features scientific methodology

Scientific methodology

Every trainer module in Kognitika targets a specific cognitive function. The table below describes what each test trains, the scientific principle behind it, the metrics it produces, and which test file validates the engine logic.

Cognitive functions glossary

Function	Definition
Attention	Sustained focus and distribution of visual or semantic control. Trained by Schulte tables, Stroop test, Alphabet table.
Working memory	Short-term retention and updating of information during a task. Trained by N-Back, Mental math, Topology memory.
Inhibitory control	Ability to suppress an automatic response and choose a more accurate action. Trained by Stroop test (color-word conflict).
Cognitive flexibility	Rapid switching between rules, contexts, and solution strategies. Trained by Stroop-Alphabet, Gorbov rules.
Metacognition	Noticing one's own thinking errors and adjusting strategy. Trained by Reframing, Hype filter, Reality check.
Stress regulation	Reducing physiological and information load without losing clarity. Trained by NeuroSilence, Noise reduction, Deep focus.
Critical thinking	Verifying claims, sources, causal links, and hidden assumptions before concluding. Trained by Scanner, Decryptor, Filter.
Logic	Working with rules, quantitative relations, inferences, and stable solution methods. Trained by Numerical analysis, Logical matrix, Situational judgment.
Sensorimotor coordination	Coordinating visual signal recognition with fast controlled motor response selection. Trained by Alphabet table, Stroop-Alphabet.
Spatial thinking	Holding shape, position, route, and object relationships in a mental map. Trained by Spatial concealment, Topology memory.

Trainer modules

Module	What it trains	Scientific basis	Key metrics	Engine test
Schulte tables	Concentration, visual search, peripheral vision	Visual scanning and distributed attention: the user learns to see more elements without unnecessary eye movements.	Total time, click stability, error count	src/tests/schulte-core.test.ts
Stroop test	Selective attention, inhibitory control	Classic executive control test: the brain must resolve conflict between automatic reading and controlled color recognition.	Accuracy, reaction time, conflict cost (difference between congruent and incongruent trials)	src/tests/stroop-core.test.ts
N-Back	Working memory, updating, distractor resistance	Continuous performance paradigm: the subject must hold and update a sequence in working memory, matching the current stimulus to the one N steps back.	Hit rate, false alarm rate, d-prime (sensitivity index)	src/tests/nback-core.test.ts
Mental math	Mental arithmetic speed, working memory, attention	Loads working memory and executive control: the brain must hold operands, execute operations, and record results repeatedly under time pressure.	Accuracy, time per problem, error distribution by operation type	src/tests/mental-math-engine.test.ts
Alphabet table	Sustained attention, letter recognition speed, sensorimotor switching	Combines visual recognition, response selection, and motor program switching — engages selective attention and sensorimotor coordination.	Total time, accuracy, average reaction time, error count	src/tests/alphabet-table-engine.test.ts
Stroop-Alphabet	Inhibitory control, cognitive flexibility, sensorimotor coordination	Two-step paradigm: first resolve color-word conflict (Stroop), then execute a motor rule (P/L/O) — adds controlled motor switching on top of automatic response inhibition.	Color errors, command errors, total accuracy, time, two-step average reaction	src/tests/stroop-alphabet-engine.test.ts
Schulte 90 (Gorbov)	Extended concentration, peripheral vision, visual search on 9x10 grid	Expanded Schulte protocol: 90 cells require wider visual field retention and sustained attention under color-switching rules.	Total time, accuracy, error count, rule comparison	src/tests/schulte90-core.test.ts
Spatial concealment	Spatial working memory, mental rotation, position tracking	The user must remember the locations of concealed objects and detect changes — engages spatial sketchpad in Baddeley's working memory model.	Accuracy, reaction time, location error distance	src/tests/spatial-core.test.ts
Topology memory	Graph memory, relational encoding	The user memorizes a graph structure (nodes and edges) and reproduces it — tests relational memory binding, distinct from item memory.	Node accuracy, edge accuracy, completion time	src/tests/topology-core.test.ts
Collision detector	Semantic filtering, conflict detection	The user identifies semantic conflicts in statements — trains the ability to detect logical inconsistencies and competing claims.	Accuracy, reaction time, false alarm rate	src/tests/collision-core.test.ts
Async dispatcher	Task orchestration, interruption management	The user manages multiple concurrent task streams, prioritizing and resuming after interruptions — models executive function in multitasking environments.	Throughput, accuracy under interruption, recovery time	src/tests/dispatcher-core.test.ts
Numerical analysis	Numerical reasoning, data interpretation, working memory	The user extracts quantitative information from charts/tables and performs calculations — trains numerical literacy and data-to-decision pipelines.	Accuracy, time per question, error type distribution	src/tests/numerical-core.test.ts
Logical matrix	Rule deduction, pattern completion	Matrix reasoning task (similar to Raven's Progressive Matrices): the user identifies the rule that governs a 3x3 grid and selects the missing element.	Accuracy, time per matrix, difficulty progression	src/tests/logical-core.test.ts
Situational judgment	Social cognition, context analysis	The user evaluates interpersonal scenarios and selects the most appropriate response — tests social reasoning and perspective-taking.	Response appropriateness score, reaction time	src/tests/situational-core.test.ts
Speed typing	Motor speed, visual-motor integration	Rapid transcription task: the user reproduces displayed text as quickly and accurately as possible — measures motor speed and visual-motor integration under time pressure.	Characters per minute, error rate, backspace count	src/tests/typing-core.test.ts
Language scanner	Pattern recognition, manipulation detection	The user identifies linguistic manipulation patterns in text — trains critical reading and rhetorical device recognition.	Accuracy, hit rate, false positive rate	src/tests/language-scanner-core.test.ts
Decryptor	Fact-emotion separation, source evaluation	The user distinguishes factual statements from emotionally charged claims — trains critical analysis and source verification.	Accuracy, reaction time, bias pattern	src/tests/decryptor-core.test.ts
Reality check	AI hallucination detection, verification skill	The user identifies fabricated data in AI-generated content — trains epistemic vigilance and cross-verification habits.	Detection rate, false alarm rate, confidence calibration	src/tests/reality-check-core.test.ts
Noise reduction	Cognitive noise filtering, signal detection	The user extracts target information from progressively noisy visual displays — trains selective attention and signal-noise discrimination.	Accuracy by noise level, reaction time, sensitivity (d-prime)	src/tests/noise-reduction.test.ts
NeuroSilence	Mental quieting, intrusion suppression	The user practices suppressing internal mental intrusions and maintaining focus — trains metacognitive control over inner speech and distracting thoughts.	Intrusion frequency, recovery time, subjective focus rating	Engine logic in src/hooks/useNeuroSilenceEngine.ts
Cognitive Trash Filter	Information quality assessment	The user evaluates information sources for credibility, bias, and relevance — trains information literacy and epistemic vigilance.	Classification accuracy, reaction time, calibration	Engine logic in src/hooks/useCognitiveTrashFilterEngine.ts
Hype filter	Signal-noise discrimination in persuasive text	The user identifies exaggerated or misleading claims in promotional content — trains resistance to persuasive manipulation.	Detection accuracy, false alarm rate	src/tests/praise-engine.test.ts
Reframing	Cognitive reappraisal, perspective shifting	The user finds constructive reinterpretations of negative events without denying the problem — trains cognitive flexibility and reappraisal skill.	Reappraisal quality score, realism assessment	Engine logic in src/hooks/useReframingEngine.ts
Rejection immunity	Social rejection resilience, recovery	The user practices responding to social rejection with constructive next steps instead of avoidance — trains emotion regulation and adaptive coping.	Response quality, recovery speed	Engine logic in src/hooks/useRejectionImmunityEngine.ts
Storytelling	Narrative structure, causal coherence	The user organizes information into a clear narrative with cause-effect relationships — trains structured thinking and explanatory communication.	Story coherence score, causal link density	Engine logic in src/hooks/useStorytellingEngine.ts
Deep focus	Sustained attention, distraction management	The user maintains focus on a single object while noticing distractions and calmly returning — trains sustained attention and metacognitive distraction awareness.	Focus stability, distraction recovery time	Engine logic in src/hooks/useDeepFocusEngine.ts
Luscher color test	Emotional state assessment, color psychology	Based on Max Luscher's color psychology: sequential color preference reveals current emotional state, stress level, and compensatory behaviors.	Color preference order, stress coefficient, autonomy score	src/tests/luscher-core.test.ts
Express knowledge hub	Combined mental arithmetic + Schulte 90	Cross-mode progress comparison: trains switching between arithmetic and visual search tasks within one practice session.	Best time, score, accuracy, session count per mode	src/tests/mental-math-engine.test.ts + src/tests/schulte90-core.test.ts

Data export and analysis

All session data can be exported in privacy-safe JSON format via api/analytics/export (see Data export). The export includes per-module metrics, trends, and normative comparisons — structured for direct ingestion by LLMs or statistical tools for personal cognitive analysis.