pith. sign in

arxiv: 2304.13901 · v2 · submitted 2023-04-27 · 💻 cs.GT

Conditional dominance in games with unawareness

Burkhard C. Schipper This is my paper

Pith reviewed 2026-05-24 09:15 UTC · model grok-4.3

classification 💻 cs.GT
keywords dynamic games with unawarenessgeneralized normal formextensive-form rationalizabilityconditional dominanceprudent rationalizabilityiterated dominance
0
0 comments X

The pith

The generalized normal form of a dynamic game with unawareness characterizes extensive-form rationalizability by iterated conditional strict dominance.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper translates dynamic games with unawareness, which consist of a partially ordered collection of extensive-form games, into an associated generalized normal form that is itself a partially ordered collection of normal-form games. It establishes that extensive-form rationalizability equals the result of iteratively deleting conditionally strictly dominated strategies in this generalized normal form, while prudent rationalizability equals the result of iteratively deleting conditionally weakly dominated strategies. The work shows that the standard iterated-admissibility concept fails to match extensive-form rationalizability under unawareness because an information set encodes both the nodes a player considers possible and the game trees of which she is aware.

Core claim

In dynamic games with unawareness represented as a partially ordered set of extensive-form games, the associated generalized normal form is a partially ordered set of normal-form games. Extensive-form rationalizability is characterized exactly by iterated conditional strict dominance in the generalized normal form, and prudent rationalizability is characterized exactly by iterated conditional weak dominance. The analogue of iterated admissibility depends on extensive-form structure because, under unawareness, a player's information set determines both the nodes she considers possible and the game trees of which she is aware.

What carries the argument

The generalized normal form: a partially ordered set of normal-form games that encodes the awareness structure of the original dynamic game.

If this is right

  • Rationalizability in dynamic games with unawareness can be computed by applying iterated conditional dominance directly to the generalized normal form.
  • Prudent rationalizability admits an analogous characterization via conditional weak dominance.
  • Standard iterated admissibility in the normal form does not coincide with extensive-form rationalizability once unawareness is present.
  • Solution concepts that ignore the extensive-form awareness structure at information sets will generally give incorrect predictions.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same generalized normal form might be usable to characterize other extensive-form solution concepts such as sequential equilibrium under unawareness.
  • Computational algorithms already developed for iterated dominance in normal-form games could be applied without modification to games with unawareness.
  • Mechanism-design problems involving agents who may be unaware of some contingencies could be analyzed by first moving to the generalized normal form.

Load-bearing premise

The generalized normal form preserves every piece of information that matters for conditional dominance and for rationalizability, including the awareness of entire game trees at each information set.

What would settle it

Construct a dynamic game with unawareness in which the set of strategies surviving iterated conditional strict dominance in the generalized normal form differs from the set of extensive-form rationalizable strategies; the characterization would then fail.

Figures

Figures reproduced from arXiv: 2304.13901 by Burkhard C. Schipper.

Figure 1
Figure 1. Figure 1: Example without Unawareness between actions n and g. This is followed by a simultaneous move game. The implication of Rowena’s action g is that Colin has all three actions B, S, and M. Otherwise, if Rowena takes action n, Colin is just left with actions B and S in the simultaneous game that follows. That is, 5 [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Associated Normal Form 5While the names of actions and the structure bears some similarity with the games discussed in Heifetz, Meier, and Schipper (2013), the payoffs differ. 6 [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Example of a Dynamic Game with Unawareness [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Associated Generalized Normal-form Game of the Example [PITH_FULL_IMAGE:figures/full_fig_p010_4.png] view at source ↗
read the original abstract

Heifetz, Meier, and Schipper (2013) introduced dynamic game with unawareness consisting of a partially ordered set of games in extensive form. Here, we study the normal form of dynamic games with unawareness. The generalized normal form associated with a dynamic game with unawareness consists of a partially ordered set of games in norm form. We use the generalized normal form to characterize extensive-form rationalizability (resp., prudent rationalizability) in dynamic games with unawareness by iterated conditional strict (resp., weak) dominance in the associated generalized normal form. We also show that the analogue to iterated admissibility for dynamic games with unawareness depends on extensive-form structure. This is because under unawareness, a player's information set not only determines which nodes she considers possible but also of which game tree(s) she is aware of.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

0 major / 3 minor

Summary. The paper extends Heifetz, Meier, and Schipper (2013) by defining the generalized normal form of a dynamic game with unawareness as a partially ordered set of normal-form games. It claims that iterated conditional strict dominance in this poset characterizes extensive-form rationalizability, while iterated conditional weak dominance characterizes prudent rationalizability. It further shows that the analogue of iterated admissibility depends on extensive-form structure because an information set encodes both possible nodes and the game trees of which the player is aware.

Significance. If the characterizations hold, the generalized normal form supplies a compact normal-form representation that preserves the conditional information needed for rationalizability analysis in games with unawareness. The result on iterated admissibility is a positive observation that highlights how awareness of trees affects dominance reasoning, distinguishing the setting from standard games.

minor comments (3)
  1. [§2] §2: the definition of the generalized normal form as a poset should include an explicit statement of the partial order and how it encodes awareness of subgames; the current description is brief and could be expanded with a small example.
  2. The proofs of the two main characterization theorems are referenced but not reproduced in the main text; moving at least the key inductive step to an appendix would improve readability without lengthening the paper.
  3. [§3] Notation for conditional dominance operators (strict vs. weak) is introduced in the abstract but first used without a forward reference in §3; a short notational table would help.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for the positive summary, the assessment of significance, and the recommendation of minor revision. No major comments were raised in the report.

Circularity Check

0 steps flagged

No significant circularity detected

full rationale

The paper defines the generalized normal form as a poset of normal-form games that encodes the conditional information from the 2013 extensive-form setup, then proves that iterated conditional strict (resp. weak) dominance in this poset characterizes extensive-form rationalizability (resp. prudent rationalizability). This is a standard mathematical characterization result, not a reduction of any quantity to itself by definition or by fitting a parameter to a subset of data. The 2013 citation supplies the underlying extensive-form games with unawareness; the normal-form representation and the dominance theorems are new constructions whose validity is established directly in the present paper rather than by self-referential equations or load-bearing self-citation chains. The additional observation that iterated admissibility depends on extensive-form structure follows from the explicit awareness-of-trees feature and does not collapse into the inputs. The derivation is therefore self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 1 invented entities

The central claim rests on the definition of the generalized normal form as a partially ordered set of normal-form games that faithfully encodes the unawareness structure from the 2013 extensive-form model.

axioms (2)
  • domain assumption Dynamic games with unawareness consist of a partially ordered set of extensive-form games (from Heifetz, Meier, Schipper 2013).
    This is the foundational setup assumed throughout the abstract.
  • domain assumption Conditional dominance is well-defined on the generalized normal form and corresponds to the extensive-form notions.
    Invoked when the abstract states the characterization.
invented entities (1)
  • Generalized normal form no independent evidence
    purpose: To represent dynamic games with unawareness as a partially ordered set of normal-form games for the purpose of applying iterated conditional dominance.
    New structure introduced in the paper to enable the stated characterizations.

pith-pipeline@v0.9.0 · 5665 in / 1460 out tokens · 30655 ms · 2026-05-24T09:15:17.847397+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

33 extracted references · 33 canonical work pages

  1. [1]

    Auster, S. (2013). Asymmetric awareness and moral hazard, Games and Economic Behavior 82, 503–521

    work page 2013

  2. [2]

    Auster, S. and N. Pavoni (2023). Optimal delegation and information transmission under limited awareness, Theoretical Economics, forthcoming

    work page 2023

  3. [3]

    Battigalli, P. (1997). On rationalizability in extensive games, Journal of Economic Theory 74, 40–61

    work page 1997

  4. [4]

    Battigalli, P. and M. Siniscalchi (2002). Strong belief and forward induction reasoning, Journal of Economic Theory 106, 356–391

    work page 2002

  5. [5]

    Brandenburger, A. and A. Friedenberg (2011). The relationship between rationality on the matrix and the tree, mimeo

    work page 2011

  6. [6]

    Cantonini, E. (2022). Iterated weak dominance does not refine strong rationalizability, New York University, Shanghai

    work page 2022

  7. [7]

    Chen, J. and S. Micali (2013). The order independence of iterated dominance in extensive games, Theoretical Economics 8, 125–163

    work page 2013

  8. [8]

    Feinberg, Y. (2021). Games with unawareness, B.E. Journal of Theoretical Economics 433–488

    work page 2021

  9. [9]

    Filiz-Ozbay, E. (2012). Incorporating unawareness into contract theory, Games and Eco- nomic Behavior 76, 181–194. 30

    work page 2012

  10. [10]

    Francetich, A. and B.C. Schipper (2022). Rationalizable screening and disclosure under unawareness, University of California, Davis

    work page 2022

  11. [11]

    Grant, S. and J. Quiggin (2013). Inductive reasoning about unawareness, Economic Theory 54, 717–755

    work page 2013

  12. [12]

    Guarino, P. (2020). An epistemic analysis of dynamic games with unawareness, Games and Economic Behavior 120, 257–288

    work page 2020

  13. [13]

    Halpern, J. and L. Rˆ ego (2014). Extensive games with possibly unaware players, Mathe- matical Social Sciences 70, 42–58

    work page 2014

  14. [14]

    Schipper (2013a)

    Heifetz, A., Meier, M., and B.C. Schipper (2013a). Dynamic unawareness and rationalizable behavior, Games and Economic Behavior 81, 50–68

    work page

  15. [15]

    Schipper (2021)

    Heifetz, A., Meier, M., and B.C. Schipper (2021). Prudent rationalizability in generalized extensive-form games with unawareness, B.E. Journal of Theoretical Economics 21, 525– 556

    work page 2021

  16. [16]

    and J.-F

    Kohlberg, E. and J.-F. Mertens (1986). On the strategic stability of equilibrium, Econo- metrica 54, 1003–1037

    work page 1986

  17. [17]

    Li, J. (2006). Dynamic games with perfect awareness information, mimeo

    work page 2006

  18. [18]

    Li, Y.X. and B.C. Schipper (2020). Strategic reasoning in persuasion games: An experi- ment, Games and Economic Behavior 121, 329–367

    work page 2020

  19. [19]

    Li, Y.X. and B.C. Schipper (2019). Disclosure under unawareness: An experiment, Uni- versity of California, Davis

    work page 2019

  20. [20]

    Swinkels (1993)

    Mailath, G., Samuelson, L., and J. Swinkels (1993). Extensive form reasoning in normal form games, Econometrica 61, 273–302

    work page 1993

  21. [21]

    Meier, M. and B.C. Schipper (2014). Bayesian games with unawareness and unawareness perfection, Economic Theory 56, 219–249

    work page 2014

  22. [22]

    Pearce, D.G. (1984). Rationalizable strategic behavior and the problem of perfection, Econometrica 52, 1029–1050

    work page 1984

  23. [23]

    Perea, A. (2022). Common belief in rationality in games with unawareness, Mathematical Social Sciences 119, 11–30

    work page 2022

  24. [24]

    Pram, K. and B.C. Schipper (2023). Efficient mechanisms under unawareness, University of California, Davis

    work page 2023

  25. [25]

    Rˆ ego, L. and J. Halpern (2012). Generalized Solution Concepts in Games with Possibly Unaware Players, International Journal of Game Theory 41, 131–155. 31

    work page 2012

  26. [26]

    Sadzik, T. (2021). Knowledge, awareness and probabilistic beliefs, B.E. Journal of Theo- retical Economics 21, 489–524

    work page 2021

  27. [27]

    Schipper, B.C. (2021). Discovery and equilibrium in games with unawareness, Journal of Economic Theory 198, 105365

    work page 2021

  28. [28]

    Schipper, B.C. (2019). Kuhn’ theorem for extensive games with unawareness, University of California, Davis

    work page 2019

  29. [29]

    Schipper, B.C. (2014). Unawareness - A gentle introduction to both the literature and the special issue, Mathematical Social Sciences 70, 1–9

    work page 2014

  30. [30]

    Schipper, B.C. and H.Y. Woo (2019). Political awareness, microtargeting of voters and negative electoral campaigning, Quartely Journal of Political Science 14, 41–88

    work page 2019

  31. [31]

    Schipper, B.C. and H. Zhou (2021). Insiders trading on unknown unknowns, University of California, Davis

    work page 2021

  32. [32]

    Shimoji, M. and J. Watson (1998). Conditional dominance, rationalizability, and game forms, Journal of Economic Theory 83, 161–195

    work page 1998

  33. [33]

    von Thadden, E. and X. Zhao (2012). Incentives for unawareness agents, Review of Eco- nomic Studies 79, 1151–1198. 32

    work page 2012