Also, I think the table is a little more complicated than the basic prisoner's dilemma ourlined by
D & R cooperate: D=2; R=2.
D capitulate to R: D=1; R=3.
R capitulate to D: D=3; R=1.
When both sides refuse to cooperate, and neither capitulates, there are three possible outcomes.
"Catastrophic Failure" for D: (D=0;R=4)
"Catastrophic Failure" for R: (D=4; R=0)
"Mutually Assured Destruction": (D=0; R=0)
Part of the problem is that these three outcomes are all possible when D and R adopt non-cooperative strategies and neither capitulates. What controls the outcome in that case are factors that are both exogenous and endogenous to the game.
D decision makers appear to have a much lower risk tolerance for the two possible negative outcomes than R decision makers, another factor in their preference for the capitulation strategy. This may be an alternative explanation to the concept that Ds persistently fail to identify the nature of the game as iterative, or may simply contribute to the cognitive dissonance theory.