Prophantasia

Mental imagery research has always had to deal with large individual differences. Galton's nineteenth-century survey already found that some people reported vivid inner pictures while others reported faint or absent ones. The modern aphantasia literature gave that low-imagery end a name and later placed it beside hyperphantasia, the unusually vivid end of the same broad field (Galton, 1880; Zeman et al., 2015; Zeman, 2024). That history matters because prophantasia should not be introduced as a brand-new faculty. It is a proposed report category at the edge of a larger, measurable, still-disputed imagery spectrum.

Vividness is the familiar scale: how clear, detailed, colorful, or lifelike an image feels. Prophantasia asks an additional location question. Is the content described as an image in the head, a conceptual knowing, a closed-eye field event, an overlay on ordinary vision, an object seeming to occupy external space, or a presence-like experience with little visual detail? Recent work makes this distinction harder to ignore. Azañón and colleagues found that a measure of imagery externalism related to interference in a way generic vividness did not, while Schwarzkopf and colleagues report that vividness ratings sit within a broader family of internally generated visual experiences, including reports of "seeing" imagery and projecting it into the external world (Azañón et al., 2025; Schwarzkopf et al., 2026).

The category therefore needs several fields, not one score. A careful report should separate vividness, external location, source judgment, confidence, modality, controllability, voluntariness, practice history, expectation, and safety signals. A person can have vivid imagery that stays plainly internal. A person can describe an externalized image while still knowing it is self-generated. A person can see visual snow, phosphenes, afterimages, or hypnagogic imagery without making any prophantasia claim. These distinctions are not pedantic. They are what keep the topic researchable.

Imagery and perception share enough machinery that the boundary between them has to be managed, not assumed. The older reality-monitoring literature asked how people decide whether information came from perception, imagination, memory, or inference. Current perceptual reality-monitoring work asks a sharper version: when imagined and perceived signals overlap, what lets a person judge which source is responsible (Johnson and Raye, 1981; Dijkstra et al., 2022)? Dijkstra and Fleming's 2023 work is especially useful here because it treats reality judgment as related to subjective signal strength and a reality threshold, not as a simple label attached after the fact (Dijkstra and Fleming, 2023).

That does not mean externally located imagery is hallucination. It means that location, signal strength, and source judgment should be measured separately. A person may say an image appears "out there" while also treating it as an intentional image. Another person may report weak imagery but strong confidence about its source. A third may report vivid closed-eye patterns from flicker, sleep onset, or visual noise, where the right comparison is stimulus or state, not volitional projected imagery.

Suzuki's 2026 C x G x D proposal gives a compact way to keep these pieces apart. In that framework, C names the classifier-like extraction of features from input, G names internal generation, and D names the discriminator-like judgment of whether a representation belongs to the outside world or internal production. For Prophantasia, this is a comparison tool, not proof. It helps explain why DeepDream, psychedelic VR, flicker, visual snow, tulpamancy, and projected imagery should not be collapsed into one mechanism (Suzuki, 2026; Gershman, 2019).

Some visual experiences are easier to provoke than voluntary projected imagery. Darkness has eigengrau or eigenlicht; the eyes and visual system produce noise, afterimages, floaters, phosphenes, and other nonretinal or entoptic material; sleep onset brings hypnagogic imagery; uniform fields and flicker can produce vivid patterns. Philosophers and psychologists have argued over what, if anything, we visually experience with closed eyes, and visual noise remains a useful case because it sits between physiology, phenomenology, and report language (Schwitzgebel, 2007; Gert, 2019/2021; Schacter, 1976; Stickgold et al., 2000).

Ganzfeld, ganzflicker, and stroboscopic work make the safety boundary more explicit. These methods manipulate input. They can generate vivid pseudo-hallucinatory reports or geometric patterns, and they are useful comparators for imagery strength and state sensitivity, but they are not evidence that a person has learned prophantasia (Wackermann et al., 2008; Konigsmark et al., 2021; Hewitt et al., 2025; Amaya et al., 2025). The same caution applies to DeepDream and psychedelic VR: they can be useful altered-vision comparisons, not proof that prophantasia shares a mechanism with psychedelic visuals.

Visual snow syndrome and hallucinogen persisting perception disorder matter here for safety and differential boundaries. They involve persistent or recurrent visual symptoms that can be distressing and clinically relevant. They should not be folded into prophantasia, and prophantasia language should not be used to normalize distressing persistent visual phenomena (Puledda et al., 2018; Sampatakakis et al., 2022; Martinotti et al., 2018; Butler et al., 2026).

If the question is whether people can voluntarily influence visual experience at all, the strongest comparison is not a forum guide. It is constrained psychophysics: binocular rivalry, ambiguous figures, multistable perception, stereopsis, free fusion, and structure-from-motion. These tasks give the visual system a narrow problem and ask whether attention, intention, imagery, or practice can bias what is perceived (Meng and Tong, 2004; van Ee et al., 2005; Pearson et al., 2008).

That evidence is useful because it is modest. It says that attention and imagery can bias perception under specific conditions. It does not say people can learn to project arbitrary images into the world. Free-fusion and Magic Eye culture add a public practice lineage because they train an unusual way of using ordinary vision, but even there the target is constrained by the stimulus and task (Stinson, 2022). This is the right temperament for prophantasia: use strong, narrow paradigms as guardrails, and do not inflate them into open-ended ability claims.

Solid: imagery variation is real and measurable by several imperfect methods. Aphantasia and hyperphantasia are active research topics. Vividness is important, but it is not the only dimension. Source monitoring is a legitimate research problem. Narrow perceptual tasks can show constrained effects of attention, imagery, or intention on perception.

Partial: prophantasia, externalism, and projected imagery are now visible enough in academic and public vocabulary to deserve careful measurement. Recent work supports separating external location from generic vividness. The construct boundary is still young, and the available evidence does not establish durable training effects.

Context: tulpamancy, imposition, Fire Kasina, autogogia, Hawkeen/NavaChing, DreamViews, public self-tests, and practice guides explain where the vocabulary and tasks come from. They are sources for provenance and practice ecology, not proof.

Open: whether any practice reliably changes projected imagery; which baseline profiles predict change; how expectation, absorption, demand, stimulus input, and report language affect outcomes; which safety signals matter most; and how to retest without coaching people toward the desired answer.