Spiking soda.

Does caffeine add flavor--or just addict you?

Why add caffeine to soda? Soft-drink makers say the chemical helps balance out soda's sweet flavor. But scientist Roland Griffiths has another theory: Caffeine has addictive properties that may hook you to buy more soda. Griffiths used the scientific method to try to prove it.

If you're an average U.S. teen, you gulp down at least two sodas every day. Ever wonder why? Is it soda's refreshing taste--or is it the way a caffeinated soft drink can make you feel: energized, alert, even ready to tackle your homework?

According to researcher Roland Griffiths, taste is not the real thing in soda. Soda's true appeal: caffeine, a mood-altering and mildly addictive stimulant drug found in nearly 70 percent of soft drinks sold in the U.S. As a pharmacologist at Johns Hopkins University in Baltimore, Md., Griffiths studies how drugs like caffeine affect mood and behavior.

Beverage makers claim caffeine--a bitter white alkaloid, or chemical base made from plants--enhances soda flavor. Griffiths thinks this claim overlooks caffeine's dramatic and possibly addictive effects on the central nervous system, the body's "information-processing system" that includes the spinal cord and brain. Caffeine may be energizing, but it can also trigger unpleasant side effects: anxiety, irritability, nervousness.

Sweet Suspicion

To debunk a claim, you need more than a hunch. Proving a theory requires an organized, systematic plan. So Griffiths turned to a researcher's most trusted tool: the scientific method, a step-by-step approach to conducting scientific studies--and a surefire way to produce reliable results.

The scientific method takes off when you make an observation, or notice something, that sparks your curiosity. Griffiths learned, for example, that Americans guzzle about 15 billion gallons of caffeinated soda each year. "Caffeine-free versions make up less than 30 percent of total U.S. soda sales," he says. Griffiths wondered why consumers prefer caffeinated sodas: is it the taste? This prompted him to brainstorm a research question: Can consumers really taste the difference between caffeinated and caffeine-free sodas?

In his quest for an answer, Griffiths delved into background research. He pored through science articles about caffeine in soda, then searched for experiments that might already have answered his question. Not satisfied with his research finds, Griffiths set out to conduct his own original research, which he began by formulating a hypothesis, or educated guess about the answer to his inquiry. He predicted that regular drinkers of caffeinated soda would be unable to distinguish between caffeine and caffeine-free soda--unless the soda contained a larger-than-normal amount of caffeine (more than 45 milligrams).

Recipe for Results

Next, Griffiths crafted a well-thought-out plan, or procedure, to put his hypothesis to the test. A good experiment tests the effects of one variable, or characteristic, on another. In this case, Griffiths tested the effect of different amounts of caffeine in soda on a consumer's ability to detect a change in taste. Griffiths' independent variable, the characteristic deliberately changed by the experimenter, is the amount of caffeine added to the soda. The dependent variable, or the characteristic that responds to a change in the independent variable, is the volunteer's taste response. The remaining variables in the experiment--such as the amount of cola--remain unchanged (constant) so they don't influence the results.

Taste-Test Time

Griffiths began his experiment by recruiting 25 volunteers (ages 21 to 55 years) who said they drank a caffeinated cola-flavored soda at least once a week and believed they could taste the difference between a cola soda with caffeine and one without it. Every participant was scheduled for a total of seven tasting sessions. During each session, the participant completed 25 individual taste-test trials.

In a trial, the volunteer tasted and swallowed two soda samples from cups labeled either "A," which contained soda with different amounts (concentrations) of caffeine, or "B," which contained caffeine-flee soda. The caffeine-flee sample served as the control, a standard for comparing the volunteer's ability to detect caffeine. For the first five trials, participants were told which sample they were drinking. The remaining 20 trials were blinded: the volunteers didn't know which sample contained caffeinated soda and which sample contained caffeine-flee soda. After each tasting, Griffiths asked a volunteer to identify the mystery drink: Is it "A" or is it "B"?

To make sure his results proved reliable, Griffiths used many different techniques. For example, between each sample tasting and each new trial, volunteers rinsed out their mouth with water and waited 30 seconds. That way, flavors from the last trial wouldn't interfere with volunteers' perceptions in the new trial.

Add It Up!

Once Griffiths collected his data (experiment information), it was time to make sense of it. That meant adding up numbers, creating charts and graphs (see p. 18), and finally drawing a conclusion, or summary of the results.

Was his hunch correct? His results show that a whopping 92 percent of those tested could not detect the presence of caffeine in soda at concentrations normally found in a can of soda. In this experiment, normal levels of caffeine had no perceptible influence on soda flavor--a conclusion that contradicts claims made by soft-drink makers. "I'd like to see the soda industry come out of denial about the role of caffeine in their products," says Griffiths. "The drug surely accounts for the fact that people drink far more sodas with caffeine than without."

Griffiths published a detailed report about his experiment in a scientific journal. He hopes his results will encourage soft-drink makers to come clean: "Given that sodas are aggressively marketed to kids, manufacturers should openly say why caffeine is added to soft drinks, and clearly label the amount found in each product," he says.

Think you can tell the difference between caffeinated and caffeine-free soda? Follow Griffiths' lead and use the scientific method to design your own taste-test experiment. The rest of this issue will help you get started!

The SCIENTIFIC METHOD Made Easy

1. BE CURIOUS. Base your idea for an experiment on an observation of something interesting.

2. BE CLEAR. State the purpose of your experiment. Usually, researchers do this in the form of a research question: What is the effect of [your independent variable] on [your dependent variable]?

3. READ ABOUT IT. Do background research to find out what has already been discovered about your topic.

4. TAKE A GUESS. State your hypothesis, an educated guess about what the results of your experiment will be.

5. MAP IT OUT. Design a step-by-step plan, or procedure (see p. 16).

6. JUST DO IT. Carry out your experiment and collect data.

7. TAKE NOTES. Record your results. Highlight your results with visuals like graphs, charts, or tables (see p. 18).

8. FIND MEANING. Draw a conclusion, discussing your results. Did your hypothesis prove true?

9. WRITE ABOUT IT. Summarize your findings in a report. Include why and how you conducted your experiment, and a description of your results.

10. SHARE YOUR WORK. Publish your report (see Teacher's Edition).

NUTS AND BOLTS of an EXPERIMENT

Variables

Characteristics or details in an experiment that change or could be changed.

Independent Variable

Characteristic you change or adjust on purpose; also called manipulated variable.

Dependent Variable

Characteristic that responds to a change in the independent variable; also called responding variable.

Hypothesis

Educated guess about how changing the independent variable will affect the dependent variable.

Constants

Details in an experiment that do not change from trial to trial.

Control

Standard for comparison in an experiment.

Trials

Number of times an experiment is repeated: the more trials, the more reliable your results.