The effect of adding a constant to all payoffs: Experimental investigation, and implications for reinforcement learning models

Ido Erev; Yoella Bereby-Meyer; Alvin E. Roth

doi:10.1016/s0167-2681(99)00028-1

The effect of adding a constant to all payoffs: Experimental investigation, and implications for reinforcement learning models

Ido Erev, Yoella Bereby-Meyer, Alvin E. Roth

Department of Psychology

Research output: Contribution to journal › Article › peer-review

71 Scopus citations

Abstract

This paper examines the effect on learning in simple decision tasks of the addition of a constant to all payoffs. Experiment 1 reveals that this effect, initially observed in a probability learning task, is not limited to single person decision making under uncertainty. Experiment 2 shows that the effect is not linear. Two additional experiments show that the non-linearity cannot be explained by whether zero is in the payoff range. The implications of these results for reinforcement learning models are evaluated and two models that capture the main results are proposed.

Original language	English
Pages (from-to)	111-128
Number of pages	18
Journal	Journal of Economic Behavior and Organization
Volume	39
Issue number	1
DOIs	https://doi.org/10.1016/s0167-2681(99)00028-1
State	Published - 1 Jan 1999

Keywords

Adaptive learning
Probability matching
Reference point
Reinforcement
Stimulus response

ASJC Scopus subject areas

Economics and Econometrics
Organizational Behavior and Human Resource Management

Access to Document

10.1016/s0167-2681(99)00028-1

Cite this

@article{50a1322a5229418b8bc91d3770771d19,

title = "The effect of adding a constant to all payoffs: Experimental investigation, and implications for reinforcement learning models",

abstract = "This paper examines the effect on learning in simple decision tasks of the addition of a constant to all payoffs. Experiment 1 reveals that this effect, initially observed in a probability learning task, is not limited to single person decision making under uncertainty. Experiment 2 shows that the effect is not linear. Two additional experiments show that the non-linearity cannot be explained by whether zero is in the payoff range. The implications of these results for reinforcement learning models are evaluated and two models that capture the main results are proposed.",

keywords = "Adaptive learning, Probability matching, Reference point, Reinforcement, Stimulus response",

author = "Ido Erev and Yoella Bereby-Meyer and Roth, {Alvin E.}",

note = "Funding Information: This research was supported in part by a grant from National Science Foundation to the Univ. of Pittsburgh. It has benefited from related research and insightful conversations with Bob Slonim, Joachim Meyer and Sharon Gilat and from the comments of the participants of the workshop on Economics and Psychology at the Univ. of British Columbia on June, 1997. ",

year = "1999",

month = jan,

day = "1",

doi = "10.1016/s0167-2681(99)00028-1",

language = "English",

volume = "39",

pages = "111--128",

journal = "Journal of Economic Behavior and Organization",

issn = "0167-2681",

publisher = "Elsevier",

number = "1",

}

TY - JOUR

T1 - The effect of adding a constant to all payoffs

T2 - Experimental investigation, and implications for reinforcement learning models

AU - Erev, Ido

AU - Bereby-Meyer, Yoella

AU - Roth, Alvin E.

N1 - Funding Information: This research was supported in part by a grant from National Science Foundation to the Univ. of Pittsburgh. It has benefited from related research and insightful conversations with Bob Slonim, Joachim Meyer and Sharon Gilat and from the comments of the participants of the workshop on Economics and Psychology at the Univ. of British Columbia on June, 1997.

PY - 1999/1/1

Y1 - 1999/1/1

N2 - This paper examines the effect on learning in simple decision tasks of the addition of a constant to all payoffs. Experiment 1 reveals that this effect, initially observed in a probability learning task, is not limited to single person decision making under uncertainty. Experiment 2 shows that the effect is not linear. Two additional experiments show that the non-linearity cannot be explained by whether zero is in the payoff range. The implications of these results for reinforcement learning models are evaluated and two models that capture the main results are proposed.

AB - This paper examines the effect on learning in simple decision tasks of the addition of a constant to all payoffs. Experiment 1 reveals that this effect, initially observed in a probability learning task, is not limited to single person decision making under uncertainty. Experiment 2 shows that the effect is not linear. Two additional experiments show that the non-linearity cannot be explained by whether zero is in the payoff range. The implications of these results for reinforcement learning models are evaluated and two models that capture the main results are proposed.

KW - Adaptive learning

KW - Probability matching

KW - Reference point

KW - Reinforcement

KW - Stimulus response

UR - http://www.scopus.com/inward/record.url?scp=0033130546&partnerID=8YFLogxK

U2 - 10.1016/s0167-2681(99)00028-1

DO - 10.1016/s0167-2681(99)00028-1

M3 - Article

AN - SCOPUS:0033130546

SN - 0167-2681

VL - 39

SP - 111

EP - 128

JO - Journal of Economic Behavior and Organization

JF - Journal of Economic Behavior and Organization

IS - 1

ER -

The effect of adding a constant to all payoffs: Experimental investigation, and implications for reinforcement learning models

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this