Protocol

IBL visual decision task

Head-fixed mice report the side of a visual grating stimulus by turning a wheel to bring the stimulus to the center. Contrast and blockwise prior probability manipulate sensory and value-related evidence.

Task structure

Stimulus, choice, feedback, and training as declared in the protocol record.

Stimulus

evidence_schedule
Grating appears on the left or right at variable contrast; sessions can include blockwise changes in prior probability.
evidence_type
mixed
modalities
visual
notes
This record deliberately separates stimulus contrast from prior probability because both can influence choice.
units
percent contrast
variables
contrast, stimulus side, block prior probability

Choice

action_mapping
Wheel turns move the visual stimulus toward the center; the rewarded turn direction depends on stimulus side.
alternatives
left, right
choice_type
2afc
notes
response_modalities
wheel

Feedback

feedback_type
mixed
notes
penalty
Noise burst and longer inter-trial interval for incorrect decisions.
reward
Sweetened water for correct decisions.

Training

notes
Standardized IBL training stages should be captured in a deeper version.
stages
shaping, visual discrimination, biased blocks

Trial timing

  1. quiescence
    variable, approximately 0.4-0.7 s in the standardized task

    Mouse holds the wheel still before trial start.

  2. go_cue
    brief auditory event

    Auditory cue announces the response period.

  3. stimulus_and_response
    response-limited

    Visual grating is displayed and coupled to wheel movement.

  4. feedback
    outcome-dependent

Vertical slices

Expected analyses

  • psychometric curve by contrast and prior block
  • chronometric curve by contrast
  • lapse, bias, and threshold estimates
  • history-dependent choice model

Interpretive claims

  • perceptual decision-making
    Confidence: high · ref.ibl-2021-standardized

    The protocol also intentionally manipulates prior probability and learned value, so records should not reduce it to sensory evidence alone.

Open questions

  • Extend Brainwide Map curation from behavior trials to probe insertions, brain regions, spike-sorting QC, and video pose features.
  • Split biased and unbiased variants if the schema needs finer protocol granularity.

Papers

Findings

Curves attached to this protocol on shared signed-evidence axes.

See all findings overlaid on shared axes.