception.

Main Results:

• The angular gyrus (AG), inferior frontal gyrus (IFG), and postcentral gyrus showed distinct neural patterns differentiating lying from truth-telling.
• Classifiers trained on instructed deception successfully identified spontaneous deception in the AG and IFG.
• Leave-one-participant-out analysis demonstrated that AG neural patterns could generalize deception detection across individuals.

Conclusions:

• Neural responses underlying instructed and spontaneous deception share commonalities.
• The findings support the feasibility of cross-participant deception validation using neuroimaging.
• This research offers a more robust approach to deception detection, independent of task difficulty and response type.