In their battle against the bulge, desperate dieters have tried drugs, surgery, exercise, counseling, creams and even electrical fat-burning belts.
Now some psychologists have a new idea: subtle brainwashing.
A team led by psychologist Elizabeth F. Loftus of UC Irvine found that it could persuade people to avoid fattening foods by implanting unpleasant childhood memories about them -- even though the memories were untrue.
In a paper published today in the Proceedings of the National Academy of Sciences, the team said it successfully turned people off strawberry ice cream and, in earlier studies, it had done the same with pickles and hard-boiled eggs -- in each case by manipulating the subjects to believe that the foods made them sick when they were children.
The scientists say they have also successfully implanted positive opinions about asparagus by convincing subjects that they once loved the vegetable.
The method, if perfected, could induce people to eat less of what they shouldn't and more of what they should, Loftus said. Good memories about fruits and vegetables could be implanted, as well as bad ones about low-nutrient, high-calorie foods.
In the strawberry ice cream experiment, Loftus and her team asked 131 students to fill out forms about their food experiences and preferences, including questions about strawberry ice cream. The subjects were then given a computer analysis of their responses that was supposed to indicate their true likes and dislikes.
However, 47 students were inaccurately told that the analysis made it clear that they had gotten sick from eating strawberry ice cream as children. Of these, almost 20% later agreed on a questionnaire that they had, in fact, been sickened by the treat and that they intended to avoid it in the future.
The findings were stronger in a second experiment where, in addition to the other steps, students were asked to detail the imaginary ice cream episode. In that case, 41% of those given erroneous information later believed the tale and said they intended to avoid it.
Several weight-control experts expressed interest in the study, but also were skeptical about using implanted memories as a dieting technique.
Michael Strober, director of the eating disorders program at UCLA's Neuropsychiatric Institute, said pressures causing people to gain weight were myriad -- including rushed lives, high-calorie convenience foods and physical inactivity.
"Such systemic lifestyle issues need to be targeted by something far more comprehensive than implanting false memories," he said.
Deliberately implanting memories also "raises profound ethical questions," said Stephen Behnke, ethics director of the American Psychological Assn.
"Say, for example, we could change a person's belief about their entire childhood," he said. "Would doing so be ethical?"
The food studies are the latest in a string of memory experiments by Loftus, a distinguished professor of psychology, social behavior and criminology.
She has shown that memories are not as reliable as once supposed and can be created from whole cloth -- persuading people, for instance, that as children they had stuck their hands through panes of glass or were lost for hours at a mall.
Loftus is most famous for her position on recovered memories of childhood sexual abuse. Based on her work, she has suggested that most of these memories were probably false.
Loftus said the food experiments were the first in which she explored a positive practical application of this manipulation of memory.
Several problems remain to be solved before a memory-manipulation diet plan is possible, she said. For instance, it is not yet known how long the false beliefs last.
Perhaps most importantly, the scientists have so far failed to implant false beliefs about two common food items, chocolate chip cookies and potato chips.
"It suggests that the ability to use this research to ... influence eating behavior may be very limited," said Dr. Michael A. Bush, founder and former co-director of Cedars-Sinai Medical Center's weight management program.
Loftus thinks that this problem might be resolved if attempts to implant memories were ramped up further using additional feedback and drills.
Parents might use this strategy to clean up their children's eating habits -- get them to eat more broccoli, perhaps, or fewer French fries, she said.
Gregory Stock, director of UCLA's program on medicine, technology and society, said this might not be the best thing for healthy parent-child relationships.
"How would the child feel later on once you told them how you had manipulated them into eating their vegetables?" he said. "If you're going that far, why not use Photoshop to doctor their childhood photos to show them having problematic experiences with junk food?"
For that matter, Stock added, "it would be much simpler to give kids the offending food -- a McDonald's burger, a pizza, whatever -- and put a little something on it that makes them harmlessly sick ... then you would really affect their eating."
Acknowledging that the issue was ethically tricky, Loftus said she and her colleagues debated including this suggestion in their manuscript.
"We finally decided that it's certainly worth some discussion. We're not taking a position on whether people ought to go out and do that or not," she said.
"People kind of cringe at the idea that anyone would suggest that they lie to their children, but they do it all the time when they tell them Santa Claus exists and so does the Tooth Fairy."