A structured method for text-based personality analysis.
TypeInteractions uses a text-based personality analysis method that examines linguistic and discursive traces in free-form writing to reconstruct the dynamic that best explains the text as a whole.
It does not look for a single trait or a hidden answer in a few isolated words. It connects multiple signals, compares competing hypotheses, and retains the profile that best withstands analysis.
From a text to a profile hypothesis.
Several dimensions of the text are considered together.
Lexical
Vocabulary, semantic fields, repetition, contrast, preferred terms, and shifts between abstract and concrete language.
Syntactic
Sentence structure, logical connections, oppositions, reformulations, precision, nuance, and degrees of certainty.
Narrative
The organization of scenes, time, causes, consequences, shifts in perspective, tensions, and turning points.
Logical
The way the text explains, distinguishes, justifies, organizes, resolves, formalizes, or reaches conclusions.
Expressive
The way an experience, intention, value, action, or situation is given form within the text.
Relational
The actual role other people play in shaping, adjusting, stabilizing, or transmitting what is expressed.
These dimensions do not operate as a checklist of independent criteria. Their meaning depends on how they combine across the text as a whole. The method does not judge educational level or literary quality.
Eight processes, eight fundamental ways of processing reality.
The core process is the operation that appears to emerge most naturally and consistently within the analyzed text.
In its simplest form, the model describes operations between four broad states: subjective, objective, concrete, and abstract.
The subjective in relation to the abstract
"Experiencing personal felt states through ideas and possibilities."
The objective in relation to the abstract
"Experiencing shared reference points through ideas and possibilities."
The subjective in relation to the concrete
"Experiencing personal felt states through real-life situations."
The objective in relation to the concrete
"Experiencing shared reference points through real-life situations."
The concrete in relation to the objective
"Experiencing real-life situations through shared reference points."
The abstract in relation to the objective
"Experiencing ideas and possibilities through shared reference points."
The concrete in relation to the subjective
"Experiencing real-life situations through personal felt states."
The abstract in relation to the subjective
"Experiencing ideas and possibilities through personal felt states."
These operations provide an accessible representation of the eight processes. The internal method uses finer distinctions to avoid confusing the product of the operation with the source being used.
The role of relationships in the process.
Each process can manifest along a “relational” scale.
This dimension does not measure social ease, the number of people mentioned, or the desire to belong to a group.
- ✓ Low relational mode: The process appears to regulate itself through protection and preservation.
- ✓ High relational mode: The process appears to regulate itself through engagement and exposure.
The key question is not simply whether other people are present in the text, but whether they act as a genuine regulatory environment for the process.
The strongest result is the one that best withstands comparison.
The analysis does not begin by selecting a type. It first builds several possible hypotheses, then determines which one best explains the central linguistic traces.
The conclusion may remain cautious
When several processes remain close, when the text lacks sufficient material, or when the relational mode remains ambiguous, the result indicates this uncertainty.
The method favors a cautious or partially determined result over an artificially confident classification.
The quality of the result also depends on the text.
A text that is long enough, natural, and sufficiently developed provides more material for comparing the different hypotheses.
- ✓ A text that is too short may not provide enough material for a preferred dynamic to emerge.
- ✓ A highly technical or purely descriptive text may contain a great deal of information while offering few signals relevant to the method.
- ✓ A strongly imposed topic or heavily imitated writing style may make the result less representative.
The confidence score reflects the estimated strength of the reconstruction based on the submitted text.
The principle is public. The detection criteria remain internal.
TypeInteractions provides a brief but public overview of the general logic of its model, its eight major operations, and its relational mode.
The detailed rules used to distinguish between competing hypotheses are not published.
What remains undisclosed
The precise signals associated with each process, their weighting, elimination criteria, comparison rules between neighboring processes, and the scoring method remain protected.
An interpretive model, not a clinical assessment.
TypeInteractions offers a structured reading of free-form writing. Its result should be understood as a personality indication and a tool for reflection.
The method is not:
a medical or psychiatric diagnosis, a clinical psychological assessment, a complete measurement of personality, definitive proof of identity, or a tool for recruitment, selection, or professional decision-making.
Text-based personality analysis in six steps.
Observe linguistic and discursive traces.
Determine the role each element plays in the text.
Reconstruct several possible operations.
Compare the candidate processes.
Estimate relational-mode score and overall strength.
Propose a structured profile.
Is TypeInteractions a scientific test?
TypeInteractions does not yet have external academic validation. It is a typological model developed over several years, based on extensive work involving observation, comparison, and the analysis of real profiles, notably through hundreds of hours of interviews. Its distinguishing feature is that it was designed in a rigorously mathematical and symmetrical way, with no known equivalent to date.
Does the analysis rely on vocabulary?
Vocabulary is one of the dimensions considered, alongside syntax, narrative structure, reasoning, and the relationships between elements. No single word is enough to determine a result.
Do I have to write about myself?
No. The process may appear in a personal narrative, but also in the description of a character, a group, a situation, a relationship, or an action. Above all, the text should be natural and sufficiently developed.
Can two texts written by the same person produce different profiles?
In theory, no. When several sufficiently natural and developed texts are written by the same person, the model should converge toward the same profile. In practice, however, differences may appear depending on the length or richness of the text, and may sometimes change the resulting profile.
Is the MBTI correspondence exact?
Both models are based on a 16-type typology, and the correspondence shows strong coherence, but it remains indicative. It serves as a complementary reference point and does not replace an official MBTI result.
Try the method in the app.
Write a free-form text, run the analysis, and discover the core process that appears most consistently within your writing.