Mice infer probabilistic models for timing
Read::
- Mice infer probabilistic models for timing Yi Li, Joshua Tate Dudman 2013 🛫 NA reading citation Print:: ❌
- print Zotero Link:: NA PDF:: NA Files:: Li_Dudman_2013_Mice infer probabilistic models for timing.pdf Reading Note:: Yi Li, Joshua Tate Dudman 2013 Web Rip::
TABLE without id
file.link as "Related Files",
title as "Title",
type as "type"
FROM "" AND -"ZZ. planning"
WHERE citekey = "liMiceInferProbabilistic2013"
SORT file.cday DESCAbstract
NA
Quick Reference
Top Comments
Topics
SIV Task
we develop a switching interval variance (SIV) operant conditioning task for mice. Optimal performance of the SIV task required mice to adapt their behavior to the mean and the SD of reward delays. We find that mice adjust their behavior to the SD of reward delays across an order of magnitude change in variability.
The SIV operant conditioning task is schematically illustrated in Fig. 1. Mice (n = 9) were placed in a behavioral chamber with two thin metal levers protruding from the walls of the box and a recessed central port (“reward port”)at which water could be delivered (Fig. 1A). For any given trial only one lever (the “baited” lever) would lead to the delivery of a delayed water reward on ∼85% of trials (Fig. 1B) and the side of the baited lever was switched in a blockwise fashion (180–200 trials/block) to ensure goal-directed and thus potentially optimal behavior (26). This design produced three distinct trial types: a correct choice of the baited lever followed by water delivery (“rewarded”), a correct choice of baited lever with no water delivered (“probe”), and an incorrect choice of the unbaited lever (“error”). In all cases, completion of a trial required approaching the reward port, and in the case of rewarded trials, the water reward had to be collected. The time delay between lever press and delivery of the water reward was randomly drawn from a Gaussian probability density function (pdf). The SD of the reward delay pdf (σdelay) in each block was selected from a set of three possible values: 50, 750, and 2,000 ms. All blocks had a mean reward delay (μdelay)of3s
Tasks
—