Anton races home at speeds well in excess of the speed limit. He's rushing to beat his parents home so that he can hide their anniversary present so it will be a surprise. Suddenly, he hits a slick patch and runs his car off the road an into a tree. He's okay, but the car is totaled and his parent's surprise anniversary party is ruined.
How much is Anton to blame for the accident? If you had to rate it on a scale of 1 to 10, maybe you'd give him a 7. After all, he was just trying to do something special for his parents.
But what if instead of hiding an anniversary present, Anton was rushing home to hide his cocaine stash? Would you now say he's more to blame for the accident? You might not when the two alternatives are placed side-by-side, but when Mark Alicke told the two versions of this story to different groups, the cocaine group rated Anton as more blameworthy than the anniversary present group.
Alicke's study provided the foundation for an array of studies on the effects of social evaluations of individuals on apparently unrelated events, and even factual recollections about episodes.
But when a team led by David Pizarro addressed this question, no study had yet shown that unrelated details about a person could literally affect witnesses' accuracy in recalling that person's actions. The researchers presented a simple story to 283 college students. The story described a person named Frank entering a restaurant, paying with cash for a drink, then ordering a three-course meal, receiving a cell phone call, and leaving without paying the $56.43 bill.
Then some of the students learned that Frank was a generally responsible person, it's just that the phone call was the hospital, telling him his daughter had been in an accident. When Frank realized his mistake, he contacted the restaurant and told them he'd return with payment the next day.
Another group of students were told the Frank enjoyed walking out on checks, and had treated the waiter at the restaurant rudely. He was obnoxiously loud on his phone call and deliberately left without paying.
A third group received no explanation for Frank's behavior.
All the students were asked to rate Frank's level of blame for his behavior on a scale of 1 to 9. As you might expect, those who heard the "bad" explanation rated him very highly -- an average of 8.14, while the "good" group gave a significantly lower rating of just 2.96.
The students returned to the lab a week later and were surprised with a quiz about the story. They were asked to recall the price he paid for the meal as well as several menu items. Here are some of the results:
The "bad" group recalled the meal cost as significantly higher than the "good" group -- those overestimating the magnitude of Frank's crime. There was also a trend towards those in the "bad" group being more likely to believe that Frank hadn't paid for his drink (he had).
Pizarro et al. say this is first study showing that social knowledge of an individual can actually distort details about an event. There have been some studies focusing on different descriptions of an event (a car crash seen on video and later described as two cars "smashing into" each other versus "hitting" each other), which resulted in memory distortions for the original event, but never has a study found that changing details about an individual's character can also distort memories of an event.
Pizarro, D.A., Laney, C., Morris, E.K., Loftus, E.F. (2006). Ripple effects in memory: Judgments of moral blame can distort memory for events. Memory & Cognition, 34(3), 550-555.
- Log in to post comments
This is really fascinating. I'd love to know if there's ever been a study in which groups of people were given a story like the ones above, but the only difference in the stories the groups heard were the genders of the characters or their names (using names associated with certain races or nationalities). Or they could vary the ages (teen in one version, grandma in the other). The possibilities are endless.
This could have implications on witness testimonies, right?
I never thought in this way. This is really thought provoking.
I'd say not morality, but social cognition is key here.. If Frank is this obnoxious guy, he probably went for the expensive menu items, knowing he's not going to pay anyway. So, since we are not that good at remembering exact information, the reconstruction of that information would get biased by what would be likely to have happened, given our (better) memory for character and motivation.
The second part of this is really interesting, but I'm confused by the first part. Did the researchers offer any explanation to the subjects of what "blame" means? That is, they're asking "rate Frank's level of blame from 1-10". Isn't it reasonable to think that the "bad" Frank has more blame than "good" Frank? It is even stated that he deliberately left without paying, which implies a conscious act on his part to break the rules. Good Frank's actions seem more accidental, and are partially due to circumstances beyond his control. I think the ratings are entirely appropriate.
Intentions are everything.
Chris,
Yes, "bad" Frank is clearly more blameworthy. This was just to verify that readers actually believed he was indeed blameworthy.
Wow! This just goes to show how fluid morality is. No wonder we have dificulty deciding what legislation should be. One person could be thinking of a new law (or punishment for breaking a law) with one stereotype in mind, another person from a much different perspective. Even something as simple as the legality of marijuana: people mainly think of users as dirty defiant youth, even though many educated people (Carl Sagan, anyone?) used marijuana. Perhaps one day we can use this kind of study to show the hazards of legislating morality.
Seems like this would have implications for celebrity and high-profile trials, too... at least what we "know" of the person's character (through the media) would also affect how guilty we really think they are.
Then again, finding out that a famous person is on trial would also count as part of our social knowledge of the person.
I don't think it's fair to say that the details given to the two groups in Pizarro's study were unrelated to the 'crime.' Particularly the detail about calling the restaurant to apologize and arrange to remit payment in combination with having received a very distracting phone call. These details make it seem as though the 'crime' was in fact completely unintentional, and would in fact not even be considered a crime by most people's standards. If you're presenting details that make the same action into an actual crime vs. a non-crime, I think that distinction is important. I'd be interested to see the results of a study that provided character-related details that did not speak to motive OR reparatory actions taken after the fact.
i agree with Shannon.
"Pizarro et al. say this is first study showing that social knowledge of an individual can actually distort details about an event" - that is not acurate, because subject were not given "social knowledge of an individual", but more facts related to the event.
Most of the commenters are completely missing the point of the study. The issue isn't whether they thought that Frank was good or bad. Of course it's a "crime vs. a non-crime", and of course they blame him more if it's intentional. This is all a non-issue.
The interesting part is that if they thought Frank was a good guy, they remembered the bill as being $55, and if they thought he's a bad guy, they remember the bill was $65. So people's memory of facts can be distorted by their opinions of the other people involved.
This could have a lot of implications for, like Freiddie (comment #2) said, witness testimonies -- if the witness thinks that the person is guilty, they might subconsciously exaggerate the facts (like the guy was going 90 mph, when he was really going 60).
Very interesting addition to the minsinformation effect literature.
The study seems, essentially, to be a cognitive perspective of pretrial publicity effects examined in a study by Kramer, Kerr and Carroll (1990), which may have been influential in establishing the policy that jurors avoid media exposure to cases for which they are involved.
In that study, participants in 108 mock juries watched a re-enactment of an armed robbery trial. Before seeing the tape, participants were exposed to newspaper clippings of the case that contained either neutral or incriminating information (e.g., that the defendent was involved in an accident in which a child was killed). Despite being told to make an innocent or guilty verdict based on evidence concerning the current case, 33% voted guilty if they were exposed to neutral material; 48% voted guilty if they were exposed to incriminating material not directly related to the current case.
It's conceivable that a similar phenomenon would take place if jurors were involved in a case concerning Frank, and were exposed to newspaper information about Frank that described him neutrally, or as a good or bad person.
Dave, i see one problem that can explain the whole mess. try this: before you evaluate the results, decide what you mean by blame. then, decide what each of the participants mean by that word. of course you can't really. it's one of the vaguest words in our vocabulary. we can however, guess that many will consider how bad the person's behaviour was- that's what they are thinking when they read the word "blame". then, when they learn of bad behaviour, they're just confirming that they read the explanation. in other words, you just end up asking: "is bad behaviour bad, or good?" we should perhaps offer little surprise at the result.
S.