History or Hogwarts? Learning from Fiction to Transform Factual Reading

Students are frequently assigned factual readings to learn disciplinary knowledge, yet such readings are often experienced as effortful, externally imposed, and assessment-driven. By contrast, fiction read for pleasure can absorb readers so deeply that they voluntarily learn characters, histories, places, rules, and lore from imaginary worlds. This contrast is not simply a matter of facts versus fiction. It reflects several overlapping dimensions, including narrative versus expository structure, assigned versus self-chosen reading, assessment pressure versus voluntary engagement, and isolated information versus richly connected worlds. In this Perspective paper, I provide: (1) an examination of why fictional worlds and fiction more generally can be so engaging, (2) an overview of mechanisms that support engagement, comprehension, and memory, and (3) design principles for making assigned factual readings more engaging. This is a conceptual synthesis rather than a systematic review or report of new empirical data. I distinguish less-mutable differences between pleasure reading and assigned reading from mutable design levers that educators and authors can use. My goal is not to suggest that educational readings should become fiction, nor that pleasure is a substitute for rigour, but rather to identify how factual readings can be made more engaging by leveraging narrative schema, mental imagery, prior knowledge, self-relevance, topic interest, and social reading, alongside assessments that reward meaningful understanding.

In educational settings, students are often required to read factual texts with the goal of acquiring knowledge about the world and remembering it later (Madan, 2024; Willingham, 2009, 2021). These readings may be accurate, useful, and carefully selected, but they are also often externally assigned—surrounded by deadlines and tied to assessment. As a result, reading for learning can become an exercise in compliance and short-term performance rather than curiosity and sustained engagement (Deci & Ryan, 1985; Guthrie & Wigfield, 2000). The challenge is not new: for centuries, practitioners of the ars memorandi used vivid imagery and spatial organisation to make prescribed content memorable (Jaeger, 1983; Patten, 1990). This problem matters because reading for pleasure appears to be declining in many contexts, with recent time-use evidence from the United States indicating marked declines in daily reading for pleasure from 2003 to 2023 (Bone et al., 2025). At the same time, recent research on reading for pleasure cautions against simplistic claims: pleasure reading is valuable, but its benefits and implementation depend on context, access, identity, motivation, and practice (Cremin et al., 2026; Cremin & Scholes, 2024; Čepič et al., 2024).

The last several years have also broadened what counts as relevant evidence for this question. Recent work has emphasised reading identities, informal interaction around texts, and the tensions that arise when schools try to institutionalise pleasure (Cremin et al., 2026; Cremin & Scholes, 2024; Čepič et al., 2024). This recent literature reframes the question: the field does not simply need another reminder that students should read more, but a clearer account of which features of pleasurable reading can be translated into assigned factual reading, and which cannot.

The contrast with fiction is striking. Many readers eagerly learn intricate details about entirely invented worlds—like Middle Earth or Arrakis—or absorb dense factual domains when they are embedded in historical fiction, such as 12th-century cathedral architecture in Ken Follett’s The Pillars of the Earth (1989), Tudor politics in Hilary Mantel’s Wolf Hall (2009), or 19th-century naval mechanics in Patrick O’Brian’s Aubrey-Maturin series (1970). They remember characters, maps, conflicts, and obscure rules with an enthusiasm that educators would welcome in classroom learning. Yet this contrast is not simply a matter of “facts versus fiction”—reading for pleasure and reading for learning differ along several dimensions at once: fictional versus factual content, narrative versus expository structure, self-chosen versus assigned reading, voluntary engagement versus assessment pressure, and socially shared enthusiasm versus isolated study. Some of these differences are difficult to change. Assigned reading usually remains assigned, and factual reading must remain accountable to evidence and accuracy. Nonetheless, some features of fiction and pleasure reading are more mutable. These include narrative schema, mental imagery, prior knowledge, self-relevance, topic interest, and social reading.

In this Perspective paper, I address three questions: (1) why fictional worlds and fiction more generally can be so engaging; (2) which mechanisms support engagement, comprehension, and memory; and (3) how assigned factual readings can be made more engaging. The argument is selective and mechanism-informed rather than systematic—I do not attempt a meta-analysis or report new empirical data, and I use the literature to build a conceptual account rather than to adjudicate the full balance of evidence. The design principles I propose are therefore best understood as testable hypotheses, not as validated interventions. My goal is not to suggest that educational readings should become fiction, nor that pleasure is a substitute for rigour. Rather, building on the framework developed in my recent book Memories that Matter (Madan, 2024), my assessment is that factual readings can be designed so that facts are encountered within forms that help readers care, imagine, question, connect, and remember.

This framing also returns the question to cognition. If some texts are easier to enter, imagine, discuss, and remember than others, then the design of a reading is part of the cognitive situation in which reading occurs. Texts and reading contexts can guide attention, support situation models, recruit schema, invite mental imagery, connect new information to prior knowledge, and give readers reasons to elaborate meaning with others. An overview of these design differences is presented in Figure 1. In this sense, the pedagogical question is also a cognitive one: which features of factual reading help the relevant cognitive processes get started and keep going? This framing aligns with the founding vision of this journal, which was established to foster genuine dialogue among psychology, neuroscience, linguistics, education, and related disciplines rather than to confine cognitive inquiry within a single tradition (Duñabeitia, 2025).

Figure 1. Conceptual overview of six mutable design levers that can support engagement with assigned factual readings: narrative schema, mental imagery, prior knowledge, self-relevance, topic interest and choice, and social reading.

History comprises sequences of events, causes, conflicts, people, places, and consequences. Fictional worlds often contain similar elements, but readers encounter them as parts of a living world rather than as isolated facts. J. K. Rowling’s Harry Potter novels (1997, 2014), for instance, introduce readers to Hogwarts School of Witchcraft and Wizardry and then invite further curiosity about its houses, rules, teachers, past headmasters, rival schools, tournaments, and magical institutions. J. R. R. Tolkien’s Lord of the Rings and The Silmarillion build Middle Earth through languages, genealogies, songs, maps, histories, migrations, wars, and myths (Tolkien, 1955, 1977; Peterson et al., 2017)—a world later extended through encyclopedias, role-playing games, and campaign atlases (Day, 2019; Dials et al., 2018; Rabuck, 1992). The appeal is not only that these worlds are invented, but that their details cohere.

Worldbuilding scholars have argued that imaginary worlds can extend beyond any single narrative, becoming transnarrative, transmedial, and transauthorial structures that audiences continue to explore across books, films, games, reference works (such as encyclopedias of the Forgotten Realms), and discussion communities (Greenwood, 1990; Grubb & Greenwood, 1990; Wolf, 2012, 2020). This is why fictional lore can resemble a knowledge domain. A reader who learns about the politics of Arrakis, the 19th-century rigging of a frigate in O’Brian’s novels, or the medieval guilds in Follett’s Kingsbridge is not memorising a random list. Each fact sits within a network of places, conflicts, values, and expectations (Follett, 1989; Herbert, 1965; O’Brian, 1970). With each new detail, the world becomes easier to navigate and more rewarding to know. Indeed, mastering these worlds is not a passive escape; it requires “hard reading,” as learners must actively map complex, unfamiliar vocabularies and rules (Shippey, 2016).

This kind of self-directed expertise has empirical precedents. Readers of The Lord of the Rings, for instance, can form spatial cognitive maps of Middle Earth’s geography that rival the accuracy of participants who explicitly study its printed map (Louwerse & Benesh, 2012). In a study of Sherlock Holmes enthusiasts, members of a Holmes society were asked to identify sentences from the stories; sentences that were concrete and relevant to the theme of the story were the strongest cues, whereas names and descriptions were less useful (Neisser & Hupcey, 1974). Similarly, Star Trek knowledge has been shown to influence recollective experience for Star Trek text, suggesting that domain expertise affects how readers remember material in a familiar fictional world (Long & Prat, 2002). Most directly, Brown and Shi (2026) found that event memories drawn from the Harry Potter series were retrieved and organised in broadly similar ways to autobiographical memories, with both showing frequent direct retrieval and clustering among related events. Together, these findings help explain why fan memory is not mysterious. Enthusiasts return to material repeatedly, discuss it socially, and fit new details into a rich prior knowledge base. The educational question is whether assigned factual readings can be designed to support some of these same processes.

Fictional worlds illustrate a broader phenomenon: people often read fiction because the experience itself is rewarding. Nell’s (1988) classic account of reading for pleasure describes readers becoming absorbed, entranced, and temporarily removed from the immediate environment—an experience that resonates with Csikszentmihalyi’s (1990) concept of flow. This emotional transportation is often the active ingredient in fiction’s psychological effects; for instance, reading fiction increases empathy, but primarily when readers are highly transported into the narrative (Bal & Veltkamp, 2013). This engagement is not a single process; neurocognitive accounts of literary reading emphasise interacting processes such as situation model building, immersion, mental simulation, imagery, affect, and aesthetic appreciation (Jacobs & Willems, 2018). Burnett and Merchant (2018) similarly describe reading for pleasure as involving affective encounters and enchantment. Reading fiction can be restful, private, and restorative as well as intellectually active. In the Rest Test, reading was ranked as the most popular restful activity among respondents (Hammond & Lewis, 2016). The pleasure of reading may also be shaped by the material and multisensory context of books, including their look, feel, sound, smell, and associated nostalgia (Spence, 2020). In medical and professional contexts, reading for pleasure has also been discussed as a potential source of reflection, empathy, stress reduction, and professional development (du Sautoy, 2021). Thus, fiction is not merely a delivery vehicle for plots and facts. It can be a form of rest, emotional experience, identity work, social connection, and cognitive play.

Fiction is also engaging because it invites readers to track minds. Readers follow what characters know, believe, want, hide, misunderstand, and infer about one another. Zunshine (2006) argues that fiction draws heavily on theory of mind, the capacity to attribute mental states to others. This makes fiction socially and cognitively rich: a reader is not only observing events, but also modelling motives, perspectives, and nested beliefs. Memories of fiction can also function in ways that resemble autobiographical remembering. Although fictional events are known to be invented, people can recall them with vivid imagery and use them in directive, social, and self-related ways (Yang et al., 2022). Fiction can therefore matter even when detailed plot memory fades (Dixon & Bortolussi, 2012); readers may retain scenes, moods, character relationships, values, or ways of seeing.

Reading for pleasure is further supported by agency and social context. Recent work emphasises identity, choice, and informal interaction around texts (Cremin et al., 2026; Cremin & Scholes, 2024; Čepič et al., 2024). These observations matter because they broaden the question. Readers rarely approach fiction with the goal of extracting facts. They read to be transported, challenged, soothed, surprised, moved, and connected. Indeed, such “deep reading” of literature can serve as a vital practice for understanding the self and the world. As Waxler (2014) argues, immersing oneself in narrative fiction counters the fragmented, fast-paced nature of modern digital life by requiring readers to slow down and reflect. In an era increasingly dominated by short-form media and a desire for neatly capsuled, easily consumable information, this deliberate engagement helps individuals construct their identities, find meaning in a chaotic world, and confront their own vulnerabilities.

The educational question, then, is not whether factual readings can become fiction. Rather, the question is which parts of fiction’s appeal are sufficiently general—and sufficiently mutable—that they can be incorporated into assigned factual reading without sacrificing accuracy.

4. From Mechanism to Design: Making Factual Readings More Engaging -

The contrast between fiction and factual reading is partly a contrast between narrative and expository texts. A meta-analysis of more than 75 unique samples and over 33,000 participants found that narrative texts were more easily understood and better recalled than expository texts (Mar et al., 2021). Crucially, this narrative advantage proved remarkably robust: it held across both comprehension and memory measures, and it was not diminished by varying study designs or reader demographics. This indicates that narrative structure supports not only initial understanding but also later retrieval, regardless of the specific learning context. This does not mean that fiction is inherently superior to factual writing. Rather, it suggests that narrative organisation can provide durable cognitive support for learning.

One way to draw these strands together is to ask what kinds of support are often present when reading feels compelling, and which of those supports can be strengthened in factual reading without weakening accuracy, evidence, or disciplinary form. Translating cognitive mechanisms into concrete educational practices has proven fruitful in other domains of the learning sciences (e.g., Weinstein et al., 2018). Table 1 takes a similar approach, offering a working framework for reading design. Each row starts with a feature often associated with fiction or reading for pleasure, identifies a cognitive or affective mechanism that may contribute to engagement and memory, and then links that mechanism to a design principle for assigned factual reading. The table is not meant to imply that engagement can be engineered one feature at a time. The rows are therefore best read as testable claims: places where future work can ask whether a particular design change improves comprehension, interest, memory, or transfer.

Table 1. Working mechanism-to-design framework for applying selected supports associated with fiction and reading for pleasure to assigned factual reading. The evidence column summarises the literatures used to motivate each design principle and notes key limits; the rows are intended as testable hypotheses rather than validated interventions.

This combination matters because memorability in fiction and pleasure reading usually reflects several conditions working together: interest, prior knowledge, repeated encounter, imagery, social exchange, and a sense that details matter within a larger world. Different disciplines will also draw on these levers differently. A history reading might lean on causal conflict, perspective, and source debate; a biology reading might rely more heavily on spatial diagrams and process models; mathematics and statistics may benefit from problem histories, worked examples, cases, and interpretation of uncertainty. The point is not that every subject should become narrative in the same way, but that factual reading can be designed with greater attention to the cognitive and social conditions under which readers understand and remember. The subsections that follow unpack each lever in turn.

Narratives organise information through time, causality, goals, conflict, and resolution. Readers learn to expect these structures and use them to anticipate what matters (Boyd et al., 2020; Campbell, 2008; Goldman & Varnhagen, 1986; Kintsch & van Dijk, 1978; Thorndyke, 1977). Narrative is therefore not only a literary form, but also a mode of organising experience and explanation (Bruner, 1986, 1991). Work on narrative comprehension similarly emphasises that readers build situation models, generate inferences, and track causal, temporal, spatial, and motivational relations among events (Graesser et al., 1994; Zwaan & Radvansky, 1998). When information fits a familiar structure, it can be easier to integrate, encode, and retrieve (Bartlett, 1932; Bransford & Johnson, 1972; Dooling & Lachman, 1971; Mandler & Johnson, 1977; Smith & Swinney, 1992). Factual readings often lose this advantage when they present information as a sequence of definitions, dates, or disconnected findings. Yet factual material can often be organised narratively without being fictionalised: a scientific discovery can be framed around a problem and its resolution; a historical period can be framed around actors, constraints, and consequences; a mathematical idea can be framed around a puzzle that previous tools could not solve.

To apply this principle, factual readings should make relationships among facts visible. This can be done through causal chains, chronology, problem-solution structures, unresolved questions, conflicts, and human-centred arcs. Consider a hypothetical high-school student, Jamie, who studies across a mandatory curriculum—history, biology, mathematics—where most readings are assigned rather than chosen and the gap between compliance and genuine engagement is at its widest. Imagine Jamie reading about the French Revolution for a history class. A history reading assigned to Jamie need not merely list events. It can show how people with different goals acted under constraints, how decisions produced consequences, and how later events changed the meaning of earlier ones. Facts become easier to understand when readers can see how they fit together, why one point follows from another, and what role a detail plays in a larger account. The goal is not to fictionalise factual material, but to give readers a path through it.

Importantly, narrative thinking is itself educable—not by teaching students to invent stories around factual content, but by scaffolding the features that support narrative comprehension: goals, causal links, changes over time, constraints, and consequences. At the same time, narrative is not a universal solution. Some disciplinary content depends on formal notation, classification, spatial representation, proof, practice, or procedural precision. In those cases, narrative may provide a motivating frame or a reason for a problem to matter, but it should not replace the representational forms that the discipline requires.

Fiction often evokes scenes, characters, places, and events. This matters because mental imagery can enrich encoding and provide retrieval cues (Paivio, 1969, 1971, 1986, 2007; Richardson, 1980). Imagery can make abstract relations more concrete and memorable (Bower, 1970; Bugelski et al., 1968; Madan et al., 2010; Madan & Singhal, 2013; Straub & Granaas, 1992; Wollen & Lowry, 1971), and imagery-based techniques such as the keyword method have been shown to support vocabulary acquisition in educational settings (Atkinson, 1975). Imagery that is vivid or bizarre can be especially distinctive and therefore more memorable (Mercer, 1996; Toyota, 2002)—one reason that the fantastical scenes of fiction may leave such lasting impressions. For example, neuroimaging research reveals that reading stories activates highly distributed networks associated with visual and motor simulation, indicating that readers actively build the world in their minds (Wehbe et al., 2014).

Neurocognitive evidence suggests that the imagery evoked by fiction is not merely visual decoration, but part of broader multimodal situation-model construction. Accounts of literary engagement place imagery alongside situation model building, immersion, mental simulation, affect, and aesthetic appreciation (Jacobs & Willems, 2018). Consistent with this view, an fMRI study of story comprehension found that visually vivid, action-based, and emotionally charged passages selectively modulated visual-spatial, motor, and affective systems, suggesting that comprehenders construct multimodal representations of described situations (Chow et al., 2014). This embodied account also helps explain why the sensory context of reading can matter: the look, feel, sound, smell, and materiality of physical books may contribute to engagement and nostalgia, even though evidence for large format differences in comprehension and memory remains modest (Spence, 2020).

Factual texts sometimes underuse this resource, especially when they present abstractions without examples, diagrams, or situations in which those abstractions do work. Visual and concrete supports can help readers form mental models. Diagrams, maps, timelines, worked examples, annotated images, comics, and recurring characters can make abstract relations easier to inspect and remember. The key constraint is that images should do cognitive work. Decorative images may make a page more pleasant, but useful images clarify a relation, show a process, locate an event, or make an invisible structure easier to imagine. Consider Jamie reading about the circulatory system for a biology class: a labelled diagram showing blood flow through the heart gives her a spatial model she can revisit, whereas the same information presented as a paragraph of anatomical terms may remain abstract and hard to retrieve.

Several existing educational works illustrate how visual and narrative scaffolding can be applied in practice. The Manga Guides series apply characters, visual sequencing, concrete examples, and dialogue to topics including physiology (Tanaka & Koyama, 2015), molecular biology (Takemura & Sakura, 2009), physics (Nitta & Takatsu, 2009), regression (Takahashi & Inoue, 2016), and linear algebra (Takahashi & Inoue, 2012). Similarly, Andy Field’s An Adventure in Statistics (2022) uses a story-like frame, visual explanation, humour, and dialogue to make statistical concepts less inert and less intimidating. These examples matter because they challenge the idea that narrative, imagery, and character are only practical for history or literature. Even abstract and technical domains can be organised so that learners have a path, a reason to continue, and a more concrete sense of what the ideas are doing.

Readers remember more when new information connects to what they already know. Prior knowledge helps readers notice patterns, make inferences, and organise details into a meaningful structure (Bransford et al., 2000; Willingham, 2009, 2021). Fictional worlds can foster this gradually. A reader first learns a few characters and places, then learns how institutions, histories, and conflicts fit together. Eventually, new details become easier to remember because they have many possible hooks.

This gradual accretion of hooks is one reason fan expertise is so powerful. Sherlock Holmes enthusiasts and Star Trek experts are not merely remembering isolated sentences or names; they are drawing on rich networks of prior knowledge (Long & Prat, 2002; Neisser & Hupcey, 1974). Assigned factual readings can borrow this principle by orienting readers before asking them to absorb detail. Timelines, maps, concept overviews, glossaries, recurring cases, and advance organisers (Ausubel, 1963) can make new facts feel less like fragments and more like additions to an emerging knowledge world.

Information is better remembered when it is processed in relation to the self (Craik & Lockhart, 1972; Nairne et al., 2008; Rogers et al., 1977; Symons & Johnson, 1997). Autobiographical memory research reinforces this point: personally significant life events are remembered with greater vividness and detail (Benuzzi et al., 2018; Rubin et al., 2010; Wardell et al., 2021). Fiction often creates self-relevance indirectly. Readers may identify with characters, imagine alternative lives, test values, or care about a character’s choices. These processes need not depend on personal similarity—a reader may become engaged because a story raises a moral question, reveals a social world, or offers a perspective that is unfamiliar but compelling (Yang et al., 2022; Zunshine, 2006).

Factual readings can use similar routes to relevance by highlighting human stakes. Rather than asking students to care about a fact in the abstract, readings can show who the fact mattered to, what problem it addressed, what changed because of it, or why it matters for a learner’s future role, community, or decisions. Existing factual works demonstrate this principle. Howard Zinn’s A People’s History of the United States (2010), for instance, shows how historical facts can be organised around perspective, conflict, voice, human consequence, and moral dispute (Cohen & Murrow, 2021). This does not make the work neutral, and it does not remove the need for source evaluation. Rather, it illustrates how factual material can become more engaging when readers encounter facts as part of contested human experience. In this sense, self-relevance overlaps with identity. Reading-for-pleasure frameworks emphasise that engagement is supported when young people are recognised as readers with interests, histories, and cultural practices rather than as pupils merely completing tasks (Cremin et al., 2026). The reading remains assigned, but it can be designed to make space for the reader.

Pleasure reading is typically self-chosen, and assigned factual reading typically is not—a difference that no amount of clever design can eliminate. Nonetheless, autonomy is not all-or-nothing. Self-determination theory suggests that motivation is supported when people experience autonomy and competence (Deci & Ryan, 1985). Curiosity and interest can also enhance learning and memory (Fastrich et al., 2018; Gruber et al., 2014; Kidd & Hayden, 2015; Litman, 2005; Murayama et al., 2019), and the resulting memory gains can persist beyond the immediate learning episode (Stare et al., 2018). Fiction often creates curiosity through mystery, conflict, uncertainty, and delayed resolution. Readers continue because they want to know what happens next.

To build curiosity in assigned reading, factual texts should give students reasons to continue. This can involve puzzles, anomalies, competing explanations, unresolved questions, or consequential stakes. Rather than beginning only with definitions, a reading might begin with a phenomenon that demands explanation. Curiosity does not remove difficulty, but it can change the reader’s relation to difficulty: the challenge becomes part of pursuing an answer.

In parallel, educators can create bounded choice. Assigned readings can include choice without abandoning curricular aims. While language classes frequently allow students to choose their own reading material for an essay or critique, the sciences rarely provide such agency. Specialised curricula, such as the International Baccalaureate (IB) and gifted and talented programmes, do often structurally embed this agency by allowing students greater latitude in topic choice (Gentry & Springer, 2002; International Baccalaureate Organization, 2017). Yet even in highly structured domains and standard curricula, bounded choice is possible. Students might choose from a curated list of suggested readings, select which case study to follow, which question to answer, which example to extend, which source to compare, or which application to explore (Cremin et al., 2026; Deci & Ryan, 1985). If Jamie is assigned a module on the French Revolution for her history class, she cannot opt out of the topic. However, the educator might design the assignment so that after reading a shared introductory text, Jamie chooses from a curated list of supplementary readings. This allows her to decide whether to focus her subsequent analysis on economic causes, military strategy, or the experience of ordinary Parisians. These small decisions matter because they can shift the student’s role from passive recipient to active reader. In this respect, bounded choice is a practical compromise between voluntary pleasure reading and the realities of curriculum.

Pleasure reading is often private in the moment, but it is frequently social before and after. People recommend books, discuss characters, debate interpretations, quote favourite lines, join communities, and form identities around reading. The appeal of literature can even extend beyond the text itself into public events, such as literature festivals, where the shared cultural experience and collective engagement draw in attendees who might otherwise rarely read for pleasure. Recent frameworks for reading and writing for pleasure similarly emphasise informal interaction, adult role models, connected communities, and recognition of readers’ identities (Cremin et al., 2026). Reviews highlight this social dimension as well (Cremin & Scholes, 2024; Čepič et al., 2024).

Assigned factual reading can borrow from this social life. Book clubs thrive precisely because reading becomes a reason to meet, talk, disagree, and discover what others noticed. In classrooms, the same principle can take many forms: discussion questions, collaborative annotation, peer explanation, debates, and comparison of interpretations can help students see reading as participation in inquiry rather than private compliance. The social activity should extend the reading rather than merely police it (Vygotsky, 1978). For instance, a useful discussion might ask Jamie to compare interpretations with peers, test an explanation, connect the text to another case, or explain why an example worked. In that form, social reading can support both motivation and understanding.

Despite these opportunities for redesign, some limits deserve attention. Assigned factual reading cannot fully become reading for pleasure. It is assigned because there is material that a course or curriculum expects students to encounter. It must remain accountable to evidence, accuracy, source evaluation, and disciplinary standards. Assessment pressures cannot be removed entirely (Ravitch, 2010), and students may not arrive with the affection, identity investment, or community support that surrounds a favourite fictional world.

Another limit concerns intention. Pleasure reading often allows incidental learning: readers may acquire maps, histories, terms, and relations without setting out to study them. Assigned reading, by contrast, often foregrounds performance, monitoring, and anticipated assessment. Those conditions may change how readers allocate attention, how freely they imagine, and how much they experience the text as an object of curiosity rather than obligation. Evidence from fiction and narrative comprehension therefore cannot simply be transferred to narrative teaching without qualification. Whether a factual text designed with narrative features produces similar engagement, imagery, comprehension, or memory is an empirical question.

Nonetheless, narratives created without explicit educational purposes are not excluded. Historical fiction, narrative nonfiction, biography, documentary writing, and even wholly fictional worlds can support incidental learning when readers become invested in people, places, conflicts, and systems. However, their use in formal education raises additional questions about accuracy, source evaluation, representativeness, and transfer. A compelling narrative can make a domain easier to enter, but it can also foreground some perspectives while obscuring others. In assigned factual reading, then, narrative is best treated as a cognitive and motivational support—not as a substitute for disciplinary evidence.

Because of these inherent limits, the design principles above are modest but concrete. My point is not to pretend that assigned factual reading can become the same as curling up with a beloved novel. Nor do I suggest choosing between facts and skills, content and pedagogy, or rigour and engagement. Learning with understanding depends on organised knowledge, conceptual structure, and opportunities to use what has been learned (Bransford et al., 2000). The question is how factual readings can preserve rigour while offering more of the supports that make fiction and pleasure reading compelling.

If assessment rewards only short-term recall of isolated facts, students will understandably read for isolated recall. Assessments can instead ask students to explain relationships, apply ideas, compare cases, transfer concepts, or evaluate evidence. Low-stakes retrieval practice can still support memory, but it should be paired with tasks that reward understanding and connection (Bransford et al., 2000; Weinstein et al., 2018; Willingham, 2009, 2021). In this way, assessment can reinforce the kind of reading educators actually want.

Beyond merely guiding study habits, this alignment between task and test is fundamentally a matter of educational trust. If students are invited to read for curiosity, relevance, and understanding, but are then assessed only on disconnected details, the instructional message is contradictory. Assessment should therefore make clear that the valued outcome is not merely having passed one’s eyes over the page, but being able to meaningfully use what the reading made available.

The next step is empirical as well as pedagogical. The most informative studies would compare versions of assigned factual readings that preserve the same core content while varying specific design features—narrative structure, problem-first framing, visual scaffolding, bounded choice, social annotation, or assessment alignment. Creating well-matched materials of this kind is difficult, which is partly why such studies remain rare; but they are needed, because they could test not only immediate recall but also comprehension, delayed memory, interest, perceived relevance, willingness to continue reading, and transfer to new examples. Work of this kind would help move the field beyond the broad claim that engaging texts are better, toward a more precise account of which features matter, for whom, and under what instructional conditions.

In practical terms, the rows of Table 1 can serve as separable hypotheses for future research. One study might hold core content, length, vocabulary, and assessment constant while varying whether a reading is organised around a problem and causal chain. Another might preserve the text but add or remove a diagram that clarifies a spatial relation. Others could examine bounded choice, social annotation, assessment alignment, or the purpose with which readers approach the text. Critically, these comparisons should include existing instructional approaches—not only minimally redesigned readings—so that the design principles are tested against meaningful educational alternatives. Such studies should also distinguish engagement from learning. A reading might feel more engaging without improving durable understanding, or it might improve comprehension and transfer without being experienced as pleasure reading. Both outcomes would be informative.

This research agenda should also remain sensitive to access and identity. Recent reading-for-pleasure work makes clear that engagement is not only an individual cognitive state; it is shaped by available materials, reader identity, classroom norms, social interaction, and the legitimacy granted to different forms of reading (Cremin et al., 2026; Cremin & Scholes, 2024; Čepič et al., 2024). A useful future literature would therefore examine not just whether a redesigned reading improves memory on average, but whether it broadens participation, reduces unnecessary barriers, and helps more students see factual reading as a meaningful route into disciplinary knowledge.

This Perspective has argued that fiction’s appeal comes from many interacting features: narrative schema, mental imagery, prior knowledge, self-relevance, topic interest, and social reading. These features do not make fiction easy in a trivial sense; many fictional worlds are complex, demanding, and full of detail. Yet readers often choose to enter them because the details are meaningful within a world they care about.

Assigned factual readings cannot fully reproduce that experience. They remain tied to curriculum, evidence, and assessment. Nonetheless, they can borrow several mutable features from fiction and reading for pleasure. Facts can be embedded in coherent structures, connected to prior knowledge, linked to human stakes, opened through topic interest and choice, represented vividly, discussed socially, and assessed for meaningful understanding. Existing works in history, statistics, and other technical fields show that these principles are already viable across classroom domains. My goal is not to make history, science, mathematics, or statistics less rigorous, but to make factual knowledge more structured, meaningful, imageable, discussable, and worth pursuing. Engagement, in this sense, is not a decorative layer added after the cognitive work has been done—it is one of the ways that attention, memory, imagery, motivation, and social cognition enter the work of reading.