A comprehensive review published in 2023 has shed light on the reasons why students sometimes reject or accept ideas that conflict with their previous ones. This systematic study reviewed 86 peer-reviewed articles that were published in science education journals, selected in order to promote a better understanding of conceptual conflict events. It aimed at documenting the preliminary conditions, the process and the outcomes of cognitive-conflict events. The studies cited here mostly used questionnaires and in-depth interviews, to ask learners about the intimate reasons why they adhere or do not adhere to discrepant ideas or conflicting information. The results show an impressive list—in terms of length—of 11 distinct and possible outcomes of cognitive conflicts: (1) Subassimilation: A response where learners attempt to assimilate the new information but do so in a fragmented or incomplete manner, falling short of full understanding, and thus being unable to even detect contradictions; (2) Ignoring: Overlooking the anomalous data as if it wasn’t presented; (3) Rejection: Outright refusal to accept the anomalous data because it conflicts with existing beliefs; (4) Exclusion: Acknowledging the data but considering it an outlier that does not affect the general understanding; (5) Uncertainty about Validity: Doubting the truthfulness, source, or reliability of the anomalous data; (6) Uncertainty about Interpretation: Being uncertain about how to interpret the anomalous data, even though its validity is not questioned; (7) Abeyance: Temporarily suspending judgment about the anomalous data, neither accepting nor rejecting it; (8) Reinterpretation: Modifying the interpretation of the data to fit existing beliefs; (9) Belief Decrease: Reduction in the confidence of the initial belief but without a shift to a new understanding; (10) Peripheral Change: Making minor adjustments to the existing belief system to accommodate the new data without fundamental conceptual change; (11) Theory Change or Belief Change: Undergoing a fundamental shift in understanding or beliefs to accommodate the new data, representing genuine conceptual change.

Surprisingly, depending on the studies analyzed, the review (Potvin, 2023) found that the distribution of responses from one category to the other was rather evenly distributed, indicating that the desired category (theory or belief change) was largely minority, and that the obstacles to the consideration on new and initially uncomfortable ideas are many. The review also suggests that resolving cognitive conflict requires additional processing activities, such as discussion or structured reflection, that can help learners integrate new information effectively, and thus that cognitive conflict alone clearly appears insufficient. Finally, it identifies several factors that influence the effectiveness of presenting contradicting information to learners. These include rather trivial things such as learner’s cognitive abilities, epistemological beliefs, affective and attitudinal states, initial confidence in their preconceptions, but also other interesting things like the initial availability of a conceptual “plan B” (p. 84). The review insists on this finding. Conceptual change seems to require that one already has a conceptual backup plan if he/she is to be productively contradicted.

In recent years, a growing body of research has shown that modifying or discarding initial conceptions might not be a necessity, nor likely, in the process of conceptual learning in science. This insight has emerged from studies using mental chronometry methods (Babai et al., 2010; Brault Foisy et al., 2021; Lin et al., 2023; Potvin et al., 2015; Shtulman and Valcarcel, 2012; Wen et al., 2024), neuroimaging (Brault Foisy et al., 2015; Masson et al., 2014), and electroencephalography (Skelling-Desmeules et al., 2021; Zhu et al., 2019) showing that even when correct responses are given, underlying misconceptions can still influence thought processes, even in experts (Allaire-Duquette et al., 2021; Potvin et al., 2020). Such findings suggest that learning new concepts does not eliminate old ones, and that new conceptions, when learning is successful, new concepts coexists with pre-existing ones, requiring the inhibition of the latter for expertise development. This notion of coexistence challenges traditional views in science education and urges a reevaluation of pedagogical approaches to incorporate the understanding that multiple conceptions can persist simultaneously within a single learner. It suggests that a conceptual change event should be understood as a conceptual prevalence shift rather that a rejection or a transformation.

Among the recommendations that have emerged from this observation (Potvin, 2017), it is suggested that we simply stop waging explicit wars on misconceptions. It argues against systematically discrediting misconceptions, suggesting instead that expertise requires the ability to inhibit rather than to eliminate them. It also recommends the use of different sequencings in teaching efforts and suggests that for students without an already available rival theory to the scientific models being taught, direct confrontation with anomalies may not lead to the desired learning outcome. An alternative might be to first introduce the desired scientific model that can then facilitate productive “theory-theory” conflicts. In this perspective, cognitive conflicts should be resolved based on the systematic and explicit evaluation of cognitive utilities of competing models rather than solely on their scientific accuracy (or inaccuracy). This could also prevent some negative impacts on students’ self-concept and interest in science, since learners often believe that discrediting an idea discredits its bearer.

In this “coexistence” perspective, it also appears crucial to reflect on the durability of prevalences. To achieve lasting expertise, educational programs and well-designed sequences may be required, not mere punctual interventions (Potvin, 2017). Indeed, apparent reversions to initial conceptions are common. So strategies that secure the durability of the prevalence of desired conceptions should be favored. It also suggests that starting science education early in life and with a focus on didactical quality could prevent the eventual and precocious establishment of undesirable conceptions, therefore avoiding the later necessity to undergo costly changes.

The constructs of coexistence and prevalence naturally lead to a representational-pluralist perspective that is even broader and more encompassing than mere conceptual prevalence framework. It also opens more educational opportunities. Representational pluralism, as it has recently been suggested and developed (Bélanger and Potvin, 2022; Bélanger et al., 2022), is based on an explicit rejection of an ideal of human knowledge consisting in the possession of a “single best” representation of a phenomenon. For instance, the intuitive idea that rolling soccer balls stop by themselves after being kicked is “true enough” in a sport context, although the Newtonian account of what happens will be expected in a school context. Cognitively speaking, such intuition does sufficiently proper job in its context of use. Generally, a pluralist perspective posits that individuals naturally tend to develop or preserve multiple representations of a given phenomenon (Horst, 2016). And this happens at any level of expertise. Just like students can keep useful intuitions despite having developed some level of understanding and belief of scientific representations, scientists often develop various theories and models that enable people to think about phenomena and act on them efficiently according to their various purposes.

One useful analogy for thinking about pluralism compares representations to tools (Bélanger and Richard, 2024). Tools are artifacts created by humans to help us achieve certain goals. No tool is said to be ‘truer’ than another; a tool can only be said to produce better performances than another for such and such a purpose. For instance, in some circumstances an impact driver is more useful than a screwdriver but not in others. Likewise, representations can be thought of as intellectual technologies, intentionally developed for a purpose, with a value relative to this purpose. In this pluralist perspective of cognition, science learning is to be seen as the acquisition of powerful representations offering new capabilities for thinking and acting in the world. But just like we do not throw away our screwdrivers after having bought an impact driver, we should not be expected to stop using intuitive representations. Conceptual change research has univocally taught us that intuitive representations are quite persistent; the pluralist perspective wishes to stress that part of this is due to their cognitive usefulness (Ohlsson, 2013). Although, as scientists, we may be despondent about the level of scientific falsehood of these intuitions and would find it quite convenient that they give way during learning, the fact is that many of them might be there to stay. Pluralism put forwards the idea that this cohabitation is not only inevitable, but normal and even something positive.

From such pluralist perspective, students’ intuitive ideas and scientific concepts can both have value in different contexts. Successful learning then implies not just the acquisition of scientific concepts but also the development of the ability to navigate between different representations and to apply them appropriately in various contexts. This is precisely what you are expected to do with a toolbox. Accordingly, teaching strategies should recognize and engage with the diversity of representations and develop ways to navigate and utilize this diversity effectively (Bélanger and Richard, 2024). For instance, the “conceptual prevalence” model suggests teaching strategies that do not aim to eliminate students’ intuitive representations but instead focus on developing strategies for when and how to prioritize scientific concepts over intuitive ones. It is important to emphasize here that the pluralism we are talking about does not call for the endorsement of false ideas that should be considered as equally valid as, or equivalent to, scientific knowledge. The project of successful science communication is not a relativistic project.

But taking pluralism seriously can bring us further in rethinking what we consider cognitive expertise about a domain to be. Pluralism insists on the fact that the truth of scientific representations is not all there is for human cognition: practically useful intuitive representations also have cognitive value. But just as the cognitive interests of humans are not limited to true representations, they are not limited to practical usefulness neither. Sometimes, we have interest in what is merely possible, because what is only possible today might reveals itself to be a truth tomorrow. Humans often have to deal with states of incomplete knowledge, and if they are to cognitively go forward nevertheless, they must do so in terms of the possibility of various competitive representations.

A strong pluralist view can go even further and acknowledge that there is some cognitive interest in what is judged entirely false and practically useless. Sometimes, it is a good thing to have some understanding of what other people think, even if we reject it. The cognitive value of such representations is social in nature. Giving students access to scientific knowledge is eminently desirable but enabling them to have empathic and fruitful interactions with people holding different views is also part of educating them to be citizen within an increasingly complex society.

Ultimately, a representational pluralist perspective on science education stresses that the learning of science consists not only in the understand scientific theories and models, but also in their effective use, as well as in the development of epistemological competency, critical thinking skills and empathic capabilities. Above all, it stresses the importance of treating representations (scientific and intuitive alike) as objects constructed by humans, whose properties (truth, usefulness, simplicity, understability, etc.) can be discussed and deliberatively weighed by students (Bélanger and Richard, 2024).

A pivotal aspect of bridging science education insights with public science communication efforts may lie in comprehending how individuals respond to conflicting information. Research in science education has highlighted a spectrum of responses, from outright rejection to the acceptance of new ideas, underscoring the complexity and difficulty of conceptual change. By anticipating the public’s reactions to scientific information that contradicts their pre-existing beliefs, science communicators can tailor their messages to mitigate resistance and foster open-minded engagement. Implementing structured reflection and discussion, as suggested -specifically for example, in socioeducational settings, could facilitate the integration of new information, making the public more receptive to scientific messages.

Given the difficulty associated with achieving conceptual change, especially in adults, science communication efforts may yield more significant impact by focusing on younger audiences. Early education presents an opportune window for shaping scientific understanding and attitudes, potentially circumventing the entrenched misconceptions that are harder to address in later life. This strategic focus does not diminish the importance of engaging adult audiences but highlights a complementary pathway to bolstering the public’s scientific literacy over the long term.

The sequencing of scientific messages and the provision of “conceptual Plan Bs” are critical for enhancing the acceptance of scientific arguments and the effectiveness of rebuttals, like in fact-checking efforts. Science communication initiatives could benefit from first introducing audiences to alternative scientific concepts that allow for a shift in understanding without triggering sterile defensiveness or principled resistance. And then begin discussion about the respective value of possibly conflicting ideas. This approach requires a nuanced understanding of different publics and the tailoring of messages to meet their specific needs and initial levels of understanding.

The complex nature of scientific knowledge demands a communication strategy that transcends simple true/false dichotomies, which can foster a dogmatic view of science. Instead, presenting scientific controversies as evolving narratives that welcome multiple perspectives encourages a more sophisticated engagement with scientific content. This approach not only respects the intelligence of the public but also nurtures critical thinking and a deeper appreciation for the provisional nature of scientific knowledge.

Embracing a pluralistic approach to science communication, which acknowledges the coexistence of multiple, often conflicting, representations of scientific phenomena, can enrich public engagement with science. By presenting rival theories (and the contexts in which they may hold validity), as well as more prospective possible theories (with a discussion of their level of plausibility), communicators can foster a more nuanced understanding of science as a dynamic and evolving field. This approach also highlights the value of scientific debate and the exploration of different theoretical perspectives.

In the face of pervasive misinformation, it is tempting to adopt a defensive stance aimed at correcting misconceptions. However, a more effective strategy may be to equip the public with the tools, like epistemic concepts and criteria, to critically discuss and evaluate the relative merits of partially or completely contradictory scientific conceptions, in light of facts. This approach fosters an environment where critical thinking and epistemological reflection are valued over the mere acceptance of information as indisputable fact. This suggestion should however be carried out with competence, if we do not want constructive skepticism to turn into dry mistrust toward partial states of knowledge necessarily omnipresent in scientific research. And looking in the other direction, critical powers backed by scientific knowledge should be used respectfully in discussions with people holding what can be considered naïve of false representations. Preserving the social fabric requires some level of empathy.

We conclude this commentary by reiterating that we believe that insights from science education research could provide an interesting framework for improving public understanding of science, although obviously not everything can be imported, as there are important operational differences. But by adopting strategies that acknowledge the complexity of learning and conceptual shifts (rather than “change”), it is not unreasonable to think that science communication initiatives could engage audiences more effectively and foster a well-informed, more epistemologically competent, and critically thinking public. As we continue to face the challenges of misinformation and mistrust, the mutual exchange of knowledge between science educators and communicators becomes increasingly important. We invite experts in public communication to share their insights in the hope of enriching the dialogue between our fields and collectively advancing public and student engagement with authentic scientific activities and outcomes.