Forgetting and false memory

Transcript

RyanIn 1984, Jennifer Thompson, a twenty-two-year-old college student in North Carolina, was assaulted at knifepoint. During the attack, she tried to memorize every detail of her attacker's face, determined to help convict him.

SoheeShown six photos, she narrowed it to two, and said of one, I think this is the guy. In a lineup, she hesitated between two men and finally said Ronald Cotton looks most like him. By the trial she had absolutely no doubt. Cotton was convicted.

RyanAnd after he'd served eleven years in prison, DNA evidence showed he was innocent.

SoheeWelcome to EPPP Prep, the podcast. I'm Sohee.

RyanAnd I'm Ryan. Last time we covered how memory is built. Today, how it fails: forgetting, distortion, and why an honest, confident witness can be wrong.

SoheeLet's start with forgetting, the loss of information from long-term memory. The oldest data come from Hermann Ebbinghaus in 1885. He memorized lists of nonsense syllables, then measured how much he retained when he relearned them, at intervals from twenty minutes to thirty days.

RyanThe result is his famous forgetting curve. Memory for new information decays quickly at first, and then levels out.

SoheeSometimes, though, nothing was lost, because nothing was stored. That's encoding failure. Most Americans can't pick out an accurate drawing of a penny. We encode just enough to tell it from other coins.

RyanIn short-term memory, two forces affect retention: trace decay and interference. Peterson and Peterson, in 1959, had people recall three-letter trigrams after delays of three to eighteen seconds. Recall fell from about eighty percent at three seconds to ten percent at eighteen, and they concluded short-term memory decayed.

SoheeBut Keppel and Underwood went back and looked only at the first trials. They found proactive interference was also affecting retention. Earlier trigrams were getting in the way of later ones.

RyanWhich brings us to the two kinds of interference. Proactive interference: old information hinders recall of newly learned information. You get a new phone number, and you keep giving out the old one.

SoheeRetroactive interference: newer learning hinders recall of older information. You study Ebbinghaus this week, Erikson next week, and now Ebbinghaus is hard to find.

RyanA memory hook. Name the direction the interference travels. Pro, forward: the old moves forward and blocks the new. Retro, backward: the new reaches back and blocks the old.

SoheePause and think. After moving, a man keeps writing his old address on forms. Proactive or retroactive?

RyanProactive. The old address, learned first, is interfering with the new one.

SoheeNow a big framework. Daniel Schacter's seven sins of memory, sorted into three groups: forgetting, distortion and intrusion.

RyanThe forgetting sins. Transience: accessibility of a memory fades over time. Absentmindedness: lapses caused by breaks in attention, like forgetting where your phone is. And blocking: information is temporarily inaccessible, the tip-of-the-tongue feeling.

SoheeThe distortion sins. Misattribution: confusing the source of a memory. Suggestibility: false memories that come from someone else, like leading questions. And bias: memories distorted by current beliefs.

RyanAnd one intrusion sin, persistence: the inability to forget unwanted memories, like a traumatic event that keeps coming back.

SoheeMisattribution has a striking example. An Australian eyewitness expert, Donald Thomson, appeared on live television talking about eyewitness memory. He was later identified by a woman as the man who raped her. The attack happened while she was watching that very program. She'd confused the attacker's face with the face on TV.

RyanAnd the difference between misattribution and suggestibility is an important one. With misattribution, you create the false memory yourself. With suggestibility, it comes from someone else, like a therapist or a police interviewer asking leading questions.

SoheeBias has subtypes, too. Stereotypical bias involves racial and gender biases. Egocentric bias enhances our memories of our own past. And hindsight bias, the I knew it all along feeling, which the reconstructive nature of memory feeds.

RyanThat word, reconstructive, is the key. Memory isn't like a video recording. Different parts of a memory are stored in different places and put back together each time we remember. Each component is a chance for an error.

SoheeWhich is where Elizabeth Loftus comes in. She developed the misinformation effect paradigm: after exposure to additional, possibly inaccurate information, a person may misremember the original event.

RyanThe classic study, Loftus and Palmer in 1974. Forty-five college students watched films of car accidents and were asked how fast the cars were going when they smashed, collided, bumped, hit or contacted each other.

SoheeThose who heard smashed estimated much higher speeds than those who heard contacted. And a week later, the smashed group was more than twice as likely to say they'd seen broken glass. There was no broken glass in the film.

RyanAnd suggestion can spread through a whole community. During the 2002 sniper attacks around Washington, D C, a white van was spotted at one shooting, and the police chief showed a picture of it on national television. Afterward, several other eyewitnesses called to say they too had seen a white van.

SoheeWhen the suspects were finally caught, they were driving a blue sedan. As the textbook puts it, it's the suggestion that causes the false memory.

RyanNow back to Jennifer Thompson. Gary Wells and Deah Quinlivan point to suggestive identification procedures, like confirming a witness's choice with good choice. After her photo pick, the detective told her she did great.

SoheeAnd practice has changed. Ever since Loftus published her first studies on eyewitness suggestibility in the 1970s, professionals have been aware of flaws in how witnesses are questioned. Interviewers who use neutral, less leading language get more accurate recall from children. A blind photo lineup is recommended, where the person running it doesn't know which photo is the suspect. And judges in some states now inform jurors about the possibility of misidentification.

RyanNow flashbulb memories, an exceptionally clear recollection of an important event. Many people can recall exactly where they were when they heard about a historic event. Brown and Kulik's original definition didn't require a widely shared event, and positive events can create them too.

SoheeThe textbook ties them to arousal theory: strong emotions trigger strong memories. Under stress, the brain secretes more of the neurotransmitter glutamate, which helps us remember the stressful event.

RyanBut vivid isn't the same as accurate. Even flashbulb memories can lose accuracy over time. The textbook's example: at least three times, President George W. Bush gave an inaccurate account of how he learned of the September eleventh attacks.

SoheeThen there's the controversy over repressed and recovered memories. Some argue traumatic childhood memories can be locked away for decades and later recovered intact. Loftus has challenged the idea that such memories can be recovered years later through techniques like hypnosis, guided visualization and age regression.

RyanAnd to be clear about her position: Loftus isn't saying childhood sexual abuse doesn't happen. She questions whether recovered memories are accurate, and she's skeptical of questioning that could create misinformation effects.

SoheePause and think. A therapist asks a client, over several sessions, whether the man at the party might have pushed her down the stairs. Months later, the client vividly remembers being pushed, though witnesses saw her trip. Which of Schacter's sins?

RyanSuggestibility. The false memory came from someone else's leading questions. If she had mixed up the source on her own, say, remembering a movie scene as her own experience, that would be misattribution.

SoheeSo how might this appear on the exam? Expect pairs. Proactive versus retroactive. Misattribution versus suggestibility. Transience versus blocking. And the misinformation effect, where the tempting wrong answer says confident memories must be accurate.

RyanNow, the part of this chapter written for students. The textbook gives evidence-based study advice, and it fits exam prep beautifully. Use distributed practice: study across time in short sessions instead of cramming. Memory consolidation takes time.

SoheeUse elaborative rehearsal and the self-reference effect. Put definitions in your own words and connect them to your own life. Take practice quizzes, and focus your review on the items you got wrong.

RyanReduce interference by studying in a quiet time without distractions. Regular aerobic exercise is good for memory. And get enough sleep, because during sleep the brain organizes and consolidates information for long-term memory.

SoheeLet's recap. Forgetting curves drop fast, then level off. Encoding failure means it was never stored. Decay and interference both affect short-term memory.

RyanProactive: old blocks new. Retroactive: new blocks old.

SoheeSchacter's seven sins. Forgetting: transience, absentmindedness, blocking. Distortion: misattribution, suggestibility, bias. Intrusion: persistence.

RyanMemory is reconstructive. Loftus showed leading questions can change memories. Flashbulb memories are vivid, not necessarily accurate. And careful procedures, like blind lineups and neutral questions, protect witnesses.

SoheeIf you've ever blanked on something you knew, maybe even on test day, please hear this: that's blocking, a normal memory sin, not a measure of you. Spaced review and good sleep are on your side.

RyanTry a handful of practice questions today, and a few more tomorrow. That's distributed practice in action. See you next time on EPPP Prep.

Loading the read-along timings…