Decoding mental processes with electrophysiological measures: in search of an effective concealed information detection protocol

Amidst the continuous global growth of terrorist activities and violent crimes, the search for an effective lie detection method to assist criminal investigation has been a priority of security agencies and law enforcement departments across countries. The polygraph test, which combines a number of physiological measures of emotional responses, has been used to distinguish liars from truth-tellers by the authorities in the past decades. Its usage was however criticized because of the high proportion of false positive rates when applied to people with high anxiety, and for its low sensitivity for psychopaths or people who are trained to apply countermeasures.

Recently, neuroscience (electroencephalography and functional magnetic resonance imaging) studies have sought for cognitive mechanisms related to deception. The two approaches of such studies are: 1) the cognitive theory of deception, which assumes that lying is cognitively more taxing than truth-telling; and 2) the concealed knowledge test, which assumes that only liars, not truth-tellers, would find guilty knowledge meaningful. However, the neural correlates of these mental processes are relatively easy to manipulate by conscious effort and are subject to the impact of countermeasures.

The present study therefore aims to fill the research gap by proposing an innovative paradigm which could potentially minimize the effects of countermeasures to the established indicators of lying. Our study will try to decode mental processes by examining the early visual processing of information that is emotionally salient and highly familiar, which are typical attributes of crime-related information. We will observe the steady-state visual evoked potentials (SSVEPs) by extracting the frequency spectrum of the EEG recorded during sensory stimulus perception with the Fast Fourier Transform procedure (FFT). SSVEP has the advantage of higher signal- to-noise ratio and is less influenced by artifacts compared to conventional ERP measures, and therefore reliable results can be obtained within a relatively short experimental duration (Morgan et al, 1996). The current study will investigate whether SSVEP can differentiate emotional and familiar faces from neutral faces (Experiment 1-3). Stimulation frequency will also be included as an independent variable in order to find out the optimal frequency for single-subject diagnosis. In Experiment 4, we will examine the impact of mental countermeasure on the SSVEP brain responses, and whether such countermeasures influence the effectiveness of our protocol. Practically, the outcomes of this study may contribute to the development of an effective forensic tool for criminal investigation; theoretically, our findings will reveal the interplay between the bottom-up sensory systems and higher level cognitive processes.