Abstract: Deception detection is an important topic in psychology. The cognitive approach to deception detection is based on the premise that lying is more cognitively demanding than truth telling. Increased cognitive load is hypothesized to result in greater behavioral differences between truth tellers and liars. By manipulating cognitive load through different interfering task of various difficulties during the concealed information test, the influence of cognitive load on memory-response conflict was investigated to better illustrate the cognitive mechanism of deception detection. Second, behavioral and physiological cues for memory-response conflict based deception detection were examined in both noncriminal and criminal group. Finally, machine learning algorithms were employed to predict liars and truth tellers via behavioral and physiological cues. These findings will serve to aid in deception detection in the fields of judicial security and human communication.

Key words: deception detection, cognitive load, memory-response conflict, interfering task, machine learning