Psychological game theory, as defined by Colman, is meant to offer a series of solution concepts that should reduce the indeterminacy of orthodox game theory when applied to a series of situations. My main criticism is that, actually, they introduce a second-order indeterminacy problem rather than offering a viable solution. The reason is that the proposed solution concepts are under-specified in their definition and in their scope.

Almost anyone seriously interested in decision theory will name John von Neumann's (1928) Minimax Theorem as its foundation, whereas Utility and Rationality are imagined to be the twin towers on which the theory rests. Yet, experimental results and real-life observations seldom support that expectation. Over two centuries ago, Hume (1739–40/1978) put his finger on the discrepancy. “Reason,” he wrote “is, and ought to be the slave of passions, and can never pretend to any other office than to serve and obey them.” In other words, effective means to reach specific goals can be prescribed, but not the goals. A wide range of experimental results and daily life behavior support this dictum.

“Psychology Game Theory” grafts social-process explanations onto classical game theory to explain deviations from instrumental rationality caused by the social properties of cooperation. This leads to confusion between cooperation as a social or individual behavior, and between ultimate and proximate explanations. If game theory models explain the existence of cooperation, different models are needed for understanding the proximate social processes that underlie cooperation in the real world.

Colman shows that normative theories of rational decision-making fail to produce rational decisions in simple interactive games. I suggest that well-formed theories are possible in local settings, keeping in mind that a good part of each game is the generation of a rational approach appropriate for that game. The key is rationality defined in terms of the game, not individual decisions.

On top of the puzzles mentioned by Colman comes the puzzle of why rationality has bewitched classical game theory for so long. Not the smallest merit of evolutionary game theory is that it views rationality as a limiting case, at best. But some problems only become more pressing.

We propose that a direct analogy can be made between optimal behaviour in animals and rational behaviour in humans, and that lessons learned by the study of the former can be applied to the latter. Furthermore, we suggest that, to understand human decisions, rationality must be considered within an evolutionary framework.

This commentary focuses on the parts of psychological game theory dealing with preference, as illustrated by team reasoning, and supports the conclusion that these theoretical notions do not contribute above and beyond existing theory in understanding social interaction. In particular, psychology and games are already bridged by a comprehensive, formal, and inherently psychological theory, interdependence theory (Kelley & Thibaut 1978; Kelley et al. 2003), which has been demonstrated to account for a wide variety of social interaction phenomena.

We argue that solving the heterogeneous problems arising from the standard game theory requires looking both at reasoning heuristics, as in Colman's analysis, and at how people represent games and the quantities that define them.

Game theory's paradoxes stimulate the study of rationality. Sometimes they motivate the revising of standard principles of rationality. Other times they call for revising applications of those principles or introducing supplementary principles of rationality. I maintain that rationality adjusts its demands to circumstances, and in ideal games of coordination it yields a payoff-dominant equilibrium.

Psychological game theory encompasses formal theories designed to remedy game-theoretic indeterminacy and to predict strategic interaction more accurately. Its theoretical plurality entails second-order indeterminacy, but this seems unavoidable. Orthodox game theory cannot solve payoff-dominance problems, and remedies based on interval-valued beliefs or payoff transformations are inadequate. Evolutionary game theory applies only to repeated interactions, and behavioral ecology is powerless to explain cooperation between genetically unrelated strangers in isolated interactions. Punishment of defectors elucidates cooperation in social dilemmas but leaves punishing behavior unexplained. Team reasoning solves problems of coordination and cooperation, but aggregation of individual preferences is problematic.

The strong predominance of right-handedness appears to be a uniquely human characteristic, whereas the left-cerebral dominance for vocalization occurs in many species, including frogs, birds, and mammals. Right-handedness may have arisen because of an association between manual gestures and vocalization in the evolution of language. I argue that language evolved from manual gestures, gradually incorporating vocal elements. The transition may be traced through changes in the function of Broca's area. Its homologue in monkeys has nothing to do with vocal control, but contains the so-called “mirror neurons,” the code for both the production of manual reaching movements and the perception of the same movements performed by others. This system is bilateral in monkeys, but predominantly left-hemispheric in humans, and in humans is involved with vocalization as well as manual actions. There is evidence that Broca's area is enlarged on the left side in Homo habilis, suggesting that a link between gesture and vocalization may go back at least two million years, although other evidence suggests that speech may not have become fully autonomous until Homo sapiens appeared some 170,000 years ago, or perhaps even later. The removal of manual gesture as a necessary component of language may explain the rapid advance of technology, allowing late migrations of Homo sapiens from Africa to replace all other hominids in other parts of the world, including the Neanderthals in Europe and Homo erectus in Asia. Nevertheless, the long association of vocalization with manual gesture left us a legacy of right-handedness.

The causes of asymmetries for handedness and cerebral speech are of scientific interest, but is it sensible to try to determine which of these came first? I argue that (1) first causes belong to mythology, not science; (2) much of the cited evidence is weak; and (3) the treatment of individual differences is inadequate in comparison with the right shift theory.

The intriguing observation that left-cerebral dominance for vocalization is ancient, occurring in frogs, birds, and mammals, grounds Corballis's argument that the predominance of right-handedness may result from an association between manual gestures and vocalization in the evolution of language. This commentary supports the general thesis that language evolved “From hand to mouth” (Corballis 2002), while offering alternatives for some of Corballis's supporting arguments.

The communicative behavior of chimpanzees has been cited in support of the hypothesis that language evolved from gesture. In this commentary, I compare gestural and vocal communication in wild chimpanzees. Because the use of gesture in wild chimpanzees is limited, whereas their vocal behavior is relatively complex, I argue that wild chimpanzee behavior fails to support the gestural origins hypothesis.

Corballis presents a plausible evolutionary mechanism to explain the tight linkage between cerebral lateralization for language and for handedness in humans. This argument may be bolstered by invoking Stokoe's notion of semantic phonology to explain the role of Broca's area in grammatical functions.

The role of Broca's area is currently unclear even with regard to language. Suggestions that this area was enlarged on the left in certain of our hominid ancestors are unconvincing. Broca's area may have nothing to do with a lateralized gestural or vocal system Handedness may have evolved more than four million years ago.

An intervening gestural stage in language evolution, though seductive, is ultimately redundant, and is not necessarily supported by modern human or chimp behaviour. The findings and arguments offered from mirror neurones, anatomy, and lateralization are capable of other interpretations, and the manipulative dextrality of chimps is under-recognized. While language certainly possesses certain unique properties, its roots are ancient.

We add evidence in support of Corballis's gestural theory of language. Using transcranial magnetic stimulation, we found that productive and receptive linguistic tasks excite the motor cortices for both hands. This indicates that the language and the hand motor systems are still tightly linked in modern man. The bilaterality of the effect, however, implies that lateralisation is a secondary issue.

Corballis's claim that the origin of syntax lies in solely gesture is contested. His scenario does not explain why constraints on syntactic “movement” are apparently part of the human biological endowment for language. It also does not pay enough attention to the internal structure of sentences, and how they contrast with other linguistic units such as noun phrases.

What do the relationships observed in the occurrence of various limb, facial, and speech apraxias following left hemisphere damage mean for Corballis's theory? What does the right hemisphere's role in nonpropositional and automatic speech production tell us about the coevolution of right hand preference and speech; how could the possibility that the right hemisphere may be “dominant” for some aspects of speech be accommodated by his theory?