Abstract
In 2011, we tested the hypothesis that people exhibit a right-oriented bias when they are approach motivated and act quickly (Roskes, Sligte, Shalvi, & De Dreu, 2011). An experiment showed that when people had to act quickly, they bisected lines farther to the right when they were approach motivated than when they were avoidance motivated. Analysis of archival data on soccer penalty shoot-outs further revealed that goalkeepers dived more to the right when their team was behind than when their team was not behind, a situation we propose induces approach motivation.
Price and Wolfers (2014) challenge whether the right-oriented bias manifests itself in goalkeepers’ behavior. They make three critiques of our findings: (a) The effect does not replicate, (b) an alternative coding of “being behind” eliminates the effect, and (c) the goalkeepers’ tendency to dive right is not dysfunctional. Our analysis suggests that the bias exists, although Price and Wolfers’s alternative coding raises interesting questions about the exact settings that evoke approach motivation. We are happy to see that more data are being collected, which is important for enhancing understanding of the phenomenon.
Prior research has demonstrated an association between approach motivation and a variety of right-oriented biases. This association is explained by increased left-hemispheric brain activation under approach motivation, which enhances attention and action readiness toward the right (Vallortigara & Rogers, 2005). For example, dogs wag their tails toward the right when they observe their owners (Quaranta, Siniscalchi, & Vallortigara, 2007), and when quickly dividing lines into equal parts, approach-motivated people divide them to the right of their centers (Nash, McGregor, & Inzlicht, 2010). In our original study, we tested whether right-oriented bias under approach motivation is more likely to appear when people have to act fast than when they have more time in which to override their automatic responses. Price and Wolfers challenge neither the theory nor the results of our experiment. Rather, they challenge whether approach motivation evokes right-oriented bias in goalkeepers during penalty shoot-outs.
Does Right-Oriented Bias Affect Goalkeepers?
In our original article, we aimed to identify real-life situations in which right-oriented biases occur. During soccer penalty shoot-outs, goalkeepers make split-second decisions between diving left and diving right. Penalty shoot-outs are likely to evoke approach motivation among goalkeepers because, as Oliver Kahn (a famous German goalkeeper) explained, “Kickers are the ones that can lose in a penalty shoot-out; goalkeepers . . . can win and ultimately become the heroes” (quoted in “Goalkeepers Give Shoot-Out Tips,” 2010, para. 25). We predicted that approach motivation would be even stronger when the goalkeepers’ team is behind and their role in winning the match is crucial. We analyzed all penalty shoot-outs from Fédération Internationale de Football Association (FIFA) World Cup matches: Goalkeepers whose team was behind were more likely to dive right (71%) than left (29%), compared with those whose team was not behind (48% vs. 49%, respectively); see Table 1.
Goalkeeper and Kicker Behavior Across Different Game Situations
Note: The original data from Fédération Internationale de Football Association (FIFA) World Cup shoot-outs between 1982 and 2010 were taken from Roskes, Sligte, Shalvi, & De Dreu (2011). The original data from shoot-outs during the Copa América tournament and Union of European Football Associations (UEFA) Champions League final games between 1984 and 2011 were taken from Price and Wolfers (2014). The chi square interaction p value compares the tendency to dive right with the tendency to dive left as a function of whether the goalkeeper’s team is behind versus not behind. The chi square simple-effect p value compares the observed frequency of diving (or kicking) right with the frequency of diving (or kicking) left with the expected frequency of choosing a side at random (50%-50%).
Price and Wolfers’s first critique is that they could not replicate the pattern in two additional competitions: Union of European Football Associations (UEFA) Champions League and Copa América. In these tournaments, the interaction between standing of the goalkeeper’s team (behind vs. not behind) and diving direction (right vs. left) was not significant (Table 1). Recent recommendations suggest that assessing accumulated data is more informative than interpreting results of single studies (Braver, Thoemmes, & Rosenthal, 2014; Stanley & Spence, 2014). We therefore aggregated the data from the three tournaments, and this revealed the same pattern as the original data: Goalkeepers whose team was behind were more likely to dive right (68%) than left (27%), compared with those whose team was not behind (52% vs. 43%, respectively), p = .022 (Table 1).
Cumming (2013) proposed that replications should not be labeled as successes (p < .05) or failures (p > .05) and that confidence intervals are a better way to present differences. Figure 1 presents the differences between the proportion of goalkeepers who dived right when their team was behind and the proportion who dived right when their team was not behind. Indeed, the replication does not rule out a zero effect in the combined UEFA and Copa América data (zero falls within the confidence interval). However, it is also consistent with the existence of right-oriented bias.
Effect of the standing of the goalkeeper’s team on the goalkeeper’s decision to dive right. For each data set, the proportion of times that the goalkeepers dived right when their team was not behind was subtracted from the proportion of times that the goalkeepers dived right when their team was behind. The top three data points show results for Fédération Internationale de Football Association (FIFA) World Cup shoot-outs between 1982 and 2010 (taken from Roskes, Sligte, Shalvi, & De Dreu, 2011), shoot-outs during the Copa América tournament and Union of European Football Associations (UEFA) Champions League final games between 1984 and 2011 (taken from Price & Wolfers, 2014), and the combined data from both studies. For these analyses, goalkeepers’ teams were considered behind if they scored fewer goals than their competitors’ teams. The lower three data points show the same data recoded according to alternative criteria devised by Price and Wolfers, in which goalkeepers’ teams were considered behind if they missed more goal attempts than their competitors’ teams. The plot was generated using Exploratory Software for Confidence Intervals (ESCI; Cumming, 2013). Error bars show 95% confidence intervals.
Price and Wolfers’s second critique relates to the situations evoking approach motivation. Because the game’s score is salient, we had originally coded goalkeepers’ teams as behind if their team scored fewer goals than their competitors. Price and Wolfers devised an alternative coding method, in which goalkeepers’ teams were considered behind if their team missed more attempts than the competitors. In this situation, the score (from the goalkeeper’s perspective) can be either behind or tied. Although the new coding includes these different situations, we agree that this may be an elegant way to assess goalkeepers’ motivation. However, as acknowledged by Price and Wolfers, the alternative coding does not invalidate our original coding.
Recoding our original data using Price and Wolfers’s method revealed no evidence of right-oriented bias (Table 1). However, the aggregated data show that goalkeepers whose team was behind were more likely to dive right (60%) than left (36%), compared with those whose team was not behind (52% vs. 42%, respectively). Although consistent with the predicted pattern, the interaction was not significant, p = .16. Price and Wolfers wrote that “the conclusion of Roskes et al. is highly sensitive to the assumption about which situations are likely to yield approach motivation” (p. 2111). It is indeed fascinating to figure out whether stronger approach motivation is evoked if one considers teams behind when they missed more attempts to score than if one considers them behind if they scored less, which can be tested in both laboratory and field settings.
Finally, Price and Wolfers suggest that the goalkeepers’ right-oriented bias is not dysfunctional. In Roskes et al. (2011), we tested the single prediction that right-oriented bias emerges under approach motivation. We had no predictions about the bias being either functional or dysfunctional. We discussed the possibility that the bias has disadvantages because it reduces accuracy and can be exploited. Price and Wolfers suggest that “the reason that diving right in these situations is a good choice is that the kicker is more likely to kick to the goalkeeper’s right: 54.2% of the time versus 37.5%” (p. 2110). However, as can be seen in Table 1, effects for kickers were not significant.
Conclusion
Approach motivation has been associated with a variety of right-oriented biases. Price and Wolfers challenge our observation that the bias exists among goalkeepers. The accumulated data support the existence of right-oriented bias; however, many questions remain open. For example, which situations evoke approach motivation, and when does it translate to right-oriented bias? In which settings does the bias emerge (e.g., drivers’ behavior)? When is the bias dysfunctional? We hope the discussion continues and that researchers will keep collecting new data to provide insight into if, when, and why motivation evokes directional biases.
Footnotes
Declaration of Conflicting Interests
The authors declared that they had no conflicts of interest with respect to their authorship or the publication of this article.