flybrainwork
Mining

FBW-P03

Two-Choice Decision-Making

Study how competing inputs produce a choice, and how stimulus strength, timing, and noise shape the outcome.

Training preparationApplications open

Apply to join training

Statistics temporarily unavailable

Project background

Create a small, interpretable task with two competing inputs and a declared action output. Progress from strength comparisons to attraction–repulsion conflict. Predeclared scoring and circuit interventions distinguish task-specific choice from broad claims about intelligence.

This project connects a defined neural-computation question to a reproducible evaluation protocol and a community participation pathway. Results will be interpreted only within their recorded model, task and evidence scope.

Training goals

  • Increase consistency with the task rule.
  • Reduce unnecessary switching between alternatives.
  • Measure reaction delay, noise sensitivity and intervention effects.

Current foundation and milestones

A/B response comparisons provide a reusable measurement foundation. Behavioral scores require a frozen two-input rule and an action readout, followed by actual evaluation.

Next milestone: Freeze two-input task rules and the action readout

  1. PlannedFreeze two-input task rules and the action readout
  2. PlannedImplement the versioned task and measurement interface
  3. PlannedEvaluate frozen conditions and independent reproduction

Content updated:

Technical approach

Freeze the two-input conditions, state duration and output readout. Search stimulus combinations, temporal windows and small readout parameter sets. Compare equal conditions across seeds and interventions. A stimulation selectivity score is not itself a behavioral decision score.

Method schematic

Competingsensory inputsNeuralresponse comparisonFixedchoice readoutPredeclaredrule scoring
A proposed project interface. Arrows describe information flow, not measured neural connections or a running training batch.

Task and evaluation design

Evaluation design

Strength comparisonCondition 1Conflicting cuesCondition 2Sequential stimuliCondition 3
Planned conditions for the project benchmark. These illustrations are protocol designs, not recorded performance.
IndicatorDefinitionDirectionUnitsObservation
Rule consistencyFraction of choices matching the predeclared task rule.Higher is better%Awaiting evaluation
Decision delayModel time between stimulus onset and a valid output choice.Lower is bettermodel timeAwaiting evaluation
Repeated-condition stabilityChoice consistency under repeated, matched conditions.Higher is better%Awaiting evaluation
Intervention differenceReport choice-distribution changes under matched interventions.Compare matched conditionsdistribution differenceAwaiting evaluation

No measured values have been published. Acceptance thresholds will be frozen in each task protocol before evaluation.

Expected deliverables

  • Versioned choice configuration packages.
  • Choice distributions with input-condition manifests.
  • Circuit-intervention control and reproduction materials.

Ways to participate

  • Parameter search
  • Input-condition design
  • Controlled experiments
  • Independent reproduction

Device requirements and task schedules will be matched by the community after runnable profiles have been evaluated. Unknown specifications are welcome in the application.

Apply to join training

Applications are received for community review and matching. Selection does not mean training has started.

Participation roles

Select at least one role.

Optional device information

Unknown specifications are welcome. These are self-reported intentions, not measured performance.

Signing in associates this application with your account. Applying does not start training or authorize payments.

View my applications ↗

References and related work

External sources provide technical background. They do not establish equivalent FBW capabilities.

View published research records ↗