If we describe the processing as:
inputs -> (statements, algorithms) -> outputs
and consider that statements and algorithms can be certain and uncertain everyone.
There are exist only 2x2 types of processing.
- calculations (certain statements, certain algorithms)
- reactions (certain statements, uncertain algorithms)
- thinking (uncertain statements, certain algorithms)
- feeling (uncertain statements, uncertain algorithms)
| Type | Statements (Data/Inputs) | Algorithms (Rules/Logic) | Description |
| Calculations | Certain (C) | Certain (C) | Deterministic and Formal. The outcome is precisely known given the input and rules. Focus on precision and proof. |
| Reactions | Certain (C) | Uncertain (U) | Adaptive and Heuristic. Known input triggers a rule set that is still being learned, optimized, or is inherently probabilistic. Focus on rapid, learned responses. |
| Thinking | Uncertain (U) | Certain (C) | Inference and Logic. Firm, formal logic is applied to fuzzy, incomplete, or probabilistic inputs to deduce a likely conclusion. Focus on logical deduction under uncertainty. |
| Feeling | Uncertain (U) | Uncertain (U) | Intuitive and Interpretive. Both the input data and the rules/model used to process it are ambiguous, subjective, or constantly evolving. Focus on holistic interpretation and intuition. |
Calculations (Certain, Certain)
Core Concept: Execution, Exactness, Determinism.
Technical Analogy: Compilers, basic arithmetic, cryptographic hashing.
Example: A cashier machine adding up prices. The prices (statements) are certain; the addition algorithm is certain. The result is absolute.
Reactions (Certain, Uncertain)
Core Concept: Optimization, Immediate Response, Training. This implies a system that is quickly adapting its internal rules based on certain stimuli.
Technical Analogy: Supervised Machine Learning Training (fixed data, optimizing the algorithm), reflexes, stimulus-response mechanisms.
Example: A self-driving car seeing a certain red light (the statement/input). Its response algorithm is not a fixed rule ("stop"), but an uncertain, optimized set of rules (e.g., brake pressure curve, timing) learned from millions of miles of data. The rules are probabilistic, but the input is a known fact.
Thinking (Uncertain, Certain)
Core Concept: Deduction, Analysis, Structured Inference. The system holds firm to its logical framework while dealing with ambiguity.
Technical Analogy: Bayesian Inference, troubleshooting/diagnosis, formal logical reasoning.
Example: A doctor diagnosing a patient. The medical rules (the algorithms/knowledge base) are generally certain (e.g., "This pattern of symptoms means X"). However, the patient's symptoms (statements) are uncertain (e.g., "I have a cough... maybe a fever..."). The doctor uses certain logic on uncertain data to arrive at a probable answer.
Feeling (Uncertain, Uncertain)
Core Concept: Intuition, Interpretation, Subjectivity. This captures the essence of open-ended, non-formal decision-making where even the rules are fluid.
Technical Analogy: Unsupervised Learning, emotional intelligence, artistic judgment.
Example: Judging the quality of a piece of art or music. The input (the sensory data) is uncertain (highly subjective, ambiguous, context-dependent), and the rules for judging it (the algorithm of taste, emotional response, aesthetic theory) are also uncertain (personal, culture-bound, and constantly shifting).