The use of coercive interrogation "is based on the assumption that subjects will be motivated to reveal veridical information to end interrogation, and that extreme stress, shock and anxiety do not impact memory," Shane O'Mara, a professor at the Trinity College Institute of Neuroscience in Dublin and the paper's lead author, said in a prepared statement. "However, this model of the impact of extreme stress on memory and the brain is utterly unsupported by scientific evidence."
Prolonged stress and subsequent changes in the body's hormone levels can have a negative effect on memory and learning, and "information presented by the captor to elicit responses during interrogation may inadvertently become part of the suspect's memory," the paper authors note. Similar false memories "recalled" during therapy or eyewitness testimony are well documented.
The report also describes such intense interrogation as a possible case of classical conditioning. If a suspect under interrogation is being repeatedly waterboarded, for example, when they are not talking, the detainee will likely come to associate talking—whether it is giving accurate or incorrect information—with safety and will be inclined to talk more (if not more truthfully). Likewise, "when the captive is talking, the captor's objective has been obtained," the authors note. Thus, the coercive techniques may encourage talking from both sides, but, as the researchers explain in the paper, "torture is as likely to elicit false as well as true information."