Chapter 12 Towards the real world
The change detection literature is an inspiring example of how lessons from a task can inform real-world practice. Without knowledge of change detection results, some practitioners, such as coaches of team sports, likely subscribed to the naive view of visual perception and attention that we are simultaneously aware of the identities of all the objects in a scene, such that unless a player actually disappears or hides behind something or someone, players should know where everyone in front of us on the basketball court, or the soccer field, is at all times (Brian J. Scholl, Simons, and Levin 2004). Similarly, during driving many people seem to assume that they are aware of all hazards in their visual field.
While change blindness demonstrations have dispelled naive beliefs about visual awareness of change, still people assume that if they are actively attending to a moving object, they will be aware of its features. As we have seen (Section 10), this is not true. Tracking does facilitate change detection however, as found in a driving simulator study by Lochner and Trick (2014).
Very few empirical studies have established strong links between real-world situations and laboratory MOT tasks or its underlying abilities. Bowers et al. (2013) found that laboratory MOT performance did not predict driving test performance as well as either the Montreal Cognitive Assessment task, a trail-making task, or a useful field-of-view task. The aforementioned driving simulator study by Lochner and Trick (2014) found that drivers were more accurate at localizing which of multiple lead vehicles braked if it was a tracking target, but there was no advantage in terms of braking response time.
Mackenzie et al. (2022) used a multiple object avoidance (MOA) task where the user, in a task reminiscent of Asteroids (Section 1), controlled one of the balls with a mouse, trying to prevent it from colliding with the other balls. Strong correlations were found with years of driving experience and driving simulator performance. Some of the same authors also found that MOA performance correlated better with driving performance than conventional MOT (Mackenzie and Harris 2017). This may be because MOA includes motor control, which is necessary for driving, but is not required for MOT.
Some teams of researchers have repeatedly found evidence that MOT performance predicts in-game performance in soccer and other sports, and have also reported evidence that training on MOT tasks can enhance skill in sports. Unfortunately, the evidence is not strong (Vater, Gray, and Holcombe 2021). Given the poor record of computer-based training tasks (sometimes called “brain training”) in improving skills in other real-world domains (Simons et al. 2016), we should be skeptical that MOT training has benefits until rigorous evidence is provided.