Questioning Anecdotal Beliefs and Scientific Findings

In their Commentary, Strayer, Medeiros-Ward, and Watson (2013) questioned the validity of our findings (Mäntylä, 2013) by arguing that our data do not provide clear and unambiguous evidence for gender differences in multitasking. They also suggested that individual differences in the ability to multitask are more likely associated with executive attention than with spatial ability.

Strayer was one of the three reviewers of the Mäntylä (2013) paper, and we appreciate his constructive comments at this and an earlier stage. Yet our conclusions are quite different. In contrast to Strayer et al.’s (2013) interpretation, our conclusion was not that there are gender differences in all real-world versions of multitasking. Rather than arguing that the gender differences are the norm or supporting anecdotal beliefs about female superiority, we reasoned that men might actually be better than women in some conditions of multitasking. We believe that such a hypothesis deserves attention, given the public interest in this issue and because the fact that the scientific community has paid virtually no attention to possible gender differences in multitasking. Questioning both anecdotal beliefs about female superiority and strong scientific assumptions about gender invariance might be one way of advancing the limited knowledge in this area.

The two experiments of the Mäntylä (2013) study showed clear gender differences in multitasking favoring men. In contrast to the interpretation of Strayer and colleagues (2013), these findings were not presented as “sufficient empirical support to justify the strong claim that men and women differ in their ability to multitask” (p. 809). Instead, our interpretation of the main findings was rather cautious:

A central assumption of the spatiotemporal hypothesis introduced here is that gender differences in multitasking are expected when the demands on temporal coordination are relatively high. In most dual-task conditions, these demands are low and less dependent on spatial abilities than are multiple tasks that may require coordination of a complex pattern of temporal contingencies. This line of reasoning is also consistent with evidence from earlier studies showing no gender-related differences in dual-task conditions. (Mäntylä, 2013, p. 519)

Furthermore, the final paragraph of the article stated:

It is reasonable to assume that, like most goal-directed tasks, everyday multitasking reflects different mixtures of task-independent cognitive functions (e.g., components of executive functioning and spatial processing) and more domain- specific skills and strategies. From this perspective, individual differences in multitasking should be considered in relative terms, given that some conditions may show reduced or even reversed gender differences because of task-specific constraints and strategies. (Mäntylä, 2013, p. 519) A reasonable interpretation of these statements is that we are not making strong claims about gender invariance.

The main criticism of Strayer et al. (2013) was based on the observation that they have systematically failed to find gender differences in the ability to concurrently operate a motor vehicle and talk on a cell phone. Questioning the validity of the spatiotemporal hypothesis proposed in Mäntylä (2013), they argued that because “driving involves spatial processing for route navigation and lane maintenance” (p. 809) and because they have not observed gender differences in these dual-task conditions, individual and gender-related differences in spatial processing should be considered a confound rather than a characteristic of multiple-task performance.

The generality of our spatiotemporal hypothesis should be examined in future work, but as noted above (and in Mäntylä, 2013), we would not expect gender differences in driving and related dual-task conditions, mainly because of their low demands on temporal processing. Driving and talking on a cell phone require minimal temporal coordination because this type of multitasking (a) involves only two component tasks, (b) is based on highly overlearned scripts and schemas (which guide the spatiotemporal coordination of component tasks), and (c) involves rather flexible time windows.

It is reasonable to assume that driving involves spatial processing for route navigation and lane maintenance, and we have no reason to question Strayer and colleagues’ conclusions about gender invariance in their dual-task conditions (and many other versions of multitasking). However, before rejecting our time-in-space hypothesis, it should be noted that route navigation and lane maintenance in simulated driving probably involve a different type of spatial processing than did the higher-level spatiotemporal coordination of component tasks in the Mäntylä (2013) study. Specifically, none of the four component tasks in that study were inherently spatial, and Mäntylä (2013) hypothesized that the temporal coordination of these (nonspatial) subtasks might be handled more efficiently if their complex temporal contingencies were represented in terms of higher-level spatial relations. Consistent with this line of reasoning, gender differences in spatial orientation (including navigation) are rather weak and mixed (e.g., Voyer, Voyer, & Bryden, 1995), and therefore the absence of gender differences in Strayer and colleagues’ driving studies does not provide a valid refutation of our spatiotemporal hypothesis.

In the Mäntylä (2013) study, gender differences in multitasking were fully mediated by spatial ability, but Strayer et al. (2013) interpreted these effects as “unwanted spatial confounds” (p. 809). In contrast, we believe that the existence of a mediator does not invalidate but, rather, explains the observed gender differences. Furthermore, although spatial ability is a more basic cognitive construct than video game experience, Strayer et al. considered experience in video games as a mediator of the male advantage in the Hambrick, Oswald, Darowski, Rench, and Brou (2010) study and spatial ability as a confound of our findings. Excessive computer gamers were excluded in the Mäntylä (2013) study, but it should be noted that video game experience and spatial ability are closely related constructs. Consistent with our main argument, several studies suggest that video game experience improves spatial attention and related cognitive functions (e.g., Bavelier, Achtman, Mani, & Föchner, 2012; Feng, Spence, & Pratt, 2007; Spence & Feng, 2010).

Finally, Strayer et al. (2013) presented an alternative view by suggesting that “a more productive approach in this research domain is to consider the association between individual differences in executive attention and the ability to multitask” (p. 809). It should be rather evident that this hypothesis was the starting point and one of the main conclusions of the Mäntylä (2013) study. Strayer et al. (2013) also used the absence of gender differences in executive functions as an explanation for “the apparent lack of gender differences in multitasking in the larger literature” (p. 810). Gender invariance in executive functioning does not necessarily justify a conclusion about gender invariance in multitasking, although individual differences in these two umbrella constructs are closely related.