An experimental test of flashbulb memory

The Stroop effect is a well-documented phenomenon that shows how easily we can be distracted from a simple task. In the classic Stroop experiment, we are shown a word, such as GREEN, and asked to indicate the color it is printed in. When the meaning of the word itself conflicts with the word's color, the task is more difficult. There's a good demo of the effect here. Go ahead and try it—it's easy and it only takes about 30 seconds. If you're like most people, you will take longer to respond to the words that are printed in an incongruent color.

The Stroop task can also be used to measure other things that might distract us, such as the orientation of a word (vertical, horizontal, or diagonal), or even emotion-laden words.

Donald MacKay and Marat Ahmetzanov of UCLA have figured out a way to use the Stroop task to test flashbulb memories experimentally ("Emotion, Memory, and Attention in the Taboo Stroop Paradigm: An Experimental Analog of Flashbulb Memories," Psychological Science, 2005). It's difficult, if not impossible, to thoroughly control a traditional flashbulb memory experiment. Even in the case of Talarico and Rubin's impressive September 11 study, the comparison of flashbulb memory to ordinary memory was colored by the events of September 11—after all, the initial questionnaire was administered on September 12! (If you haven't read our post on Talarico and Rubin's study, you might want to do so now. Their basic finding is that we believe memory for shocking events to be more accurate than memory for normal events, but in fact it is not.)

MacKay and Ahmetzanov believed that the Stroop task could provide an analog to an emotionally charged flashbulb memory. They asked volunteers to participate in a test similar to the classic Stroop task: a computer flashed colored words on the screen, and rather than reading the words, participants had to indicate the color the word was displayed in. But instead of using color words, the researchers substituted "taboo" words: obscene words or epithets like shit, cunt, or nigger that often arouse a strong emotional reaction in readers. Each taboo word was paired with a neutral word (an animal name) the same length and number of syllables; each word was presented in one of six possible locations on the screen.

The Stroop test was primarily a distraction; the real test came afterwards, when participants were surprised with a test asking them which location particular words were presented in. Six of the taboo words and six of the neutral words had consistently been presented in the same location; the other twelve words appeared in random locations. Participants remembered the location of the taboo words more than twice as often as the neutral words. When asked to write down words that had appeared in the task, they identified almost twice as many taboo words as neutral words.

In a second experiment, MacKay and Ahmetzanov gave a similar test, but every word was displayed in a random color and location. But this time, in two of the six possible locations, whatever word that was displayed there was always displayed in the same color, and was the same word type. So, for example, words displayed in the upper left-hand corner of the screen would always be a green taboo word, and words displayed in the lower right-hand corner of the screen would always be a red neutral word, but words displayed in any other location would be randomly chosen from the other four colors and word types: blue, gray, orange, and pink, with no consistent pattern.

Again, after the task, there was a surprise test on where each word had appeared. This time, there was no difference in recognition of location of the taboo versus neutral words. However, participants did recall many more of the taboo words than neutral words when simply asked to write down the words that had appeared in the task.

So what does all of this say about flashbulb memory? Well, if the way flashbulb memory works is that we store a mental image (like a "fuzzy photograph") of an emotionally charged event, then we should have a better memory of the location of the taboo words in both experiments. Instead, participants only had a better memory of location when words consistently appeared in the same location. So this suggests that these emotionally charged memories are not like photographs.

Why did viewers remember the locations in the first experiment? In that case, the same word often appeared in the same location, so there was a contextual reason for viewers to remember the location of the word. In other words, flashbulb memories appear to be stronger than normal memories when the details of the memory are relevant to the general sense of the memory.

So you might not remember what you were wearing on September 11, but you may remember what the weather was like in New York that day, since you repeatedly saw images of the twin towers with a clear blue sky in the background.