The Truth About Lie-Detection – What Works And What Doesn't: - Barking up the wrong tree

The Truth About Lie-Detection – What Works And What Doesn't: - Barking up the wrong tree

German scholars have pioneered Criteria Based Content Analysis (CBCA). Armed with a 19-point list of identifiers, analysts ask if the story was incoherent or disorganized. They count the number of details, how frequently the storyteller self-corrected wrong facts, or admitted not knowing something about his own story.

According to CBCA proponents, liars tell stories in chronological order to keep the facts straight. They rarely correct a misstatement, and they're less willing to say, “I don’t know.” Some scholars using CBCA can accurately predict lying as high as 78 percent of the time. But that's nowhere near perfect, and it’s not a method easily used in real-time conversation.

Another intriguing lie-detection test is Reality Monitoring. The idea behind Reality Monitoring is that a truth-teller will, without prompting, relay spatial and sensory details. They won't just say where the man stood in the room: they'll include if the man was near or far from the window, how the room smelled, the sudden bang of door slamming. Liars are creating a story intended to make sense, so they rely on logic to supply the details. For example, a truth-teller might say, “I remember he had an umbrella, because it was dripping on the floor,” while the liar would say, “Well, he must have had an umbrella with him because it rained earlier.” The liar’s story is based in a rational inference, compared to the truth-teller’s sensation. Reality Monitoring, like CBCA, has shown some surprising success.