The first version of Mindscape A tool for shaping your inner workspace; the project this entire glossary refers to.
The problem wasn’t the questions. The problem was the format. Numerical scales don’t measure internal state — they measure how people respond to numerical scales. When asked to rate emotional wellness from 1 to 10, most people land on 7. Not because 7 is accurate, but because of how scales work cognitively. Norbert Schwarz documented how the numeric range and labels of a scale alter what respondents think the scale is asking — the same verbal label means something different sitting above a 0 than it does above a −5. Krosnick identified the complementary problem: when questions demand cognitive effort, respondents satisfice rather than deliberate, latching onto the first answer that doesn’t feel wrong rather than the one that feels most right. The number you get back tells you about the scale. It doesn’t tell you much about the person.
Two research threads pointed toward a different approach. The first was projective assessment theory — Frank’s 1939 principle that when people impose structure on ambiguous stimuli, that structure reflects their internal state. The second was embodied cognition, specifically the work on conceptual metaphor showing that abstract emotional states are grounded in physical sensation. Heavy. Light. Frictionless. Stuck. These aren’t figures of speech. They’re how emotional experience is actually stored and recalled.
From those two frameworks came a design question: what if onboarding didn’t ask for a self-report at all? What if it created conditions for the user to act — and read the action instead?
The result is the Subconscious Interface A UI that mirrors the user’s pre-articulate intent — listens before it speaks, surfaces what’s almost-thought.
0. Goal Input
Before any mechanics run, the user is asked one direct question via a plain text field: what are you here to work on? The answer can be a sentence or a few words — burnout, anxiety, a creative block, a relationship, a transition. There’s no structure imposed on the response.
This input does two things. First, it gives the companion a working context before any behavioral data exists — a starting point for generating the first version of the regenerative portal and the initial quest framing. Second, it seeds the Narrative Swipe card pool. The cards shown to a user who writes “grief” are drawn from a different subset than the cards shown to someone who writes “career direction.” The scenarios are still abstract and surrealist, but they’re selected for thematic resonance with the stated context. The mechanics that follow aren’t generic — they’re oriented from the start.
1. Narrative Swipe
The Narrative Swipe presents a deck of cards, each depicting an abstract or surrealist scenario — a figure at the edge of something, a door in an unexpected place, a path that forks without reason. The prompt is minimal: you are in a dream. Do you go toward this, or away from it? Swipe right to approach. Swipe left to retreat.
The mechanic works because it bypasses deliberation. A questionnaire gives users time to consider what their answer implies about them. A card that requires a swipe decision doesn’t. The pattern across cards is where the data lives — not in any single response, but in what a user consistently approaches, what they consistently avoid, and where the contradictions appear. Each card in the sequence is drawn from a designed pool; which card appears next depends on the previous answer. The deck adapts as it goes, following the behavioral thread the user is laying down.
No single swipe is interpretable in isolation. The meaning is in the intersections — approach/avoidance motivation, coping style under ambiguity, threat sensitivity — surfaced through instinct rather than introspection.
2. Inertia Scroll
The conceptual foundation here comes from cognitive linguistics: depression is a burden. Sadness is heavy. These aren’t figures of speech — they reflect cross-modal mappings in how the brain encodes emotional experience. The physics of how something moves can feel emotionally calibrated or wrong in a way that’s immediate and pre-verbal.
The Inertia Scroll gives the user a scrollable area and a slider that controls its friction. At one end, the scroll is slippery and fast. At the other, it’s thick and resistant. The instruction: tune the scroll until it matches your energy today.
The friction setting the user chooses gives the system a reading of psychomotor state that scale-based self-report wouldn’t reliably surface. The preference for resistance or lightness corresponds to how people experience effort in their daily movement through the world. The mechanic doesn’t diagnose. It registers.
3. Entropy Calibration
The third mechanic presents a field of dots and a single slider. The slider controls the overall behavior of the system — moving it from one end to the other transitions the dots from a completely ordered grid to a fully chaotic scatter. The user moves the slider until the state of the field matches how their life feels right now.
The user has no direct control over individual dots. The slider determines the system state, and the dots respond accordingly — combinations of order and chaos emerge from where the slider sits, not from the user arranging things manually. This is intentional: the reading comes from two data points. The first is the position on the slider — a ratio from 0% to 100% that indicates where the user sits on the chaos-to-order spectrum. The second is the distribution: at any given slider position, some particles are ordered and some are in motion, and the ratio between them carries information about how much of the user’s life they perceive as stable versus unstable. These two values, combined with the data from the other mechanics, contribute to the composite profile.
4. Pareidolia Slider
The Pareidolia Slider addresses a specific design problem: Mindscape’s companion needs to know what level of abstraction each user can work with. Not everyone processes symbolic content the same way. Some find it immediately legible; others find it alienating. Asking directly — “how comfortable are you with abstract themes?” — produces answers shaped by how users want to present themselves, which is not the same as what they can actually engage with.
The mechanic works differently. The user sees a fluid ink-blot field and a slider that controls its coherence. At one end: high-entropy noise, unresolvable. At the other: a recognizable form. The instruction: adjust the image until you’re just at the edge of seeing something in it.
Where the user stops is the data. Stopping early, at high entropy, indicates comfort with ambiguity. Moving immediately to maximum coherence indicates a preference for clear signal and low tolerance for uncertainty. Oscillating before settling shows something else again. The slider doesn’t ask about abstraction tolerance — it measures it by making the user demonstrate theirs.
5. Mood Gradient
The Mood Gradient closes the onboarding. Two glowing orbs — the user moves them, overlaps them, blends them. A hue slider shifts their colors. The prompt: adjust the orbs until they match how you’ve been feeling lately. When it looks right, they press a button that doesn’t say “Finish.” It says: This feels right.
Color bypasses the problems of language. Describing an emotional state in words requires vocabulary, introspective clarity, and the willingness to commit to a label. Color doesn’t. It’s immediate and non-committal in a way that a written answer isn’t.
The challenge this introduced was interpretation — how to assign consistent meaning to a color a user chose, across users with different histories and associations. General color theory offers a starting point, but it’s too blunt to carry the weight. The solution was not to interpret on day one. The Mood Gradient becomes a daily practice: every morning and evening, the user sets their gradient. Over time, the companion builds a record of what this user’s colors mean for this user — not based on general theory, but based on their longitudinal pattern. This also resolved the accessibility concern: because meaning is constructed over time rather than assumed at the start, no particular color association is required from the user.
What the Mechanics Do Together
Each mechanic targets a different axis: coping orientation, psychomotor energy, order perception, abstraction tolerance, affective state. Individually, none of them is conclusive. Together, and cross-referenced against each other, they build a starting profile that the companion uses to calibrate everything downstream — the tone of its language, the framing of quests, the visual character of the regenerative portal, how directly or symbolically it communicates.
The profile is a first read, not a fixed one. The companion updates it continuously through daily use — mood gradients, journaling, quest completion, conversation. The onboarding is where the model starts. Everything that follows is how it learns.