Decision_theory - Wikiparency

# Decision theory Decision theory models rational (and boundedly rational) choice under uncertainty by combining probabilities with utilities or other value measures. Normative expected-utility prescriptions, descriptive heuristics and biases, intertemporal discounting, and multi-agent extensions link it to economics, cognitive science, and AI. Strategic interaction is the province of [[Game_theory|game theory]].  ## Microsims (promoted from legacy — three.js first) ### Microsim (three.js) <div class="microsim-player"> <iframe src="https://wikitube-3d-microsims.netlify.app/Decision_theory.html" width="100%" height="620" frameborder="0" loading="lazy" sandbox="allow-scripts allow-same-origin"></iframe> </div> *Every saccade, sniff, or reach is a small wager settled in spikes: sensory neurons deliver noisy evidence that a downstream [[Neural_circuit]] integrates until a Motor control command commits the body to one option. This microsim opens that black box and shows how the brain trades speed against accuracy — the same accumulating signal that Neural decoding recovers from population activity.* > The drift-diffusion model (DDM) treats a fast decision as a noisy walk. Evidence favouring one option over another is summed moment by moment until the running total touches a boundary, and the brain commits to whichever wall it hits first. Strengthen the evidence and the walk drifts briskly to the correct side; add noise or move the walls inward and it wanders, sometimes crashing into the wrong one. Here you can launch many such walks at once and watch the choices and reaction times they generate. ### About this microsim The sketch animates the diffusion process directly. **Drift rate v** (−2 to 2) sets how strongly evidence pushes toward one boundary; **Noise σ** (0.2 to 2) scales the random moment-to-moment jitter; **Threshold a** (0.5 to 4) sets how far a walk must travel to commit; and **Starting bias z₀** (−1.5 to 1.5) offsets where every walk begins. **Racers N** (1 to 20) launches that many independent trials simultaneously so a distribution of outcomes forms before your eyes, while **Speed** (1 to 60 steps per second) changes only playback rate, not the underlying model. **Play**, **Step**, and **Reset** run, single-step, and restart the animation. ### Related microsims - [[Neural_circuit]] — the parietal–frontal loops that pool sensory evidence toward a bound - Neural decoding — reading the momentary decision variable out of population spike trains - Action potential — ramping firing rates are the spike-rate substrate of the accumulator - Motor control — the saccade or reach that reports the committed choice - Electroencephalography — the centro-parietal positivity indexes human evidence accumulation - Neuromodulation — arousal and reward signals adjust the boundary and urgency - [[Evolutionarily_stable_strategy]] — related GENOMICS microsim  ## Reveal %%REVEAL:d3%% %%REVEAL:mermaid%% --- *Concept aligned with [Wikipedia](https://en.wikipedia.org/wiki/Decision_theory); adapted text, where present, is licensed [CC BY-SA 4.0](https://creativecommons.org/licenses/by-sa/4.0/).* ## Overview The formal core weds [[Probability|probability]] and [[Statistics|statistics]] to preference through the [[Mathematical_model|mathematical model]] of the [[Rational_choice_model|rational choice model]], optimized with [[Mathematical_optimization|mathematical optimization]] from [[Operations_research|operations research]] — while [[Bounded_rationality|bounded rationality]] keeps the descriptive side honest about real [[Decision-making|decision-making]]. Dilemmas like the [[Prisoner's_dilemma|prisoner's dilemma]] mark the border where single-agent choice becomes strategy. The systems lineage claims it too: [[Cybernetics|cybernetic]] regulators choose actions under [[Control_theory|control-theoretic]] costs, [[Information_theory|information theory]] prices what an observation is worth, [[Artificial_intelligence|artificial intelligence]] operationalizes utility in agents, and thinkers from [[Norbert_Wiener|Wiener]] and [[W._Ross_Ashby|Ashby]] to [[Stafford_Beer|Stafford Beer]] treated every organization as a decision machine inside [[Economics|economics]] and beyond — dynamics included, via the [[Dynamical_system|dynamical system]] view of sequential choice.  *Built to the [[WT!Three_js_Microsim_Master_Class|three.js Master Class]].*  ## Wikipedia : Wikitube **Strict pair:** [Wikipedia](https://en.wikipedia.org/wiki/Decision_theory) : [Wikitube](https://en.wikitube.io/wiki/Decision_theory) ## Previous hub tags Hubs: `GENOMICS`, `Systems`. Portals: [[PORTAL_Systems]], [[PORTAL_Decision_theory]].