If you’ve been coaching for any length of time, you’ve seen it play out in real time.
Two clients read two “credible” sources. They walk in with two different conclusions. Both are certain. Neither is interested in nuance.
At that point, you’re not really coaching nutrition behaviors. You’re coaching interpretation, and sometimes you’re coaching identity.
That’s not a character flaw. It’s a human brain feature, and it’s one of the reasons nutrition science can feel uniquely polarizing in public conversation (Garza et al., 2019; Goldberg & Sliwa, 2011; Ruxton, 2022).
This blog is meant to connect the first two CEC articles in this series. Article 1 focused on nutrition/eHealth literacy and why online evaluation skills can be inconsistent. Article 2 examined the “food as medicine” movement and why literacy and access determine whether programs work in the real world. This piece addresses the bridge variable: cognitive bias, especially confirmation bias, plus the overload and media dynamics that turn nutrition into a two-camp argument (Peng & Shen, 2024; Ramírez & Carmona, 2018; Goldberg & Sliwa, 2011).
Nutrition Information Isn’t Processed Like a Spreadsheet
We like to imagine humans as rational consumers of evidence. We read, compare, and weigh the data, then decide.
In practice, nutrition information is filtered through cognitive biases, which are predictable patterns of deviation from purely rational judgment. These biases shape what we notice, what we trust, and what we dismiss (Larrick et al., 2022; Peng & Shen, 2024).
Bias is not limited to “the public.” It can be applied to research design, guideline development, media reporting, and social media dissemination. In other words, bias can influence both the production and the consumption of nutrition information (Kraak, 2023; Larrick et al., 2022; Ruxton, 2022).
That’s one reason “nutrition confusion” doesn’t resolve itself with more content. Sometimes the system is producing friction faster than individuals can process it (Goldberg & Sliwa, 2011; Ramírez & Carmona, 2018).
Confirmation Bias: The Great Evidence Filter
If there is one bias that explains a large portion of nutrition debates, it’s confirmation bias.
Confirmation bias is the tendency to interpret new information in ways that support existing beliefs, while discounting information that threatens those beliefs (Larrick et al., 2022; Kraak, 2023). It can show up as selective attention, biased interpretation, or a stronger willingness to accept evidence that “fits” one’s values, identity, or prior conclusions (Peng & Shen, 2024).
This isn’t limited to clients scrolling social media. It can appear in research as early as the study’s conceptualization, when investigators search for supportive information, prioritize certain outcomes, or interpret ambiguous information in belief-consistent ways (Larrick et al., 2022). It can also appear in high-stakes policy processes, where advisory committee members may bring pre-existing professional commitments that shape how evidence is weighed (Kraak, 2023).
This is also why conflicts of interest matter so much in public perception. Industry funding doesn’t automatically invalidate a study, but it can heighten scrutiny, trigger skepticism, and amplify polarization when the public perceives incentives that could bias conclusions (Garza et al., 2019; Larrick et al., 2022).
The Confidence Trap: When Feeling Certain Signals Risk
One of the most practically useful insights in this literature is uncomfortable: confidence does not reliably track accuracy.
Peng and Shen (2024) found that misinformation beliefs across different health domains were correlated, suggesting a generalized susceptibility rather than topic-specific “gullibility.” They also reported evidence consistent with overconfidence in health literacy and with metacognitive monitoring errors, in which individuals were unaware of their limitations in identifying misinformation (Peng & Shen, 2024).
This matters for trainers because clients rarely bring you misinformation timidly. They usually bring it with conviction. They often say, “I did the research,” and they genuinely believe they did.
What they often did was assemble a story that felt coherent and consistent with their identity.
More Effort Doesn’t Always Mean Better Conclusions
It’s tempting to assume that people who think harder will do better.
Peng and Shen (2024) complicate that assumption. They found that greater susceptibility to health misinformation was associated not only with lower objective health literacy but also with certain information processing patterns, including elaboration and selective scanning (Peng & Shen, 2024).
In theory, elaboration is more effortful and systematic. In practice, elaboration can become directional when it is driven by motivated reasoning. People may process deeply, but they may process in a way that defends a belief rather than evaluates a claim (Peng & Shen, 2024).
Selective scanning is even more straightforward. It describes attending to information that aligns with prior beliefs while skimming past discordant cues. It feels like research, but it functions like reinforcement (Peng & Shen, 2024).
Information Overload: When the Noise Becomes the Message
Not all nutrition confusion starts with certainty. Sometimes it starts with overload.
Ramírez and Carmona (2018) found that even when individuals understood the link between diet and disease, conflicting messages from public and interpersonal sources created confusion and a sense of being overwhelmed. They also suggested that what is sometimes labeled “fatalism” may reflect a reasonable response to an information environment that feels contradictory and unresolvable (Ramírez & Carmona, 2018).
That’s a major coaching point.
Some clients are not resisting change. They are trying to make sense of a nutrition environment where the loudest message is that nothing can be trusted. When overload is the problem, adding more information can backfire. Clarity, simplicity, and actionable skills often work better (Ramírez & Carmona, 2018; Goldberg & Sliwa, 2011).
Why Nutrition Science Gets Reduced to “Good” and “Bad”
Nutrition science is complex, context-dependent, and often probabilistic. Communication rarely is.
Goldberg and Sliwa (2011) described how nutrition communication is challenged by multiple competing voices, including government agencies, researchers, industry, advocacy groups, healthcare professionals, and media outlets. These messages can differ in rigor, transparency, and incentives, producing a cacophony that overwhelms consumers and erodes trust (Goldberg & Sliwa, 2011).
When nuance gets compressed into binary claims, reductionism becomes the default. Foods become “good” or “bad.” Compounds become “miracle” or “toxic.” Diet patterns become moral identities (Goldberg & Sliwa, 2011).
This is not only a media issue. It is also a human preference for cognitive simplicity in the face of complexity, which is strengthened under overload (Ramírez & Carmona, 2018).
Polarization, Trust, and the Methodological Reality of Nutrition Research
Some of the public’s frustration with nutrition science stems from real methodological limitations.
Ruxton (2022) highlighted common limitations in observational nutrition research, including measurement error, confounding, bias, and representativeness. Those limitations can contribute to inconsistent findings, which then become easy material for sensational headlines (Ruxton, 2022).
Williams et al. (2020) also discussed the challenges inherent in developing dietary recommendations for noncommunicable disease prevention, including issues with dietary assessment, misreporting, confounding, and changing exposures over time. Even strong frameworks cannot erase all uncertainty (Williams et al., 2020).
When the public is exposed to shifting headlines without a clear explanation of why science evolves, trust can erode.
Garza et al. (2019) identified multiple domains that influence public trust in nutrition science, including transparency, reproducibility, conflicts of interest, accountability, equity, and information dissemination. When those domains are perceived as weak, polarization becomes more likely because people search for certainty elsewhere (Garza et al., 2019).
Social Media: The Bias Amplifier
Social media doesn’t just spread information. On the contrary, it sorts it, rewards it, and repeats it.
Molenaar et al. (2023) noted that nutrition, food, and cooking content on social media carries substantial emotional tone, and the broader social media nutrition environment often includes misinformation, conflicting messages, and idealized, restrictive dietary narratives. The lack of regulation and the role of algorithms can amplify content that is attention-grabbing rather than evidence-based (Molenaar et al., 2023).
Probst et al. (2025) found that engagement metrics were higher for biased and lower-quality videos in a multiple sclerosis nutrition context. That pattern aligns with the idea that belief-consistent content is more likely to be accepted and shared, even when it does not accurately represent evidence (Probst et al., 2025).
Furthermore, an important nuance from Probst et al. (2025) is that “reputable sources” do not guarantee good information. Their analysis suggested that source credibility heuristics can override critical appraisal. In practice, people may lower scrutiny when content comes from a physician or another authority figure (Probst et al., 2025).
Why This Matters for “Food as Medicine”
Per article two in this series, the food-as-medicine movement is growing, and many programs now rely on digital tools, media messaging, and broad public buy-in. That means cognitive bias is not a side issue; it is a delivery issue (Williams et al., 2020; Garza et al., 2019).
If people interpret food-as-medicine as a rigid ideology, they may adopt overly restrictive patterns that reduce adherence, increase anxiety, or worsen their relationship with food. If they interpret it as an evidence-informed framework, they’re more likely to build sustainable patterns that support health and performance (Goldberg & Sliwa, 2011; Hickson et al., 2024).
This is also where ethnonutrition matters. Food choices are cultural, social, and identity-linked, and nutrition messaging that ignores cultural context risks being ineffective or counterproductive (Jacob et al., 2021).
What Trainers Can Do: Coach the Process, Not the Position
Your best move is rarely to argue a nutrition “side.”
Your best move is to coach your client in deciding what to believe.
Here are a few questions that work well in-session because they reduce defensiveness and promote reflection:
- “What would you need to see for you to change your mind?” (Peng & Shen, 2024)
- “Where did you first hear this, and what else have you read that disagrees with it?” (Goldberg & Sliwa, 2011)
- “Is the message giving you a usable pattern, or is it selling certainty?” (Garza et al., 2019)
- “How is this affecting your training, recovery, and consistency?” (Hickson et al., 2024)
That last question keeps the conversation grounded. Trainers are uniquely positioned to connect nutrition choices to performance outcomes, adherence, and day-to-day functioning.
Scope Reminder: Where Trainers Stay Strong
You do not need to be a dietitian to be effective here.
However, you do need to stay inside scope and refer out when appropriate.
If a client’s nutrition behavior is showing rigidity, anxiety, or disordered patterns, remember that biased food evaluation can persist even when nutrition knowledge is intact (Martinelli et al., 2023). That is a referral moment.
You can support literacy, skill-building, and behavior change. You can also encourage clients to seek credible sources and help them slow down before adopting extreme dietary changes based on social media certainty (Garza et al., 2019; Molenaar et al., 2023).
The Takeaway
Nutrition polarization is not just about the science. It is also about how humans process information under uncertainty, overload, and identity threat (Ramírez & Carmona, 2018; Peng & Shen, 2024).
If you want continuity across this series, here’s the thread:
- Article 1: evaluation skill and literacy are uneven in the real world.
- Article 2: Food-as-medicine works best when access and literacy are built into the design.
- This blog: biases and overload shape what clients accept as true, which determines whether any evidence-informed approach is usable.
In my next article, we’ll move from “what’s happening” to “what to do about it,” focusing on practical strategies trainers can use to support stable decision-making in a noisy nutrition environment (Goldberg & Sliwa, 2011; Garza et al., 2019).
References
Garza, C., Stover, P., Ohlhorst, S., Field, M., Steinbrook, R., Rowe, S., … Campbell, E. (2019). Best practices in nutrition science to earn and keep the public’s trust. American Journal of Clinical Nutrition, 109(1), 225–243. https://doi.org/10.1093/ajcn/nqy337
Goldberg, J., & Sliwa, S. (2011). Communicating actionable nutrition messages: challenges and opportunities. Proceedings of the Nutrition Society, 70(1), 26–37. https://doi.org/10.1017/s0029665110004714
Hickson, M., Papoutsakis, C., Madden, A., Smith, M., & Whelan, K. (2024). Nature of the evidence base and approaches to guide nutrition interventions for individuals: a position paper from the Academy of Nutrition Sciences. British Journal of Nutrition, 1–20. https://doi.org/10.1017/s0007114524000291
Jacob, M., Teixeira, C., Bautista, D., & Ramos, V. (2021). Ethnonutrition. Ethnobiology and Conservation, 10. https://doi.org/10.15451/ec2021-10-10.35-1-8
Kraak, V. (2023). Perspective: Examining conflicts of interest for professional service within the 2020 Dietary Guidelines Advisory Committee. Advances in Nutrition, 14(3), 432–437. https://doi.org/10.1016/j.advnut.2023.03.009
Larrick, B., Dwyer, J., Erdman, J., D’Aloisio, R., & Jones, W. (2022). An updated framework for industry funding of food and nutrition research: Managing financial conflicts and scientific integrity. Journal of Nutrition, 152(8), 1812–1818. https://doi.org/10.1093/jn/nxac106
Martinelli, C., Chami, R., & Reid, S. (2023). The investigation of biases in the evaluation and knowledge of foods’ healthiness and disordered eating in a community sample. Canadian Journal of Behavioural Science / Revue Canadienne Des Sciences Du Comportement, 55(1), 14–22. https://doi.org/10.1037/cbs0000316
Molenaar, A., Jenkins, E., Brennan, L., Lukose, D., & McCaffrey, T. (2023). The use of sentiment and emotion analysis and data science to assess the language of nutrition-, food- and cooking-related content on social media: A systematic scoping review. Nutrition Research Reviews, 37(1), 43–78. https://doi.org/10.1017/s0954422423000069
Peng, R., & Shen, F. (2024). Why fall for misinformation? Role of information processing strategies, health consciousness, and overconfidence in health literacy. Journal of Health Psychology, 30(8), 2030–2045. https://doi.org/10.1177/13591053241273647
Probst, Y., Saffioti, E., Manche, S., & Eaton, M. (2025). Examination of social media nutrition information related to multiple sclerosis: A cross-sectional social network analysis. Public Health Nutrition, 28(1). https://doi.org/10.1017/s1368980025100943
Ramírez, A., & Carmona, K. (2018). Beyond fatalism: Information overload as a mechanism to understand health disparities. Social Science & Medicine, 219, 11–18. https://doi.org/10.1016/j.socscimed.2018.10.006
Ruxton, C. (2022). Interpretation of observational studies: The good, the bad and the sensational. Proceedings of the Nutrition Society, 81(4), 279–287. https://doi.org/10.1017/s0029665122000775
Williams, C., Ashwell, M., Prentice, A., Hickson, M., & Stanner, S. (2020). Nature of the evidence base and frameworks underpinning dietary recommendations for prevention of non-communicable diseases: A position paper from the Academy of Nutrition Sciences. British Journal of Nutrition, 126(7), 1076–1090. https://doi.org/10.1017/s0007114520005000
