Cognitive Nutrition Update:

British Crown
Debates Nutrient / IQ Link

An interview with Dr. Stephen Schoenthaler
by John Morgenthaler and Steven Wm. Fowkes

Do vitamin supplements increase intelligence in children? Why is this question being debated in a British court of law? Do vitamin supplements decrease behavior problems and acts of violence in juvenile correctional facilities? Dr. Stephen Schoenthaler, Professor of Sociology and Criminology at California State University Stanislaus, answers these questions, and more.

Ceri: How did you get into this field?

Schoenthaler: I sort of stumbled into it in 1980 when one of my students told me that sugar caused crime. I’m a criminologist and I had never heard of such a thing, so I looked it up in my criminology textbooks. I discovered that it was an idea that was out there and growing, but that no one had taken it seriously enough to conduct a study or trial. It seemed to me to be a good idea, either way it would be important: if it wasn’t true I could kill a bad idea, and if it was true I could lower crime rates by changing sugar consumption.

Ceri: And you could make a lifelong enemy of the sugar industry?

Schoenthaler: (laughs) Maybe so. I was a keynote speaker at their ’83 convention in New Orleans. People tried to warn me that they were going to get me—bump me off—but I said, “No, they want to know what is going on in this field,” so off I went into the lion’s den to be open minded and honest, expecting them to be open minded and honest. I figured they might try to convince me that the results were something else, or that maybe I had missed something; what I found was diet drinks, and artificial sweetener for the coffee.

Ceri: How did you first get into examining intelligence?

Schoenthaler: Back in ’88, we did a double-blind study of children who were locked up in juvenile correctional facilities in Oklahoma. We wanted to know what the effect of improved nutrition was on the children’s behavior. We had already examined this question with foods in about 830 institutions in the US and each time we got better behavior and/or better academic performance. But a good scientist would say that the food itself was responsible for the improvement in behavior. And there is the psychological effect that is associated with getting more attention and better food.

Ceri: Or more attention from the observing scientific team.

Schoenthaler: That’s also a legitimate possible influence. The only conclusive thing we could say was that when children were fed better, they did better. But we didn’t know exactly why. The only way we could properly control for biological versus psychological influences was to do the equivalent of a drug trial. This calls for a placebo group as a control and an experimental group which is given the vitamins and minerals.

Ceri: And this was the study in Oklahoma?

Schoenthaler: Yes, we did the study in 1987 and the findings were released on British TV in January of 1988. At the same time, a colleague was doing a similar study on English school children; same time frame and almost the same study. Both studies had a control group getting a placebo and an experimental group receiving the vitamins, with random assignment to each group — the ideal and perfect design. When we analyzed our data, our supplemented children gained six IQ points compared to the placebo group.

Ceri: What about the behavioral benefits in regard to violence and aggression?

Schoenthaler: That was the main thrust of the study, and we basically cut the violence, the antisocial behavior, hyperactivity, and all types of rule violations in half. While the placebo group did show some improvement, the massive improvement was seen only in the supplement group. We examined multiple measures of behavior, for example, how often children attacked each other, how often they attacked staff members, how often they got in trouble at the correctional-facility school, and the rate of injuries from the attacks—every way we could validate serious violence. The improvement in behavior was coming from the supplement group, and the entire institution was running much smoother as a result — all of this from an improvement in nutrition.

Ceri: Did you find comparable improvement in all the different measurements of violence, or were some of them pretty small and others dramatic?

Schoenthaler: They were consistently cut across all the categories. We kicked the daylights out of things like hyperkinesis, hyperactivity, insubordination, fighting, truancy, and assault and battery — a diverse collection of anti-social behaviors. It was really quite remarkable.

Ceri: It sounds like the cost of running that institution would be dramatically lowered?

Schoenthaler: Yes. It was interesting. The staff decided for itself that the improvement was too important to wait for the state government to decide the long-term policy implications of the results. The improvement took place virtually overnight—within 48 hours; it chopped the rate of violence in half. Rather than go back to the way it was before the study started, the institutional staff pooled their money and bought supplements for the kids instead of waiting for the state to eventually take action.

Ceri: Do you know anyone who has done an informal economic analysis of the savings that might be gained.

Schoenthaler: It’s been done several times. The savings from the use of supplements can be astronomical. The cost of giving these supplements to a incarcerated delinquent for a year, depending on the caliber of what one uses and the wholesale cost, varies from a low of $15 to a high of $30-$40 per person. The cost to keep them locked up is about $36,000 per child per year. The violence is a significant fraction of that total cost. The violence results in a lot of medical bills and lawsuits. The average cost of one of these court cases runs $6,000-7,000.

Ceri: That amount would pay the cost for supplements for the entire institution.

Schoenthaler: If you look at your ratio of the savings to the cost of vitamin supplementation, it is 100 to 1, or 1000 to 1.

Ceri: So how did you get involved in the IQ issue in England?

Schoenthaler: The Oklahoma study was the first one where we measured intelligence. We wanted to look at intelligence because previously, when we changed the diet of children in public schools, their grades shot up dramatically. At the same time we were doing our study, Benton and Roberts were conducting an almost-identical study of 90 children in England. When the results of both studies were released together — both showing a six-point IQ gain, neither showing any gains in verbal IQ, both showing gains in non-verbal IQ—the only difference was that they used children in Wales and we used kids in Oklahoma. The effect was dramatic and startling to the people of the UK. They concluded that their children were undernourished and rushed out to buy vitamin supplements. Some store shelves were stripped bare in a matter of two or three days. I’ve read estimates that 80-150 million Pounds Sterling were spent in one week. That translates into 300 million US dollars in a country with only twice the population of California. The dramatic influence of the agreement between these two small studies was too powerful to downplay or ignore and it became a matter of public policy; it put a lot of egg on the face of the British medical establishment which had been telling the British public that their diet was fine and they needn’t worry. If that had been true, Benton and Roberts would never have gotten the results that they did.

Ceri: Did the medical establishment try to resist this conclusion?

Schoenthaler: Their reaction was that this needed to be replicated to find out if it was true or not, and quickly. This started the scientific controversy. They started attacking Benton’s study, which was published in Lancet, very severely. They did not like the way he measured the nutritional status of children, and those criticisms were valid. However, in his defense, even if he did not measure nutrition status right, it does not matter in regard to whether or not children do better on a supplement or a placebo.

Ceri: The results were the same...

Schoenthaler: Yeah. The IQ results hold true regardless. We just don’t know why if we don’t have the measure of their nutrition.

Ceri: You got a six-point IQ increase on average?

Schoenthaler: Net gain. That’s different, you always get a little gain in the placebo group, but the gain in the supplement group is larger by an average of six points.

Ceri: What was the spread? Some got more than six points, some got less...

Schoenthaler: IQ gains were all over the board. Truthfully, they’re not important in an early study of this size. The trial groups were too small to provide robust statistical information about IQ distributions. Later studies used hundreds of subjects.

Ceri: Out of curiosity, what were some of the highest improvements on IQ scores?

Schoenthaler: I think the larger ones were 16-20 points on the first study.

Ceri: What about the studies that followed that concluded there was no result.

Schoenthaler: There were two studies that came out negative — at least some people consider them negative. The first came out in May ’88 in the form of a letter to Lancet that said the work of David Benton was wrong (they didn’t have my study at the time) and that there was no link between supplementation and IQ.

Ceri: Was that true?

Schoenthaler: Well, that’s an interesting question. The people who did the study were very good nutritionists, but they had no background in psychology and they apparently didn’t consult a psychometric expert. They tried to measure nonverbal IQ using two measurements. One was a group-administered IQ test where you take a class and set them down for 20-25 minutes while they take a test which measures intelligence only in a very broad way. It’s normally used as a screening test because it’s only accurate to about 20 IQ points. In other words, it’ll sort out retarded children from normal children, and gifted from normal. With a test of this kind, you can’t pick up subtle IQ differences.

This kind of screening test has excessive variation. Subtle things that the tester does can drastically change the way the children perform on the test. Kids that don’t try the first time but do the second show huge IQ gains that aren’t real. Or, some children will try hard the first time and not the second and get huge deteriorations that aren’t real. To measure IQ accurately, you need one trained psychologist with one child, working one-on-one. With group tests, the kids know that in 20 minutes, as soon as the test is done, they can go outside and play. There is no substitute for accurate measurement.

The second IQ test they used was the same one we used, however, they decided not to follow the test’s instructions. The test has five parts to it and the instructions say to give all five parts and add up the scores to get the nonverbal IQ score. The test does let you get away with giving only four of the five parts of the test, and then multiplying the four-part total by 5/4 to get the IQ score. The researchers on this study came up with the brilliant idea of using only one of the five parts, not four out of five, not five out of five, only one.

Ceri: Did they report that in their paper?

Schoenthaler: Yes. But nobody picked up on it. It’s never been reported in the press. The single scale they chose correlates with IQ at 0.1. It is interesting that the test’s instruction manual cautions against such use of the test, warning that it is not a valid measure of IQ when used in this manner.

Ceri: What do you mean, an IQ correlation of 0.1?

Schoenthaler: When all five parts are used together, the correlation between the test score and intelligence is 0.9; with only one part it is 0.1. It’s analogous to asking someone to spell potato in a spelling quiz, and stopping them after they’ve given only the first two letters.

Keep in mind that this study started the controversy by claiming that supplementation didn’t work to increase IQ based on 1/5 of an IQ test (which is not reliable) and a screening test (that never claimed to be an IQ test).

Ceri: Who funded this study?

Schoenthaler: According to the authors, it was funded by the World Sugar Association.

Ceri: Do you think they provided funding to try to defend the poor quality of food that they produce?

Schoenthaler: I believe so. There has been a long debate in the UK literature about the caliber of the food between those who blame the sugar industry and those who support by saying that the UK diet is just fine. So it makes sense that the sugar industry would support a study to disprove their possible culpability for the poor diet of children in the UK. This study was formed and carried out rather quickly. From the timetable I’ve seen, they had to have their funds and protocol drawn up and implemented in a matter of days. What came out in the trial was that they had erred in entering their data into a computer file and didn’t even have the time to correct the errors before they published it in Lancet.

Ceri: You were able to get the actual raw data of their study, but you can’t talk about it. Can you go on record as to why you can’t talk about it?

Schoenthaler: I was notified by the prosecution lawyers that the defense was being given the data for the purposes of the trial, that they were copyrighting the data, and that I, in particular, was not allowed to discuss what the data shows. It was given to me to analyze only in relation to the trial. It had been subpoenaed by the defense and we had a legal right to look at it, but the data can go no further.

Ceri: Is your re-analysis of their data also restricted?

Schoenthaler: Yes.

Ceri: Is there any way to get it through scientific channels?

Schoenthaler: I would hope to see that, but it will be their choice and I have to honor their conditions in the meantime.

Ceri: Why do you think that they would be so concerned about their data being seen?

Schoenthaler: That’s real simple: there was another study that was published in the Lancet that reported negative findings—that the IQ could not be raised—and they gave us their raw data with permission to re-analyze and discuss it. We found that their statistics were wrong, and when properly re-analyzed, they showed a significant increase in IQ.

Ceri: Of what magnitude? What was the increase, compared to the six points in yours and Benton’s results?

Schoenthaler: The first two small studies showed six-point gains. This study showed a 2.4 point gain. Since then there have been seven other studies done and the average gain seems to be 3.5 points. So we had larger-than-average gains in the first two studies which can be explained by the small sample sizes of those studies.

Ceri: So this other study, this negative study, came out positive when you re-analyzed the data?

Schoenthaler: Yes. They published it in Lancet in 1990. It was a very well-controlled and proper replication of the Benton and Robert study — a well-done study.

Ceri: Did they ever acknowledge those errors? Did they testify at the trial?

Schoenthaler: The second author took the witness stand and admitted under oath that the data as a whole did indicate that IQ could be raised with supplements, and indicated that, upon re-analysis, their own data approaches statistical significance. We disagreed slightly on whether the final p value was 0.054 or 0.045, but either way it rounds out to 0.05 significance level which is quite a bit different from the conclusions of the original article which said that vitamin supplementation does not raise intelligence.

I was very pleased with their courage in being forthright and truthful about their own data, and I think that the re-analysis of their data was probably the reason that the first study’s research team did not openly cooperate.

Ceri: Especially when it’s based on such gross measurements and methodological flaws. What’s your guess about how the trial will come out?

Schoenthaler: My guess is that they are going to be found guilty. [Editor’s note: they were found guilty and fined less than a week after this interview was conducted].

Ceri: Of what?

Schoenthaler: The prosecution, in their opening statement, amended the charges against the company. The charges started out being that vitamin/mineral supplements could not raise the IQ of a child. The operative words are “a child.” It was not a question of whether or not the supplement could raise intelligence in general, it was whether the supplement could raise the intelligence of anyone. That was the initial charge when I left California for England, and that was the charge that we had prepared 200 pages of documents for. When I arrived in court on Monday morning, the opening statement of the prosecution (the Crown) changed the charges. No longer were they claiming that the supplements were not fit to raise the IQ of “a child,” it was now “the vast majority of children,” which they refused to define. They wouldn’t say if it meant 98%, 80%, or what.

I thought them [the Crown] very clever to realize they were going to lose on the original charge and to change the charge. I was surprised that the Court allowed them to change the charge at the beginning of the trial. Normally, the defense has a right to advance notice, at least in this country.

Ceri: Did the Defense object?

Schoenthaler: They objected strenuously that they were there prepared to defend what they were charged with, not the new charge, and the prosecution argued that the defense was notified that the new charge might occur over a year ago and considering that they were given notice over a year ago that it was proper notice.

Ceri: Were they given notice?

Schoenthaler: Apparently yes, since the Defense indicated that they were notified of dozens of other possible charges, but they came prepared to defend against the three charges that were brought forward, not all possible charges. This new charge was not one of the three. The court ruled for the prosecution. I thought this was a mistake.

Ceri: That might indicate that the chips were stacked against the Defendants.

Schoenthaler: It’s interesting. The case ultimately evolved to a test of the first and third charges. The first, the new charge, was that this particular brand of supplements was not fit to be “widespread and generally effective in raising IQ.” The second charge was that the outside of the box conveyed a false impression that the pills had a general widespread effect. Then there was the leaflet inside the box that explained what was published in Lancet and that also gave a false impression of a widespread and general effect. The trial came down to the question of whether or not the outside of the box or the leaflet gave, in theory, the impression that the supplement gave a widespread and general effect.

It was clear from the scientific evidence that unfolded that they definitely had a widespread and general effect as the defense claimed. The prosecution’s own experts kept backing off and conceding more and more. It was clear after their experts testified that the defense clearly won that point.

The reason I think they will probably lose the case, even though they won the war, is that you have to interpret the scientific evidence within the context of British law. It was clear that the supplements did result in widespread and general increases in IQ in children, but the British definition of false advertising, unlike American law, is based on a single person being misled. If one customer is misled by the box or the leaflet, then the company is guilty of false advertising. It does not have to be a “reasonable” person like it does in the United States.

Ceri: So all you have to do is bring someone forward to testify that they misinterpreted the label and you have your case.

Schoenthaler: Right, but the prosecution couldn’t find one. That’s one of the uncertainties about how the court will rule. This case has been pending for four years and the prosecution couldn’t bring forth a single person who said they didn’t understand the instructions. What it comes down to is whether or not the word “IQ” on a box of vitamin pills implies that it will raise the IQ of every child or just some children. That’s why I have a hunch that the company will lose on a technical point of law even though they won on the scientific merits of the case.

I do think that what we are going to see come out of it is in a sense ironic. This is the equivalent of the FDA in Great Britain. They openly admitted in court that these supplements will raise the IQ of children. I imagine that the manufacturer will change their advertising accordingly to say the equivalent of “as endorsed by the FDA” — that their vitamin pills are the “only ones in Great Britain” with “official testimony” that they do in fact raise the IQ levels of some children.

Ceri: We can see it: “Government Endorsed to increase IQ!”

Schoenthaler: The false advertising conviction will probably carry a 2000 Pound fine.

Ceri: The advertising value would probably be worth 100 times that amount.

Schoenthaler: I know the trial probably cost them a pretty penny because of the number of attorneys, expert witnesses and other things—over œ100,000 is my assumption. Plus the fine.

Ceri: How did your recent study of the kids in a California correctional facility come about? What is the situation in the U.S. relating to the connection between crime, violence and nutrition?

Schoenthaler: Following the 1987 Oklahoma study, California realized that we don’t have a handle on how to stop crime and solve it. The Oklahoma experiment was so positive, they wanted to try bringing it to California. In 1987, it was illegal in the state of California to do a placebo-controlled double-blind trial with juveniles, so the legislature had to change the law. In 1990, Senator Robert Presley led a bill through the Senate (35 to 0) and then through the House (67 to 1), signed into law by the Governor to explicitly authorize this particular trial. The Oklahoma study had 65 participants, the California study has over 400. Rather than looking at just minor offenses of any type, the California Youth Authority studied only those individuals who had violated rules that were so serious that formal charges were brought against them. The accused actually went to a hearing were the charges were found either true or unfounded. These are very serious offenders. These are violent people already inside the institution.

Ceri: You looked only at those?

Schoenthaler: Those were the focus because the state of California was particularly interested in cutting serious crime, rather than minor rules violations or simple insubordination. They wanted to look at the most serious offenses, in a large population. They also decided to look at intelligence. We hope to publish this research in the next five or six months, sometime after the first of the year.

Ceri: What about the makers of the Tandem IQ vitamin product? What are they doing in response to the trial?

Schoenthaler: In anticipation of the Court’s decision, Larkhall Natural Health (the manufacturer of Tandem IQ) has now re-released their product with a disclaimer stating that their supplement is for children who are not eating a nutritionally balanced diet or who may have nutritional deficiencies. They are still able to claim that the product increases IQ in some children.

Ceri: Can you summarize your experience of the trial?

Schoenthaler: There were seven studies presented at the trial, five positive, two negative. In one of the “negative” trials, statistical errors were found, and when re-analyzed, there turned out to be a significant gain in IQ among the supplemented children, compared to the placebo group. The other “negative” trial never really measured intelligence at all. In effect, all studies of the effect of vitamin supplementation on children’s IQ support a significant increase in supplemented children. This is why the Crown conceded their original allegation that IQ was not raised by supplements, and instead argued that this case was a question of what percentage of children had their IQs raised. How’s that for effect?

One of the British papers released the fact that the world’s leading expert on intelligence, Professor Hans Eysenck, concluded that the best estimate of the percentage of children that could show these IQ gains was 40% for both the UK and US.

Ceri: Those were fairly short-term studies. What would the cumulative increase in IQ be over a 10 year period?

Schoenthaler: We don’t know. But what we did find was that the average overall increase in IQ was 3.5 points. What isn’t often mentioned, but is very important, was that roughly one out of three kids showed a massive gain of 10 or more IQ points.

This is an interesting observation. Many people don’t get anything from smart drugs, but those who do often get major changes. Roughly one out of three people got these massive IQ changes of over ten points. When this gain is divided over the whole group, it comes out to only 3.5 points. This may seem small, but a 10 point increase in the responsive kids is really a massive gain.

Ceri: That’s a significant fraction of a standard deviation on the intelligence curve, isn’t it?

Schoenthaler: The standard deviation is 15 points, but to put it in a real-life perspective, the IQ difference between an “average” American and a doctor, lawyer, or professor is only about 11 points. The gain observed in one out of three people is the same as might be required for an average American to aspire to be a doctor, lawyer or professor.

Ceri: How long did that much gain take?

Schoenthaler: Three months for that much intellectual gain.

Ceri: Thank you for your interesting responses, Dr. Schoenthaler. It has been a pleasure speaking with you.