Framing has been widely shown to affect decision making. In this paper, we investigate experimentally whether, and to what extent, cooperative behaviour in a Game of Chicken may be impacted by a very basic change in the labelling of the strategies. Our within-subject experimental design involves two treatments. The only difference between them is that we introduce either a socially-oriented wording (I cooperate/I do not cooperate) or colours (red/blue) to designate strategies. The level of cooperation appears to be higher in the socially-oriented context, but only when uncertainty as regards the type of the partner is manipulated, and especially among females. . We thank Tim Cason as well as two anonymous referees for their insightful comments and suggestions.1 From a normative point of view, only the 'deep structure' of the decision problem should influence decision making. 43 44 Bulletin of Economic Research et al., 2011); and (ii) 'label' framing, which consists in manipulating the game's title, the game's characterization, the labelling of players' strategies or type of player, and so on (see, for instance, Rege and Telle, 2004;Zhong et al., 2007;Dufwenberg et al., 2011). In this paper, we are interested in how a change in the labelling of the strategies is likely to affect cooperative behaviour.To avoid confounding framing effects and better control the data, interactive decision making has often been investigated using context-free experimental designs that 'typically remove informational cues that might provide richer meaning to the game' (Zhong et al., 2007, p. 432). In particular, defective/cooperative strategies are usually designated using a neutral wording, using such labels as 'A and B; C and D; X and Y; red, blue, yellow, or white; up and down; or left and right ' (pp. 433-4).On the other hand, real life usually involves explicit decision settings, which may impair the external validity of data collected using a context-free experimental design. A number of experimental studies have investigated the influence of 'valence' and 'label' framing on cooperative behaviour, be it by comparing behaviour in a context-free setting with behaviour in a context-rich setting, or by comparing behaviour in different context-rich settings (for example, depending on whether the focus is made on either cooperation or competition, sharing money or earning money for oneself, and so on). Most of the time, the level of cooperation appears to be higher when frame manipulation is intended to put the subjects in a socially-oriented mood.Decision making in interactive settings in general, and especially cooperative behaviour, has also been shown to be gender-dependent, even though the direction of the evidence remains unclear. 2 However, even though men may exhibit a more cooperative (or contributive) behaviour than women (e.g., Schwieren and Sutter, 2008), the opposite pattern seems to be more frequent (e.g., Gächter et al., 2004; see also Eckel andGrossman, 2008a andCroson andGneezy, 2009 for comprehensive reviews).Severa...