Correction: Humans perseverate on punishment avoidance goals in multigoal reinforcement learning

Article
Main text
Article and author information
Metrics

Main text

Sharp PB, Russek EM, Huys QJM, Dolan RJ, Eldar E. 2022. Humans perseverate on punishment avoidance goals in multigoal reinforcement learning. eLife 11:e74402. doi: 10.7554/eLife.74402.

Published 24 February 2022

Paul Sharp, the lead author, noticed two errors in the naming of variables in two the key equations in the article.

Equation 6 is incorrect. The equation in the published article read::

\begin{array}{ll} [\begin{array}{ll} Q_{M B_{t + 1}} (p r e s s = g) \\ Q_{M B_{t + 1}} (p r e s s = j) \end{array}] \\ = [\begin{array}{ll} β_{M B r e w a r d} P (f = r e w a r d | p r e s s = g)_{t} & β_{M B p u n i s h} P (f = p u n i s h | p r e s s = g)_{t} \\ β_{M B r e w a r d} P (f = r e w a r d | p r e s s = j)_{t} & β_{M B p u n i s h} P (f = p u n i s h | p r e s s = j)_{t} \end{array}] [\begin{matrix} 0 \\ - 1 \end{matrix}] \end{array}

The correction to equation (6) is as follows:

\begin{array}{ll} [\begin{array}{ll} Q_{G P_{t + 1}} (p r e s s = g) \\ Q_{G P_{t + 1}} (p r e s s = j) \end{array}] \\ = [\begin{array}{ll} β_{G P r e w a r d} P (f = r e w a r d | a = p r e s s g)_{t} & β_{G P p u n i s h} P (f = p u n i s h | a = p r e s s g)_{t} \\ β_{G P r e w a r d} P (f = r e w a r d | a = p r e s s j)_{t} & β_{G P p u n i s h} P (f = p u n i s h | a = p r e s s j)_{t} \end{array}] \end{array}

Equation 8 also also had a slight error in how the action variable was named. It is now consistent with how all other equations reflect the action taken. The original equation in the article was:

P (a = p r e s s j) = \frac{e^{Q_{I n t e g r a t e d} (p r e s s = j)}}{e^{Q_{I n t e g r a t e d} (p r e s s = j)} + e^{Q_{I n t e g r a t e d} (p r e s s = g)}}

The equation is now correction to read:

P (p r e s s = j) = \frac{e^{Q_{I n t e g r a t e d} (p r e s s = j)}}{e^{Q_{I n t e g r a t e d} (p r e s s = j)} + e^{Q_{I n t e g r a t e d} (p r e s s = g)}}

We apologize for any confusion this might have caused.

Article and author information

Author details

Paul B Sharp

For correspondence
paul.sharp@mail.huji.ac.il

"This ORCID iD identifies the author of this article:" 0000-0003-4949-1501
Evan M Russek
Quentin JM Huys
Raymond J Dolan

"This ORCID iD identifies the author of this article:" 0000-0001-9356-761X
Eran Eldar

Publication history

Version of Record published: October 10, 2022 (version 1)

Copyright

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.