Voice Stress Analysis is a developing technology that can aid in the detection of deception. The underlying principle of VSA is that lying causes most people to experience involuntary psychological and neurological changes that occur at the subconscious level. Liars experience increased psychological stress and hightened sympathetic arousal (for example, fear or guilt) that manifest as physiological anomaly. These anomaly can be detected, measured and analyzed to predict innocence or guilt. In the case of Voice Stress Analysis, the particular physiological anomaly of interest relate to the modulation of the human voice.
The underlying theory of VSA
When we speak, our lungs and chest cavity contract and force air through the glottis and between the two vocal folds (commonly known as one’s vocal chords or “voicebox”). Audible sound is created by the vibration of the vocal chords as air passes through them. Apart from the sound we hear, a number of natural infrasonic anomaly occur, for example formant frequencies generated by resonance in the nasal, sinus and throat cavities, and both amplitude modulation and frequency modulation at a subsonic level. These inaudible modulations are the focus of Voice Stress Analysis or more accurately, the changes in modulation at low frequency are what is measured to determine deception. This change in modulation is not discernible to the human ear, but can be detected and measured with specialized equipment and training. The VSA apparatus is able to capture, identify and correlate changes in physiological states (for example, micro-tremors and the dampening of selected frequency variations) with questions and responses. An experienced analyst can interpret the information in order to differentiate between unstressed (truthful) and stressed (deceptive) vocalization.
The VSA apparatus
In order to measure and interpret these vocal changes a multidirectional microphone is used to capture sounds in the inaudible 8 to 12Hz frequency range. The audio is digitized and processed by a complex algorithm to produce a graphical chart that allows a sufficiently well-trained and experienced examiner interpret the results. Voice stress analysis can also be conducted using prerecorded audio or at a distance via telephone.
A critical component of voice stress analysis centers around the use of appropriate questioning techniques, such as the Control Question Test (CQT), Guilty Knowledge Test (GKT) or Guilty Actions Test (GAT).