Verified by 
Remembering is always a constructive process. Do you remember when you were 5 years old and attended the wedding of a family friend? At the reception, you were running around with some other kids, bumped into a table, and spilled punch on the bride's parents. Do you remember doing that? I hope not, because I am relatively sure you didn't knock over the punchbowl at a wedding reception. A few years ago, however, my students and I led many individuals to construct memories of spilling punch at a wedding reception. We asked people to remember this false event in the midst of asking them about other events that actually happened to them. We interviewed them two or three times within one week. We encouraged them to keeping thinking about the events and suggested they might remember more over time. In various experimental conditions between 25 and 40% of the individuals constructed punch spilling memories. These participants did not simply agree that the event might have occurred, but instead constructed full memories. They added details we never suggested and claimed the story as their own.
Memory researchers have led people to construct a wide variety of false memories. Steve Lindsay and his colleagues convinced people that they pulled a mean prank on their second grade teacher. Maryanne Garry and her colleagues led people to remember a hot air balloon ride that they never experienced.
When we did our research, I was surprised at how easy it was to create false memories. I worried, when we first started the research, that we would need to gradually increase the social pressure before people would create memories of spilling the punch. However, the experiment worked the first time we tried and it worked with very little social pressure. The constructed memories lasted as well. A few years after the research, a young woman who had participated told me that she could still see herself spilling the punchbowl on the bride's parents.
This line of work has always caused me to worry about the power of interviews to change and create the past. Lately I've been particularly worried about the use of enhanced interrogation techniques. For example, should the Christmas day pants on fire terrorist be subjected to enhanced interrogation techniques? Many individuals have suggested that national security requires aggressive interviews of potential and actual terrorists. With the best of intentions, these individuals want all the knowledge in that terrorist's memory and are willing to use enhanced interrogation to retrieve it.
Will enhanced interrogation lead individuals to reveal generally accurate information? Probably not. From the public descriptions, enhanced interrogations employ many features that will lead to false memories, including: repetitive questions, multiple interviews, asking about individuals and events based on potentially erroneous information, using heavy social and physical pressures, and isolating individuals. Admittedly neither I nor any cognitive researchers have studied the effects of some of these techniques. I don't think my university's ethics review board would allow me to use water-boarding, isolation, sleep deprivation, or many of the other techniques I've heard about. But these techniques aren't needed - we led people to construct false memories with much less pressure over shorter periods of time. With enhanced interrogation, I suspect I could convince you not only that you spilled the punch and rode in a balloon, but also any other event suggested by Dick Cheney.
This is the problem: Enhanced interrogation techniques will cause people to provide piles of false information they will come to fully believe. The stronger the interrogation, the less reliable the information will be. If the goal is accurate information, then these techniques are the wrong method.
There are valid moral, ethical, and political arguments against the use of enhanced interrogation techniques. My point is that there is a practical argument as well - these techniques will lead to false memories.
