--- # MIF Ontology: psycholinguistics # # A research ontology for psycholinguistics and computational stylometry: # psychological and linguistic constructs, stylometric features, psychometric # indices, elicitation protocols, research instruments, linguistic frameworks, # and existing framework references. Supports voice-elicitation, personality- # language mapping, and authorship/readability research. # # Compatible with: engineering-base v0.1.0 + mif-generic v1.0.0 # Version: 0.1.0 # Last Updated: 2026-06-26 # # Sources: # - Big Five / OCEAN (Costa & McCrae, 1992) https://doi.org/10.1002/per.2410060304 # - LIWC (Pennebaker et al., 2015) https://www.liwc.app/ # - Flesch Readability (Flesch, 1948) https://doi.org/10.1086/386080 # - MATTR / MTLD (McCarthy & Jarvis, 2010) https://doi.org/10.3758/BRB.42.2.381 # - Stylo / Burrows's Delta (Burrows, 2002) https://doi.org/10.1093/llc/17.3.267 # - Cognitive Interview (Fisher & Geiselman, 1992) https://doi.org/10.1007/978-1-4612-4736-2 ontology: id: psycholinguistics version: "0.2.0" description: >- Psycholinguistics and computational stylometry ontology — constructs, features, indices, protocols, instruments, frameworks, and references for personality-language mapping, voice elicitation, and authorship research. schema_url: >- https://mif-spec.dev/schema/ontology/ontology.schema.json extends: - engineering-base - mif-generic # Namespace Hierarchy namespaces: semantic: children: constructs: description: "Psycholinguistic and psychological constructs" type_hint: semantic features: description: "Stylometric and linguistic features extracted from text" type_hint: semantic indices: description: "Psychometric indices and derived scores" type_hint: semantic instruments: description: "Research instruments, tools, and platforms" type_hint: semantic frameworks: description: "Linguistic and psycholinguistic frameworks" type_hint: semantic protocols: description: "Elicitation and interview protocols" type_hint: semantic # Entity Types entity_types: - name: psycholinguistic-construct description: "An abstract psycholinguistic or psychological construct measurable via language (e.g. Big Five trait, attachment style)" aliases: - "psycholinguistic trait" - "personality construct" - "cognitive-linguistic construct" - "psychological writing construct" exemplars: - "Style features fall on a spectrum from fully self-reportable, like punctuation preferences, to largely unconscious, like sentence complexity" - "A personality trait correlates with pronoun and emotion-word usage frequency across a large body of analyzed text" - "First-person singular pronoun usage is elevated in self-focused, lower-status, or genuinely emotional writing" base: semantic subtype_of: - concept traits: - cited schema: required: - name - construct_family properties: name: type: string description: "Construct name (e.g. Openness to Experience)" construct_family: type: string description: "Parent family (e.g. Big Five, HEXACO, attachment, need-for-cognition)" measurement_instrument: type: string description: "Primary instrument used to measure this construct (e.g. NEO-PI-R, BFI)" dimension_count: type: string description: "Number of facets or sub-dimensions" corpus_application: type: string description: "How this construct is operationalized in a text corpus" - name: stylometric-feature description: "A quantifiable textual feature used in authorship analysis or style profiling (e.g. type-token ratio, function-word frequency)" aliases: - "stylometric marker" - "authorship feature" - "writing-style feature" - "style fingerprint" - "linguistic style marker" exemplars: - "Function word frequency remains a robust, largely unconscious marker of authorship that stays stable across topics" - "The ratio of words used only once in a text indicates a writer's willingness to deploy unique vocabulary" - "Deep-learning models outperform traditional n-gram methods at identifying an author's fingerprint in short text samples" base: semantic subtype_of: - concept traits: - cited schema: required: - name - feature_class properties: name: type: string description: "Feature name (e.g. MATTR, mean sentence length, POS trigram)" feature_class: type: string description: "Class of feature (lexical, syntactic, character, semantic)" extraction_method: type: string description: "How the feature is computed (tool, algorithm, window size)" discriminating_power: type: string description: "Reported effectiveness for authorship discrimination" prior_art: type: string description: "Key studies or tools establishing this feature" - name: psychometric-index description: "A derived psychometric index or score computed over text or survey responses (e.g. Flesch Reading Ease, LIWC Analytic)" aliases: - "psychometric measure" - "style metric" - "readability index" - "lexical diversity metric" exemplars: - "Tree-based models using measurable stylistic features achieve high accuracy distinguishing human from AI-generated text" - "Two lexical-diversity metrics are considered psychometrically superior to a traditional type-token ratio for measuring vocabulary range" - "A set of syntactic-complexity indices measures sentence length, subordination, and coordination across four structural dimensions" base: semantic subtype_of: - delivery-metric traits: - cited schema: required: - metric_name properties: metric_name: type: string description: "Index name (e.g. Flesch-Kincaid Grade Level, LIWC Analytic)" index_type: type: string description: "Index category (readability, personality, affect, complexity)" formula: type: string description: "Computational formula or algorithm" normative_range: type: string description: "Expected range or population norms" tool: type: string description: "Software or platform that computes this index" - name: elicitation-protocol description: "A structured protocol for eliciting language samples or self-report responses for psycholinguistic analysis" aliases: - "interview elicitation protocol" - "survey elicitation method" - "questioning protocol" - "data-elicitation procedure" - "probing protocol" exemplars: - "An AI-conducted adaptive interview produces response data comparable in quality to a human-led interview, with longer average responses" - "A four-stage cognitive model of comprehension, retrieval, judgment, and response mapping explains how respondents process survey questions" - "Long survey instruments risk significant respondent dropout past a certain length, motivating modular section design and pacing strategies" base: semantic subtype_of: - design-pattern traits: - cited schema: required: - pattern_name - context - solution properties: pattern_name: type: string description: "Protocol name (e.g. Cognitive Interview, Think-Aloud, Voice Elicitation)" context: type: string description: "Research context and target constructs" solution: type: string description: "How the protocol structures the elicitation" elicitation_type: type: string description: "Modality (written, spoken, structured, semi-structured)" question_sequencing: type: string description: "Order and branching logic of prompts" respondent_burden: type: string description: "Estimated time and cognitive load for respondent" validity_notes: type: string description: "Known validity or reliability constraints" - name: research-instrument description: "A software tool, survey platform, or analysis package used in psycholinguistic research" aliases: - "research toolkit" - "analysis instrument" - "measurement tool" - "academic software tool" exemplars: - "An open-source academic package remains the dominant toolkit for computational stylistic analysis, supporting word and character n-gram extraction" - "A hybrid model combines a psycholinguistic category dictionary with transformer embeddings for interpretable style profiling" - "A latent-space interpolation approach conditions text generation on an arbitrary writing style from only a few samples" base: semantic subtype_of: - technology traits: - cited schema: required: - name properties: name: type: string description: "Instrument name (e.g. LIWC-22, stylo R package, APA PsycINFO)" instrument_type: type: string description: "Type (text-analysis-software, survey-platform, corpus-tool, statistical-package)" vendor_or_author: type: string description: "Developer, author, or institution" version: type: string description: "Version or edition" primary_use: type: string description: "Main analytical use case" licensing: type: string description: "License or access model (open-source, commercial, academic)" - name: linguistic-framework description: "A theoretical or computational framework grounding psycholinguistic research (e.g. LIWC dictionary, Universal Dependencies)" aliases: - "linguistic analysis framework" - "register analysis framework" - "stylistic framework" - "text-analysis framework" exemplars: - "A multidimensional register-analysis framework identifies functional dimensions of text variation through factor analysis of many linguistic features" - "A literature review of a forensic analysis method identifies persistent unresolved methodological gaps in evidence-strength standards" - "A survey of personality-detection methods finds transformer-based models now dominate over earlier bag-of-words approaches" base: semantic subtype_of: - concept traits: - cited schema: required: - name properties: name: type: string description: "Framework name (e.g. LIWC, Universal Dependencies, Systemic Functional Linguistics)" framework_type: type: string description: "Type (dictionary-based, dependency-grammar, discourse, register)" primary_author: type: string description: "Lead author(s) or originating institution" publication: type: string description: "Canonical publication or specification" harness_application: type: string description: "How this framework is applied in the research harness" - name: existing-framework-reference description: "A reference to an existing psycholinguistic or stylometric framework for comparative or reuse analysis" aliases: - "reference framework" - "prior-art reference" - "existing tool reference" - "comparable-system reference" - "benchmark framework reference" exemplars: - "A commercial creative-writing tool calibrates its voice model to only a couple of author writing samples" - "An enterprise brand-voice platform enforces granular style-guide rules for a large customer base learning brand voice at scale" - "An AI-text detection tool expanded from binary detection into word- and sentence-level authorship attribution" base: semantic subtype_of: - concept traits: - cited schema: required: - name properties: name: type: string description: "Referenced framework or tool name" framework_origin: type: string description: "Originating research group or project" capability: type: string description: "What the framework offers" limitation: type: string description: "Known limitations or scope boundaries" comparison_basis: type: string description: "Basis for comparing with or adopting from this framework" # Traits traits: cited: description: "Requires citations/sources" requires: - citations # Relationships (RO-grounded, mint) relationships: measures: description: "A research instrument measures a psycholinguistic construct, stylometric feature, or psychometric index (RO:0000056 participates-in)" from: - research-instrument to: - psycholinguistic-construct - stylometric-feature - psychometric-index symmetric: false operationalizes: description: "An elicitation protocol operationalizes a psycholinguistic construct, making it empirically observable (RO:0002234 has-output)" from: - elicitation-protocol to: - psycholinguistic-construct symmetric: false grounds: description: "A linguistic framework grounds a psycholinguistic construct, providing theoretical or categorical basis (IAO:0000136 is-about)" from: - linguistic-framework to: - psycholinguistic-construct symmetric: false # Discovery Configuration discovery: enabled: true confidence_threshold: 0.8 patterns: - content_pattern: "\\b(Big Five|OCEAN|openness|conscientiousness|extraversion|agreeableness|neuroticism|HEXACO|LIWC)\\b" suggest_entity: psycholinguistic-construct suggest_namespace: _semantic/constructs - content_pattern: "\\b(TTR|MATTR|MTLD|Flesch|Kincaid|readability|type.token ratio|lexical density)\\b" suggest_entity: psychometric-index suggest_namespace: _semantic/indices - content_pattern: "\\b(Stylo|Burrows.?s Delta|stylo|Delta method|stylometry|authorship attribution)\\b" suggest_entity: research-instrument suggest_namespace: _semantic/instruments - content_pattern: "\\b(elicitation|cognitive interview|think.aloud|voice interview|structured elicitation)\\b" suggest_entity: elicitation-protocol suggest_namespace: _semantic/protocols