Shuichiro Ogawa
日本語

Notes · updated 2026-09-22

What Do Science Fiction Prototyping and Design Fiction Actually Do?

This note reads 81 academic works on the "effects" of science fiction prototyping (Johnson 2011) and design fiction (Bleecker 2009), sorted into four settings: the route by which technology shown in films moves the real world, organizational foresight, HCI research tools, and education.

Contents (7)
  1. Two Names and Their Sources
  2. What Is Reported in Organizational Foresight
  3. What Has Been Established as an HCI Research Tool
  4. What Has Been Measured in Education
  5. Criticism Directed at Claims of Effect
  6. What Has Not Yet Been Measured
  7. Footnotes

In corporate training and university courses, people run workshops where participants write short science fiction to think about future products and societies. The participants seem to enjoy it, and there is a sense that ideas came up that would not have come up otherwise. Surely it had an effect.

But ask what that effect was measured by, and the answer stalls. Imagination? Interest in the future? Ethical sensitivity? What counts as an effect has not been settled in the first place.

This note gathers the academic literature on the effects of science fiction prototyping and design fiction, and sorts out what is claimed, what has been established, and what has not yet been measured. The search ran across OpenAlex, Crossref, Semantic Scholar, J-STAGE, and other sources, and 81 works remained in the ledger after screening.

Two Names and Their Sources

Science fiction prototyping (SFP) is the method Johnson named in a 2011 book. As a method that uses the writing of science fiction to make people consider a technology within its social context rather than in isolation, it has been used in security education (Kohno and Johnson) and in organizational foresight.1

Design fiction is the term Bleecker popularized in a 2009 essay. Bleecker opens by writing that “Design Fiction is making things that tell stories,” positioning it as a practice that brings certain matters of concern into focus by making things that tell stories.

Behind both lies the diegetic prototype described by the film scholar Kirby (a prototype depicted as actually working inside a story world). Kirby argued that depictions of technology in films show audiences a technology’s need, viability, and benevolence, and can lead to real-world funding opportunities. The book that popularized the term speculative design is Dunne and Raby’s Speculative Everything (2013).

Kirby’s argument is not about effects on workshop participants; it concerns a route by which fiction pushes technological development forward within society. Much of the literature discussed below calls both of these “effects” without distinguishing them.

What Is Reported in Organizational Foresight

In corporate and policy foresight, science fiction prototyping has been used as a tool for building scenarios.

Merrie et al. used science fiction prototyping to create four radical ocean-future scenarios and claimed that it complements existing scenario methods and reaches non-experts. Stahl et al. presented a science fiction prototype about a care robot as an example of responsible research and innovation (RRI) practice, one that surfaces ethical issues in advance. Bennett and Vijaygopal built a prototype of mobility technology from an expert workshop, presented it to people with ambulatory disabilities, measured which features they valued with conjoint analysis, and divided them into three groups.

In a Japanese corporate setting, Monoi et al. brought a science fiction-based live-action role-play (LARP, an experiential method in which participants act out roles) into an automaker’s foresight activities. Monoi et al. report that participants explored complex relationships and gained a comprehensive understanding of future scenarios, and state that follow-up interviews suggested lasting effects.

In 2026, Schimpf et al. identified 17 science fiction-related methods from roughly 60,000 publications and mapped where each can be used across a six-stage innovation process. A map of the methods is taking shape.

Still, every study listed here either describes a case in which a method was used or organizes where the methods fit. No study in this corpus compared organizations that used science fiction prototyping with organizations that did not to see whether decisions or outcomes changed. Wu’s (2013) “Imagination Workshops,” whose title announces an “empirical exploration,” could not be checked for what it measured, because its abstract could not be retrieved.

What Has Been Established as an HCI Research Tool

In HCI and design research, design fiction is used as a tool for getting people to talk about technologies that do not exist yet. This is the area where reports of effects are relatively concrete.

Noortman et al. placed a design fiction prototype about future dementia-care technology in the homes of eight participants for three weeks. They also held sessions with six HCI experts and reported that the method drew out empathic and rich discussion about the consequences of future technology. Elsden et al. presented, through three cases, speculative enactments, which involve participants in circumstances that are speculative but carry real consequences. Dillahunt et al. held a five-week workshop series with Black and Latino working-class residents of Detroit and, by centering Afrofuturism (a cultural movement that imagines futures from the perspective of the African diaspora), drew out alternative economic visions such as community capitalism and childcare collectives.

What these studies call an effect is not a change in participants but the quality of the accounts and values that researchers obtain. Design fiction is used as a tool for getting participants to voice their concerns and hopes about technologies that do not yet exist.

Researchers in the same area have also pointed out that how to evaluate this is not settled. Baumer, Blythe, and Tanenbaum describe confusion over the criteria for evaluating design fiction and show that several evaluative traditions sit side by side within HCI: critical design, narratology and literary theory, studio critique, user studies, scenarios and personas, and thought experiments. Which tradition one adopts can change the criteria for what counts as a good work. Kozubaev et al. organized what design futuring researchers should reflect on into five modes: formgiving, temporality, researcher positionality, real-world engagement, and knowledge production. Barendregt and Vaage argued that understanding speculative design as a thought experiment makes it easier to achieve the goal of provoking public discussion.

What Has Been Measured in Education

The educational literature is large; more than 20 studies in this corpus are set in education. Most of them, however, are reports of course design or qualitative case studies.

Only Nicholes (2020) reported a change in learning with numbers. Nicholes had 83 college students (35 English majors and 48 STEM majors) write short science fiction prototypes and compared their perceptions of writing before and after. Before the intervention there were differences between majors (for example, English majors rated creative writing as more relevant to their future careers, p = .04, r = .23), and after the intervention those differences disappeared. Among STEM majors, ratings of creative writing as relevant to their majors and future careers rose after the intervention (p = .015, r = .35). English majors’ ratings did not change. There was no control group, and what was measured was perception of writing, not the ability to think about the future or ethical judgment.

Qualitative studies shed light on the content of the effects from other angles. Andersson and Brodin report that in an elective engineering-ethics course that used fiction, students combined perspectives from the ethics of individual judgment and the ethics of social institutions, but their reflections rarely connected to their own future professional roles. Bendor and Lupetti state that a graduate course in speculative design draws attention to sociotechnical dilemmas and prompts discussion of the moral and political implications of innovation, while discussing barriers to adoption specific to design and engineering programs. Petersen and Gundersen drew three themes from fiction-based practice in higher education: challenging conventional norms in education, student frustration accompanied by signs of transformation, and learning that unfolds beyond the classroom.

Rahm’s report questions the direction of the effect itself. Rahm ran three design fiction workshops with 36 teacher-education students and analyzed how, in a setting meant to make them think about critical futures, the students produced an “ironic imitation” that traced existing, dominant views of technology. Rahm does not treat this failure as wasted, but as a chance to make visible where imagination is constrained. Set beside Andersson and Brodin’s findings, it shows that one cannot say a critical perspective grows automatically once fiction is used.

In Japanese-language research, Mitani et al. proposed an “open” science fiction prototyping method that can be carried out without creative-writing expertise, and Minegishi and Osawa built an agent that supports idea generation in science fiction prototyping and compared it with a human facilitator. Minegishi and Osawa state that the agent evened out the variation in the number of ideas across participants and showed a possibility of increasing the number of characters created. Neither abstract states the number of participants or any statistical test, so the size of the effects could not be confirmed.

Criticism Directed at Claims of Effect

There is also criticism of the premises on which effects are discussed.

Coulton, Lindley, and Akmal examined design fiction works in three groups: those clearly recognizable as fiction, those identifiable as design, and those whose status is ambiguous. They argued that the more familiar viewers are with an existing format, the more easily they mistake a work for the real thing, and that the pursuit of plausibility can tilt designers toward deliberately deceptive strategies. The power to draw out discussion and the risk of misleading viewers both come from the same plausibility.

Szollosy described Johnson’s framework for science fiction prototyping as “resolutely and obsoletely humanist” and proposed reconstructing it as a posthumanist method that treats cultural imaginaries self-reflexively. The criticism is that using imagination as a tool of technological development itself narrows which futures can be imagined.

What Has Not Yet Been Measured

Laid side by side, this literature shows that the word “effect” refers to at least four different things. Fiction pushing technological development forward within society (Kirby); organizations building scenarios and surfacing issues (Merrie, Stahl, Monoi); researchers drawing accounts and values out of participants (Noortman, Dillahunt); and changes in learners’ perceptions and reflections (Nicholes, Andersson and Brodin). Each calls for a different method of confirmation and a different standard of evidence.

Of these, the evidence is relatively consistent for the third: drawing out discussion and values. Even so, that evidence consists of researchers describing the quality of the accounts they obtained, not results compared against a condition in which the method was not used.

The gaps can be summarized in four points.

  • No comparisons: no study in this corpus compared a group using science fiction prototyping or design fiction with a group that used neither or used a different scenario method. The education search and the verification step looked for pre/post comparisons and control groups, and the only quantitative report was Nicholes.
  • Inconsistent definitions of effect: several evaluative traditions coexist (Baumer et al.), and which one is used differs from study to study.
  • No test of persistence: Monoi et al. suggest lasting effects from interviews, and Wuyts and Vandamme name the assessment of long-term impact as future work. No follow-up study was found.
  • Combination with generative AI: studies that co-create design fiction with generative AI began to appear in 2025 and 2026. Sipos (2025) combined generative AI with participatory scenario-building in a workshop with urban designers. Lehnert et al. (2026) co-created 18 design fictions about education with GPT-4 using six prompting approaches, and reported that the quality of the narratives emerged not from the model alone but from the interplay of human framing, prompting strategies, and model behavior. Both analyzed the narratives or the method itself; neither measured effects on participants (at the abstract level). Mishra et al. (2026), by contrast, is not about co-creation with generative AI: it reflects, through the writers’ own collaborative autoethnography, on a speculative fiction workshop series about AI ethics.

Going back to the opening workshop: the participants’ enjoyment and the ideas that would not otherwise have come up overlap with the “drawing out discussion” effect that researchers have reported. Whether that sense of achievement reached changes in participants’ thinking or in organizational decisions is something this body of literature cannot yet say. Who will measure, and on what scale, what was different compared with not running the workshop remains open.

References

All accessed 2026-09-22.

Footnotes

  1. The wording of the definition in Johnson’s book itself could not be confirmed from the publisher’s description or an abstract. The description here is based on the abstracts of other works that discuss SFP (Kohno and Johnson 2011; Szollosy 2024). ↩

Author: Shuichiro Ogawa (Design Researcher / Consultant) About me →