Argument Formation in the Reasoning Process: Toward a Generic Model of Deception Detection