Three Very Simple Games and What it Takes to Solve Them

Abstract

We study experimentally the nature of dominance violations in three minimalist dominancesolvable guessing games. Only about a third of our subjects report reasoning consistent with dominance; they all make dominant choices and almost all expect others to do so. Nearly two-thirds of subjects report reasoning inconsistent with dominance, yet a quarter of them actually make dominant choices and half of those expect others to do so. Reasoning errors are more likely for subjects with lower working memory, intrinsic motivation and premeditation attitude. Dominance-incompatible reasoning arises mainly from subjects misrepresenting the strategic nature (payoff structure) of the guessing games. AbstractWe study experimentally the nature of dominance violations in three minimalist dominancesolvable guessing games. Only about a third of our subjects report reasoning consistent with dominance; they all make dominant choices and almost all expect others to do so. Nearly twothirds of subjects report reasoning inconsistent with dominance, yet a quarter of them actually make dominant choices and half of those expect others to do so. Reasoning errors are more likely for subjects with lower working memory, intrinsic motivation and premeditation attitude. Dominance-incompatible reasoning arises mainly from subjects misrepresenting the strategic nature (payoff structure) of the guessing games. Table 4 116 8 1 1 0 3 3 Choice=1 & Belief=0 3 3 0 0 0 0 0 Choice=1 & Belief=1 47 4 6 1 0 0 0 0 Three Very Simple Games and What It Takes to Solve Them AbstractWe study experimentally the nature of dominance violations in three minimalist dominancesolvable guessing games. Only about a third of our subjects report reasoning consistent with dominance; they all make dominant choices and almost all expect others to do so. Nearly twothirds of subjects report reasoning inconsistent with dominance, yet a quarter of them actually make dominant choices and half of those expect others to do so. Reasoning errors are more likely for subjects with lower working memory, intrinsic motivation and premeditation attitude. Dominance-incompatible reasoning arises mainly from subjects misrepresenting the strategic nature (payoff structure) of the guessing games.

Cognitive Ability, Character Skills, and Learning to Play Equilibrium: A Level-kAnalysis

“…In related dominance-solvable and guessing games, working memory (Rydval et al, 2009) and depths of reasoning in the red hat puzzle (Bayer and Renou, 2012) correlate with behavior. Cognitive ability also influences behavior in public good games (Millet and Dewitte, 2007) and in the prisoner's dilemma (Burks et al, 2009).…”

“…More agreeable subjects send more in a sequential prisoner's dilemma (Rustichini et al, 2012). Higher need-for-cognition and premeditation give rise to fewer dominated choices in dominance-solvable games (Rydval et al, 2009). More neurotic subjects earn more in the stag-hunt game (Al-Ubaydli et al, 2013).…”

Three very simple games and what it takes to solve them

“…The order of cognitive tests and personality scales was the same across sessions, with the former generally preceding the latter. The working memory and short-term memory tests were computerized using E-prime (Schneider et al, 2002) while the remainder of the experiment was administered in a paper-and-pencil format. Table 2 displays the number of subjects in the reasoning classes defined earlier, aggregated across the three guessing games.…”

“…27 As an exception, arithmetic ability affects both reasoning classes and choices when included instead of or besides working memory, but its impact is generally weaker than that of working memory. We measure arithmetic ability using a "two-digit addition and subtraction" test under time pressure (see Rydval et al, 2008, for more details). Since we lack arithmetic ability scores for the first experimental session, we focus on the effect of working memory in the full sample, noting that working memory is correlated with arithmetic ability at the 10 percent significance level (Spearman correlation of 0.19) and hence that part of the explanatory power of working memory may reflect the impact of arithmetic ability.…”

Feature-Based Choice and Similarity in Normal-Form Games: An Experimental Study

“…Behavioral models estimated using large data sets (Weizsacker 2003), and experiments that try to track down individual reasoning processes (Devetag and Warglien 2008;Rydval, Ortmann and Ostatnicky 2009) suggest that players reason through incomplete models of the strategic situation at hand, either tending to ignore their opponents' incentives when making their choices, or treating them as mirror images of their own. Experiments that elicit both choices and beliefs about opponents' play point at a general inconsistency between choices and beliefs, suggesting that in the absence of learning opportunities and feedback, the assumption that actions are driven by beliefs about the opponent must not be taken for granted (Costa-Gomes and Weizsacker 2008;Stahl and Haruvy 2008).…”