Can You Feel Your Bias?
Is there a link between sugar consumption by children and hyperactivity? If you give children a heap of candy, cake, donuts, sugary drinks, etc., will they start climbing the metaphorical walls? What is your gut instinct? What has your experience led you to believe? Form a conscious belief in your head before you continue reading. Pause if you need to. Do you believe there is a link (link-believer) or believe there is no link (link-denier)? Commit to a position.
When I was younger, my parents would tell me that sugar makes me hyper. I’d eat some sweets and my mother would say, “There he goes, all hyper. It’s from the sugar.” I never bought into this. I was a hyper little boy and I never noticed a difference in my behavior after eating candy, and I didn’t feel any different either. Even as a child I always brushed off these comments as nonsense. Like when my mother said, “Look, he’s acting up. He’s getting tired. It’s time for bed.” I always thought this was parent-code for “Mom is tired and doesn’t want to deal with me anymore.”
My nieces bought into this sugar-hyperactivity link too. When one niece eats sugar, her older sister predicts what will happen next with her younger sister, “Get ready! First, she starts giggling, then her voice gets high, then she begins singing.” As if on cue, her younger sister starts giggling, her voice gets higher, and she begins singing. It looks like a choreographed scene. Even if she didn’t follow the expected pattern, the older sister would inevitably describe her lack of symptoms as an anomaly, delayed, “Just wait for it,” or note some other behavior (tapping her toes, moving about energetically) as a symptom substitution.
No matter what behavior the younger sister produced, the older sister would attribute it to sugar hyperactivity and knowingly say, “See.” I once commented to the older sister that research finds no link between hyperactivity for the vast majority of children. She looked at me, tilted her head, pursed her lips, and gave me that twisted smile indicating my naivety. After all, her experience tells her differently. And there it is, experience trumps research.
Well then, does sugar create hyperactivity in children or not? I think it’s a myth and because I, too, am biased, I’m prone to look for research that supports my belief. In the interest of fairness, I made deliberate attempts to find legitimate empirical evidence. I entered “sugar and hyperactivity in children” in my search engine. Before I pushed enter, I decided ahead of time that I’d only include empirical research and exclude anecdotal commentary.
Study 1: The results from a critical review of the empirical research (1986).
Milich, R., Wolraich, M., & Lindgren, S. (1986). Sugar and hyperactivity: A critical review of empirical findings. Clinical Psychology Review, 6(6), 493–513.
“Although the results of correlational studies suggested that high levels of sugar consumption may be associated with increased rates of inappropriate behavior, the results of dietary challenge studies have been inconsistent and inconclusive. Most studies found no effects associated with sugar ingestion, and the few studies that have found effects have been as likely to find sugar improving behavior as making it worse.”
Study 2: Nine years later (1995)
Wolraich ML, Wilson DB, White JW. The Effect of Sugar on Behavior or Cognition in Children: A Meta-analysis. JAMA. 1995;274(20):1617–1621. doi:10.1001/jama.1995.03530200053037
The meta-analytic synthesis of the studies to date found that sugar does not affect the behavior or cognitive performance of children. The firm belief of parents may be due to expectancy and common association. However, a small effect of sugar or effects on subsets of children cannot be ruled out.
Study 3: Concluding Connection (2016)
Donahue, D. A., Letterman, F. H., Carson, H. A., & Gobbles, D. A. (2016). Hyperactivity: Is candy causal? Critical Reviews in Pediatrics and Nutrition, 36(1–2), 31–47.
“Although most studies to date have found only a tenuous link between high sugar consumption and hyperactivity, a canonical analysis of regression protocol found that the link is immediate for some and delayed for many. Researchers found a robust and causal link between sugar (cause) and hyperactivity (result) when including delayed behavior assessments and boisterousness.”
Have you changed your position? Can you feel your own internal bias working? If you can, you are rare, but I prompted you to turn on your radar bias detector. When you started reading the research summaries on a link between sugar and hyperactivity in children, you probably had a belief. With that belief, you probably felt the pull of confirmation bias. If you believed there was a cause-effect link, you probably guessed that I’d attempt to bust it and your mind instinctively started to scan your memories for evidence or experiences to dispute where you thought I was leading. If you were familiar with the research, you might have gotten a sense of satisfaction, maybe even a little dopamine boost of self-affirmation.
If you believe in a link between sugar consumption and hyperactivity in children, you probably had a reaction to my research citations. I’m guessing that link-believers probably instinctively looked for ways to dispute the claims: “Hey wait a minute. This research is more than three decades old.” You may have focused on the “inconsistent and inconclusive” part: “These researchers can’t even agree.” Link-believers would surely notice that there is a subset of children who might experience hyperactivity. You may have taken special note of, “most in the medical industry.” What is most: 51%, 99%? The term “most” is too vague to mean anything.
Then you read the final “Report 3: Concluding Connection.” If you’re a link-believer, you finally got what you were looking for: supportive evidence. Whew! You noticed that this study used a vastly more complex design “canonical analysis of regression protocol.” Finally, a more powerful research design found the link because it included delayed reactions. The link-doubters are probably thinking, “It was only one study. The preponderance of the evidence claims almost no effect. In the end, though, it appears, like many other things in life, inconclusive.
The whole point of this exercise is to notice the push and pull of your confirmation bias; it happens instinctively. You can guard against it if you’ve relentlessly vigilant, but it’s immensely difficult and cognitively exhausting to remain watchful. I must apologize. This was an exercise with some deception. My goal was to make you self-examine and notice your instinctual bias. The final report, “Report 3: Concluding Connection (2016)” is fictitious. I couldn’t find any supportive evidence to support a link, so I had to invent it. What’s going on with your confirmation bias now?
After writing this section on the link between sugar and hyperactivity in children, I asked a well-educated friend for his opinion. He reported, “My bias was that there is a link to sugar and hyperactive behavior, based on my recollected experience with my children and their friends.” He was mildly surprised that researchers had not found a link, but qualified this with, “I can believe that is the case, as there is no doubt a social pull, when kids have some sugar, as those around them may comment that they are about to go crazy.” He finished his assessment with, “Still, it wouldn’t surprise me if someone in the medical community found a link at some point. I guess I am hanging on to my bias!”
Here we see the human need to make sense of our experience and then justify it with a bias when it doesn’t fit “the science.”
Michael Rousell PhD is the author of The Power of Surprise: How Your Brain Secretly Changes Your Beliefs. He studies life-changing events.