Figure - PMC

Skip to main content

An official website of the United States government

Here's how you know

Here's how you know

Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS
A lock ( ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

View full-text article in PMC

. 2021 Aug 13;47(1):58–71. doi: 10.1038/s41386-021-01123-1

Search in PMC
Search in PubMed
View in NLM Catalog
Add to search

© The Author(s) 2021

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

PMC Copyright notice

Fig. 1 — A The goal of learning is to obtain certain outcomes by selecting appropriate actions based on presented stimuli while considering the context that includes internal state as well as external cues that reflect the latent state of the environment. This requires linking stimuli, actions, and outcomes, which can be done in multiple ways each with different levels of flexibility. B Different types of learning strategies for linking stimuli (S), actions (A), and outcomes (O) and their main shortcomings. (1) S-Rew associations link reward values (Rew value) of the outcomes to certain stimuli that precede those outcomes, allowing for the computation of stimulus value. Such a model cannot correctly link S and O if reward that follows the same stimulus (Rew’) or the state of the animal changes. (2) A-Rew associations link reward values (Rew value) of the outcomes to certain actions that precede those outcomes, allowing for the computation of action value. Such a model cannot correctly link A and O if reward (Rew’) that follows the same action or state of the animal changes. (3) S–A associations or selective models link the chosen action and the stimulus that precedes this action using experienced rewards. Such models cannot link S and A if reward type or state of the animal changes. (4) S–O (similarly feature–outcome, F–O) associations or predictive models link S (respectively, F) and O by learning the probability of outcomes contingent upon stimuli and/or their features regardless of their rewarding values through encoding the statistical occurrences of these outcomes. (5) A–O associations or predictive models link A and O by learning the probability of outcomes contingent upon actions regardless of their rewarding values. Predictive models cannot easily transfer learning from one context to another context. C Flexible link between stimuli, actions, and outcomes through creation of task sets consisting of multiple internal models (see text for more details).