The role of context and team play in cross-game learning

One of the dividing lines between economics and psychology experiments is that economists favor abstract context while psychologists favor meaningful context. We investigate the effects of meaningful versus abstract context on cross-game learning in a signaling game experiment. With individual decision makers (1 × 1 games) meaningful context promotes positive cross-game learning in moving from a pooling equilibrium to a separating equilibrium, whereas abstract context yields negative cross-game learning. In 1×1 games a change in the (meaning-ful) context which accompanies "superficial" changes in the game stalls the learning process compared to an abstract context that does not change. In contrast, with two-person teams the same change in meaningful context has no disruptive effect on strategic play, with teams also having substantially higher levels of strategic play than the 1 × 1 games. We relate the effects of meaningful versus abstract context on cross-game learning to the psychology literature on deductive reasoning processes.