Detecting concealed autobiographical information using feedback and recognition P300 ERPs. Exploring the impacts of changed feedback proportions and social pressure in the fCIT

The feedback concealed information test (fCIT) incorporates feedback on memory concealment performance within the classical CIT. Providing feedback facilitates the analysis of feedback related ERPs in addition to initial recognition related ERPs within each trial. Combining multiple ERPs aims to improve the reliability and efficiency of the test. The present study investigated the efficiency of combining recognition and feedback related P300 ERPs in detecting concealed birthdates in a personal item fCIT. Feedback proportions were altered to improve believability in feedback. Detection efficiencies using combined P300 amplitudes within this altered paradigm and the impacts of social pressures on feedback P300s were explored. Results indicated that with altered feedback proportions, both P300s could be used to categorise innocent and guilty participants with AUCs = 0.79-0.91. Across both measures, trials in which information had been concealed evoked the largest P300 potentials. Social pressure introduced via witness observation during the test impacted feedback P300 amplitudes differently for different stimulus types and warrant further investigation into social motivational factors in the fCIT. The study has theoretical implications, with feedback P300s being manipulated by increased social pressure during the test, independently of recognition P300s, providing support for theories that they may reflect separate mechanisms.

Background & Context

The Concealed Information Test (CIT) has emerged as a scientifically robust alternative to traditional polygraph methods, relying on involuntary physiological responses to recognized information rather than deception per se. The feedback concealed information test (fCIT) represents an innovative enhancement that incorporates feedback on memory concealment performance within the classical CIT, facilitating analysis of both recognition-related and feedback-related event-related potentials (ERPs) within each trial.

Combining multiple ERPs aims to improve the reliability and efficiency of the test, addressing longstanding concerns about single-measure approaches in psychophysiological detection. This 2025 undergraduate research project from the University of Plymouth advances fCIT methodology by examining two critical practical considerations: the credibility of feedback proportions and the influence of social pressure through witness observation during testing.

The P300 component—a positive ERP deflection occurring 300-800ms after stimulus presentation—has demonstrated effect sizes of d = 1.89 in CIT applications, outperforming traditional autonomic measures. The fCIT leverages both recognition P300s (elicited when participants encounter their concealed information) and feedback P300s (evoked when participants receive performance feedback), creating a dual-indicator system for detection.

Research Design & Methodology

The study investigated the efficiency of combining recognition and feedback related P300 ERPs in detecting concealed birthdates in a personal item fCIT, with feedback proportions altered to improve believability in feedback. Using autobiographical information (participants' own birthdates) represents ecologically valid stimuli with strong personal salience, enhancing the strength of recognition responses.

Key methodological features included:

• Modified feedback paradigm: Altered feedback proportions from standard protocols to enhance perceived authenticity
• Dual P300 measurement: Extraction of both initial recognition P300s and subsequent feedback-related P300s from EEG data
• Social pressure manipulation: Introduction of witness observation during testing to explore social motivational factors
• Guilty/innocent paradigm: Participants assigned to conceal their birthdate (guilty) or tested without concealment (innocent)

The personal item CIT paradigm using birthdates provides high motivation for concealment while avoiding ethical concerns associated with mock-crime scenarios. EEG recording captured event-related brain potentials time-locked to both stimulus presentation and feedback delivery, allowing independent analysis of recognition and feedback processing mechanisms.

Results & Key Findings

Results indicated that with altered feedback proportions, both P300s could be used to categorise innocent and guilty participants with AUCs = 0.79-0.91. This represents strong diagnostic accuracy, with performance ranging from acceptable to excellent across the different ERP components.

Critical findings include:

• Stimulus discrimination: Across both measures, trials in which information had been concealed evoked the largest P300 potentials
• Detection efficiency range: Area under the curve values of 0.79 to 0.91 indicate 79-91% accuracy in classifying participants
• Independent mechanisms: Recognition and feedback P300s responded differently to experimental manipulations, supporting distinct processing pathways
• Social pressure effects: Social pressure introduced via witness observation during the test impacted feedback P300 amplitudes differently for different stimulus types

The differential impact of social pressure on feedback P300s versus recognition P300s provides compelling evidence that these components reflect separate cognitive and motivational processes. Recognition P300s remained relatively stable under observation, while feedback P300s showed stimulus-specific modulation, suggesting the feedback system is more sensitive to social-evaluative concerns.

Discussion & Significance

The study has theoretical implications, with feedback P300s being manipulated by increased social pressure during the test, independently of recognition P300s, providing support for theories that they may reflect separate mechanisms. This dissociation challenges unitary models of concealed information processing and suggests feedback monitoring engages distinct neural systems related to outcome evaluation and social cognition.

The improved believability of feedback through altered proportions addresses a critical weakness in earlier fCIT research. If participants doubt the authenticity of feedback, the motivational significance of performance information diminishes, potentially reducing the amplitude of feedback-related ERPs. The strong classification accuracies (AUCs 0.79-0.91) achieved with the modified paradigm suggest this methodological refinement successfully enhanced ecological validity.

Compared to traditional CIT accuracy rates of 80-90% and standard P300-CIT performance, the fCIT's dual-indicator approach provides comparable or superior detection efficiency. The independence of recognition and feedback P300s creates redundancy in the detection system—if one measure is compromised (by countermeasures, encoding failure, or individual differences), the other may still provide diagnostic information.

Limitations & Considerations

As an undergraduate research project using a convenience sample of university students, generalizability to forensic populations requires caution. Real-world suspects differ from laboratory participants in motivation, stress levels, cognitive strategies, and stakes of detection. The specific sample size is not reported in available sources, limiting assessment of statistical power.

Additional methodological considerations include:

• Birthdate concealment, while ecologically valid, differs from crime-relevant information in emotional valence and memory characteristics
• The specific alterations made to feedback proportions are not detailed in accessible sources, preventing replication
• Social pressure via witness observation represents only one form of motivational manipulation; other contextual factors (consequences, interrogation pressure) remain unexplored
• Individual differences in anxiety, self-consciousness, and performance monitoring may moderate social pressure effects on feedback P300s

Social pressure effects warrant further investigation into social motivational factors in the fCIT, as the interaction between stimulus type and observation conditions requires systematic parametric studies to identify boundary conditions and optimal testing protocols.

Practical Applications

The fCIT's enhanced detection accuracy and theoretical robustness position it as a promising evolution in psychophysiological credibility assessment. For polygraph examiners, the dual-P300 approach offers increased confidence in classifications by providing convergent neurophysiological evidence across independent processing streams. The resistance of recognition P300s to social pressure effects suggests core detection validity remains intact even under high-stakes conditions.

Practical implementation considerations include the need for specialized EEG equipment, trained technicians for electrode application, and sophisticated signal processing capabilities. The altered feedback paradigm requires careful calibration to maintain believability while preserving diagnostic validity. For consumers and policy-makers, this research demonstrates continued scientific refinement of CIT methodology, with neurophysiological measures offering advantages in objectivity and resistance to voluntary control compared to autonomic measures. The separation of recognition and feedback processes provides forensic safeguards against both false positives and false negatives.