Rethinking Distraction

Abstract

When one’s attention is directed away from the task at hand, especially when there is explicit motivation to maintain task engagement, it is tempting to conclude that the person has become distracted, implying a failure of attentional control or a lack of focus. Similarly, when a person fails to process information that they know to monitor for, such as a warning signal or a familiar hazard, it is easy to assume that they must have been distracted by something else, again failing to maintain their focus. Informed by recent findings concerning the control of attention, I argue that such a view of distraction is often misguided. Rather than reflecting a failure of the attention system, what is typically referred to as “distraction” in everyday life can in many instances be described as a predictable feature of a well-functioning attention system. This has implications for the responsibility we ascribe to people for being distracted, and what we try to do to help people cope with so-called distractions. For context, I apply this framework to our understanding of “distracted” driving, “distractions” in the classroom and at work, and inattention to hazards in work environments. Broader theoretical implications concerning the nature of distraction are discussed.

Advances in technology, and the impact of these advances on access to information, have both scientists (e.g., Weksler & Weksler, 2012; Strayer, 2015) and popular media (Alexander, 2022; Goldman, n.d.; Monroe, n.d.) decrying an “epidemic of distraction.” Drivers are distracted on the road (Strayer, 2015; Strayer & Drews, 2004; Strayer & Fisher, 2016; see also Stothart et al., 2015), students are distracted in the classroom (Hall et al., 2020; Kim et al., 2019; Seidman, 2005; Smallwood et al., 2007; Taneja et al., 2015), employees are distracted at work (Rosen & Samuel, 2015; Stich et al., 2018; Yin et al., 2018), and people are simply distracted from being “present” in their daily lives (Aagaard, 2016; Newcombe & Weaver, 2016). Directing attention to a text message on a smartphone while driving or during a class activity is a quintessential example of distraction in the modern age, and there is widespread belief that the degree to which people are distracted in their daily lives has reached unprecedented levels. The question then turns to why people are “distracted” in this way and what we can do as individuals and as a society to better “handle” such distraction.

Distraction in Real-World Situations

To help contextualize some of the arguments made in this paper, I will focus on three contexts for the allocation of attention that have been extensively studied in the more applied literature on the nature of distraction. Findings from this literature speak to some of the core issues that are exemplified by the “epidemic of distraction.” These contexts encompass distraction while driving, distraction in the classroom and at work, and inattention to workplace hazards. This section is by no means intended to offer a comprehensive accounting of the state of the literature concerning the control of attention in these real-world contexts, but rather to provide examples highlighting some of the ways in which the control of attention can be limited by apparent distractions.

Distracted Driving

Most traffic accidents are believed to be the result of human error (e.g., Theeuwes, 2021; Treat et al., 1977), often with respect to insufficient attention to some aspect of the demands of driving (e.g., Charlton & Starkey, 2013; Sabey & Taylor, 1980). In fact, with respect to collisions, in many cases drivers fail to react to the cause of the collision at all, rather than act too late (Guo et al. 2010). Drivers often fail to process a road sign that has changed along a familiar route (Martens & Fox, 2007) and may completely ignore a bicyclist approaching from an unexpected direction (Räsänen & Summala, 2000). Such failures of the perception of the driver can be attributed to the manner in which they allocated their attention (Theeuwes, 2021).

The biased competition theory of attention posits that attended stimuli are robustly represented in capacity-limited stages of perceptual information processing, while unattended stimuli are represented with considerably less fidelity, if at all (Desimone & Duncan, 1995; see also Mack & Rock, 1998; Most et al., 2001; Rensink et al., 1997). Presumably in many of the aforementioned driving situations, the driver was attending to something, just not what was situationally most important to the demands of safe driving. The object of attention in these cases might be the car radio, a text message, a conversation, or even just internal thoughts unrelated to driving (e.g., Nemme & White, 2010; Overton et al., 2015; Strayer et al., 2015, 2016; Yanko & Spalek, 2014); when attention is directed in this capacity, we might say that the driver has become distracted. Indeed, distracted driving can be formally defined with respect to inattention occurring when drivers focus on an activity other than driving (National Highway Traffic Safety Administration, 2012; Overton et al., 2015).

Consistent with an attentional account of failures to process information pertinent to safe driving, drivers are less efficient at detecting potential hazards when the attentional demands of driving are high, for example in the case of inexperienced drivers or drivers navigating an unfamiliar environment (Kahana-Levy et al. 2019; Underwood et al. 2002). This may be related to a more general phenomenon by which people are prone to miss important information that is directly in their line of sight (“looked but failed to see”), particularly when engaged in a secondary task that is attentionally and/or cognitively demanding (Wolfe et al., 2022). In this kind of situation, we might not be inclined to say that the driver is necessarily distracted, but at the least, their attention is in the same way directed elsewhere, highlighting the critical role for attention in processing safety-relevant information while driving.

An attentional account of failures to process safety-related information while driving helps to explain why a significant component of cell phone-related impairments in driving can be attributed to the cognitive demands of cell phone use (Brookhuis et al., 1991; Wikman et al., 1998), and why hands-free cell phones do not eliminate the detrimental impact of cell phone use on the situation awareness of the driver (Drews et al., 2008; Horrey & Wickens, 2006; McCarley et al., 2004; McEvoy et al., 2005; Strayer & Drews, 2007). With that said, when distractions cause the eyes to deviate from the road, the ability to detect the sudden onset of a potential hazard may be particularly reduced (Wolfe et al., 2019). Although the direst consequences of distracted driving in many cases involve instances of looked but failed to see, distracted driving is associated with generally less efficient and situationally-responsive driving, which increases the risk of an accident and disrupts traffic flow (e.g., Beede & Kass, 2006; Cooper & Strayer, 2008; Cooper et al., 2009; Shinar et al., 2005; Stavrinos et al., 2013; Strayer & Drews, 2004).

Distraction in the Classroom and at Work

The use of technology in both the classroom and on the job is commonplace, and there is substantial evidence that the use of technology can serve as a distraction from learning and the completion of work. Many students report feeling like they overuse smartphones and that smartphone use served as a distraction during class (e.g., Dontre, 2021; Ko et al., 2015). The cost of a technology-related interruption can quickly compound, resulting in a 20–30 min delay until the person reengages the interrupted task (Chen et al., 2020; Dontre, 2021; Gazzaley & Rosen, 2016). In the workplace, when such engagement with media is unrelated to the content of the work, it can be referred to as “cyberloafing” (Varol & Yıldırım, 2019).

Media multitasking in the classroom is associated with poorer learning outcomes and scholastic performance (Carrier et al., 2015). Even the media content on another student’s laptop can serve as a distraction to nearby students, detrimentally impacting their comprehension and retention of instructed material (Hall et al., 2020). Strikingly, the extent to which an individual engages with social media is negatively correlated with grade point average, suggesting that social media use competes with the amount and quality of study time (Kirschner & Karpinski, 2010). In addition to the use of media, daydreaming/mind-wandering (Kane et al., 2017; Ralph et al., 2017; Smallwood et al., 2007) and conversations with peers (Boice, 1996; Frisby et al., 2018; Johnson et al., 2017; Seidman, 2005) have also been described as sources of distraction with respect to students in the classroom and workers in the office.

Inattention to Hazards

Some professions involve routine exposure to specific hazardous situations. For example, construction workers may be exposed to heavy machinery moving in their vicinity or ledges without railings over which they could fall, and electricians may be exposed to materials that would be dangerous to touch. Routine exposure to a particular hazard is associated with reduced vigilance for that hazard (e.g., Anderson et al., in press; Kim et al., 2021b, 2023b, 2023c). Although workers engaged in hazardous professions receive periodic safety training and consistent reminders about the importance of workplace safety, frequently exposed hazards often go ignored, which significantly contributes to workplace accidents (Anderson et al., in press; Herslund & Jørgensen, 2003; Lee & Kim, 2022; Ortiz et al., 2000; Weyman & Clarke, 2003). The more workers are exposed to workplace hazards without detrimental consequences, the less likely they are to direct their attention to these hazards and the warning signals associated with them (Anderson et al., in press; Kim et al., 2021b, 2023b, 2023c). Instead, experienced workers often spend more time focusing their attention on other things, including the performance of their immediate work task, non-safety-related information in their environment, and internal thoughts. Given that these workers were aware of the importance of maintaining their attention on safety-related information, it can be argued that failures to do so are reflective of distraction by other, less important things.

Summary

Although drivers understand the importance of carefully attending to traffic signals, the roadway, and vehicular and pedestrian traffic, they are prone to inattention to such information as attention shifts to processing other things that are unrelated to driving. Students and employees spend considerable amounts of time attending to non-work-related material when they are expected to maintain their focus on work or study, resulting in reduced productivity/performance and in some cases missing important information. Workplace accidents can result from inattention to critical safety-related information, which may be the result of distraction from task-irrelevant stimuli and off-task thoughts, or too strong of a focus on (and thereby arguably some measure of distraction from) a work task. In each of these situations, there is a discrepancy between what people know they “should” pay attention to and how they actually allocate their attention, and as such it is tempting to characterize each as a failure of attentional control.

Distraction in the Basic Science of Attentional Control

In the basic science domain of experimental psychology, distraction is defined narrowly and objectively. The centerpiece of the definition of distraction here is an impairment in the performance of a task that can be measured and attributed to a specific stimulus. The task in which such a performance impairment is assessed is one that is unambiguously instructed to the participant. In a typical experiment, participants search for a specific stimulus called a target, and they are instructed in advance concerning what this target looks like, often with the provision of disambiguating examples. Then, the speed and accuracy with which they are able to identify the target repeatedly over a variety of instances of performing the task, called trials, are measured. All trials contain objects drawn from a core set of possible objects with one exception, which is a special stimulus referred to (often presumptively) as the distractor. The distractor typically only appears on a subset of trials, allowing the researcher to (a) compare performance on trials that differ only with respect to the presence or absence of the distractor and/or (b) compare the frequency or depth with which the distractor is processed (e.g., with respect to how often participants look at it) relative to a simultaneously-presented non-target stimulus drawn from the base set. If target report is slower and/or less accurate in the presence of the special distractor and/or the distractor is processed more robustly than other non-targets, the “distractor” can be said to be just that—something that causes distraction. Given the (lack of) relationship between the distractor and the requirements of the task, it is typically assumed that any observed distraction is beyond the control of the participant (i.e., involuntary).

A large group of scientific experts on the control of attention, which includes the author of this review paper, offer a consensus definition of the phenomenon of distraction and the stimulus defined as a distractor that it is tied to. Distraction is defined as the “Processing of irrelevant information that impairs search performance.” (Liesefeld et al., 2023; p. 22), with irrelevance defined with respect to the information needed to localize a target and make a required response (which is generally in accordance with explicit task instruction). Further, these authors recommend that scientists “Reserve the word distractor to refer to the stimuli for which the potential to cause distraction is examined,” (Liesefeld et al., 2023; p. 23, italics maintained from original version), contextualizing this with respect to the kind of manipulation of stimulus set described in the preceding paragraph.

Using this basic experimental psychology approach, scientists are often interested in understanding the kinds of stimuli that tend to cause distraction and the conditions under which distraction may be more or less severe. For example, the critical distractor might differ with respect to its physical salience (e.g., high-contrast, uniquely colored objects; Theeuwes, 1992, 2010) or reward history (e.g., something that predicts a monetary reward, positive social feedback, or water when thirsty; Anderson, 2016a; Anderson et al., 2011; De Tommaso & Turatto, 2021; Kim & Anderson, 2020b), leading to the conclusion that physically salient objects and objects with learned value tend to cause distraction. Alternatively, physically salient stimuli have been shown not to result in a performance cost in certain situations and can even be processed less robustly than other non-target stimuli (Gaspelin et al., 2015; Gaspelin & Luck, 2018), leading to the conclusion that it is sometimes possible to suppress distraction. Likewise, previously reward-associated stimuli are less distracting when an individual feels threatened (Kim & Anderson, 2020a, 2020c), leading to the conclusion that there is an antagonistic relationship between the processing of aversive information and distraction by appetitive stimuli.

In all of these cases, the experimenter decides what constitutes appropriate task performance and, by extension, distraction. It is assumed that the participant faithfully follows task instruction as articulated in the study materials and genuinely tries to resist distraction, in many cases to no avail. Given the strict definition and the highly reductionist approach to isolating distraction to a single stimulus or class of stimuli, the “cost” of distraction in experimental psychology is often quite small, in many cases on the order of tens of milliseconds in time to report the target (e.g., Anderson et al., 2011; Theeuwes, 2010). This can be contrasted with the consequences of distraction often of interest in the more applied attention literature, which tend to focus on felt costs tied to complex behavior (e.g., an accident in a virtual environment that would have had significantly detrimental consequences if it were to have happened in the real world) or at least an elevated probability of incurring such a cost (e.g., riskier driving behavior, such as following a car at an unsafe distance).

A parallel basic science literature that is pertinent to our understanding of distraction, particularly with respect to how it is typically defined in real-world contexts, is the science of mind wandering. In a typical mind wandering experiment, individuals are periodically probed concerning whether their immediately preceding thoughts were on-task or off-task, and such subjective reports can be related to task performance leading up to the report (e.g., Randall et al., 2014; Thomson et al., 2014) and/or brain activity evoked by task-relevant stimuli (e.g., Fox et al., 2015). A common finding is that wind wandering is associated with less efficient task performance and increased brain activation in the default mode network (e.g., Fox et al., 2015; Randall et al., 2014; Smallwood et al., 2007). These outcomes can be thought of as the distracting influence of internal thoughts with respect to task-related information processing. Although the primary focus of this review paper will be on distraction evoked by stimuli in the environment (in keeping with the focus of the special issue), any consideration of distraction in the real world is not complete without some consideration of internal sources of distraction linked to the construct of mind wandering. As such, some consideration will be given to internal sources of distraction as well, and all of the arguments made in this paper will be pertinent to both internal and external distractions.

Rethinking Distraction: An Integrated Perspective

Few would debate that in everyday life, people can spend significant amounts of time directing their attention in ways that are detrimental to personal safety and occupational or scholastic performance, and perhaps detract from quality of life more broadly. The idea that the distraction captured by the real-world contexts described in the preceding section can have negative consequences for a person is uncontroversial and needs no argument. A key issue then becomes how we think about such apparent distraction—what exactly happened and why?

In the basic science literature, distraction is typically viewed as a failure of attentional control in which volitional processes are unable to overcome biases in what information is selectively processed, to the detriment of the person. The content of this special issue speaks for itself in that regard. Volitional, goal-directed attentional control and the processes that lead to distraction are two forces that battle for control over the information content of the mind. In the context of academic research, we might focus on what makes something like a billboard or text message alert attention-grabbing, characterizing the limitations of goal-directed attentional control, or we might explore the conditions under which attention to task-irrelevant stimuli can be reduced in severity, offering insights into the fundamental mechanisms of attention that a person can leverage in order to more effectively (albeit imperfectly) restrict attention to the information that is pertinent to the task at hand. These are worthwhile endeavors, which are at the core of my own research program, but they still leave us with the broader question of what it really means to be “distracted” by something.

As contextualized above, when thinking about distraction in the real world, it can be easy to assume that distracted people were either insufficiently attentive to the task at hand to their own fault or they were the victims of attention-grabbing tactics or realities that overwhelmed their attention systems. Maybe distracted students simply do not try hard enough to maintain their focus on their course instructor, and maybe distracted drivers are prone to succumbing to the temptation to direct their attention to something other than the roadway in front of them (e.g., an attention-grabbing billboard or the unchecked message on their smartphone). On the most fundamental level, there seems to be widespread agreement among scientists and the general public that distraction is a categorically distinguishable experience that happens to people, and a bad one at that, in need of explicit countermeasure. After all, such experiences are becoming “epidemic,” and as the title of this special issue states explicitly, distraction is something we are burdened to manage or “handle.”

Here, I will make an empirical, theoretical, and philosophical case that these default assumptions about distraction are at best misleading, and in some cases probably untrue. People generally do not try but fail to resist distraction; in most cases, their attention system merely operates as we should expect it to given the information they have been exposed to, which the individual is quite content with (at least in the moment). This is true of attention in both the real world and in the lab. We just call it distraction when the consequences of attention allocation seem detrimental, either in retrospect or from the vantagepoint of a third-party observer (e.g., a scientist designing an attention experiment, an employer who values productivity). To ascribe responsibility to a “distracted” person in many cases belies any principled account of their volitional contribution to the contents of their attention, barring egregious instances of sustained behavioral neglect (e.g., texting while driving). When it comes to attempts to curb the epidemic of distraction, a person will probably never learn to handle distraction better than they already do in any direct sense; again, in many cases distraction is simply a product of how the attention system of a person works in the context of the situations in which they find themselves. Blaming the situation (e.g., smartphones and the immediate access to information that they provide) and considering what might be done to remove people from such situations (e.g., policies or strategies to limit smartphone use), although not entirely unproductive, ignore the fact that the propensity towards “distraction” is fundamental.

Below, I will offer six arguments that collectively serve as the backbone for an alternative view of attentional control that normalizes the phenomenon of distraction. At the heart of these arguments is a framework for understanding the contents of attention as the product of interactions between attention and memory, with these interactions constituting “control.” This framework argues in favor of the idea that distraction in the real world is often what we should expect from an adaptive attention system rather than a failure of attentional control, and provides recommendations for how to more meaningfully combat the so-called “epidemic of distraction.”

The Link Between Attentional Control and Volition

The attention system itself lacks volition. Although a person can, of course, direct their attention to something as a result of a voluntary process (e.g., attending to the left side of space to check if a certain person is standing there), the allocation of attention is actualized through a process that is not itself under volitional control. Debates concerning the nature of free will aside (Dennett, 1984, 2004; Sali et al., 2018), attention to the contents of the outside world is best characterized as weights applied to both sensory and non-sensory (i.e., internally generated thought) input from a variety of mechanisms in the brain. As the next section will explore in more detail, some of these weights are the product of mental processes that are under volitional control and some are not. The attention system is subservient to the collective influence of these mental processes, such that which source of input “wins” the battle for attention (e.g., whether a student attends to their instructor or their smartphone) is not itself a choice that a person gets to make.

From a phenomenological perspective, directing attention to something is unlike the initiation of a voluntary motor movement (e.g., Libet et al., 1983; Wolpe & Rowe, 2014). When trying to find someone in a crowd, an observer likely never actually thinks about what their attention system is doing, nor do they try to do something intrinsically attentional while searching. They likely just think about the person they are looking for—what they look like—and attention just sort of happens from there. The act of attentional selection is therefore not tantamount to a choice. Drivers that are distracted by billboards or other salient advertisements generally do not enter the roadway intending to direct their attention to such stimuli, and students and workers can easily find themselves shifting their attention to something they did not come to class or the office intending to focus on. Such individuals were likely thinking about what they came to the situation intending to focus on, but there came a moment where their attention did not follow from such thinking.

Consistent with this perspective on attention, awareness for how attention has been directed is strikingly poor. When people view naturalistic images, their later memory for how and where they directed their attention cannot distinguish between their own patterns of looking and the patterns of another person (Vo et al., 2016; see also Horowitz & Wolfe, 1998). People will attentionally process a salient but task-irrelevant non-target (“distractor”) in a traditional attentional capture task and exhibit a limited ability to identify when such “distraction” occurs (Adams & Gaspelin, 2020, 2021; Theeuwes et al., 1998; see also Anderson & Mrkonja, 2021, 2022). It is possible to shape how people direct their attention when viewing a scene by punishing certain attentional “actions” (defined with respect to eye movements), with learning-dependent change occurring in the absence of awareness concerning what causes the aversive outcomes (Anderson, 2021d). Directing attention can in many cases reflect a cognitive process for which people possess limited awareness, which is at odds with the idea that directing attention is itself the product of a conscious choice. People generally have robust awareness of what they intended to pay attention to, but an impoverished understanding of what they actually directed their attention to.

Attentional Control as Memory-Dependent Biases in Information Processing

As asserted at the outset of the prior section, it is intuitively obviously that people can exert some degree of volitional control over the contents of their attention (see, e.g., Gmeindl et al., 2016). Were this not the case, employees would hardly ever get any work done and drivers would seldom have their attention on the road at all. From the standpoint of volition, we can think about volitional influences on attention as biasing signals from activated memory states, with the contents of activated memory being under some degree of volitional control. In the terminology of attention, we might call these memories templates (Carlisle et al., 2011; Woodman & Arita, 2011). Both representations brought online in working memory and activated long-term memory representations have been argued to serve as input to the attention system (Woodman et al., 2013). Uncontroversial is the idea that these activated memory representations can increase the attentional bias applied to a stimulus that sufficiently matches the contents of the memory or template (Anderson & Folk, 2010; Desimone & Duncan, 1995; Woodman et al., 2013). More controversial is whether activated memory can be used to suppress the representation of known-to-be-irrelevant stimuli and at what stage of information processing such suppression occurs (e.g., Arita et al., 2012; Becker et al., 2015; Forschack et al., 2022; Gaspelin et al., 2015; Gaspelin & Luck, 2018). It is clear, however, that we can to some degree control what we direct our attention to by influencing the contents of our memory. With respect to such volitional influences, the memory system thinks, and attention reacts. This naturally leads to a discussion of the role of memory in the control of attention, and how volitional that influence really is.

If only volitionally activated memories influenced attention, we might have some reason to characterize distraction as a failure of attentional control. However, not all memories that are brought to bear in the control of attention are brought online volitionally; in fact, it is likely that the majority are not (see Anderson, 2021a). The reward history and punishment history of objects influence the degree to which the attention system biases these objects, even when they are known to be irrelevant to the task at hand (e.g., Anderson et al., 2011; Anderson & Halpern, 2017; Kim et al., 2021a; Schmidt et al., 2015; see Anderson et al., 2021, for a review) and even when directing attention to these objects is explicitly counterproductive when it comes to maximizing rewards (e.g., Kim & Anderson, 2019a; Le Pelley et al., 2015) or avoiding aversive outcomes (e.g., Anderson & Britton, 2020; Kim & Anderson, 2021 Nissens et al., 2017). Similar influences on attention have been observed with respect to habit learning (e.g., how frequently an object has served as a searched-for target or how frequently a searched-for target has appeared in a particular region of space; Kim & Anderson, 2019a, 2019b; Jiang et al., 2013; Kyllingsbaek et al., 2001; Shiffrin & Schneider, 1977; Sha & Jiang, 2016) and statistical learning (e.g., how frequently an object has been irrelevant to the task or how frequently a task-irrelevant object has appeared in a particular region of space, resulting in a reduction in the biasing signal applied; Britton & Anderson, 2020; Kim & Anderson, 2022; Kim et al., 2023a; Stilwell et al., 2019; Wang & Theeuwes, 2018a, 2018b). The content of thought when mind wandering is often linked to the broader concerns of the individual (e.g., Kopp et al., 2015; McVay & Kane, 2010; Mooneyham & Schooler, 2013; Poerio et al., 2013). The common theme, it would seem, is that memories have consequences for how we process information, whether we want them to or not.

This alone has implications for how we characterize distraction. If a person has strong memories concerning reward, the influence of these memories would be expected to dominate attention when they are in an active state. This has been argued to help explain why it is so difficult for individuals who struggle with drug dependence to ignore drug cues (Anderson, 2016c, 2021b; Field & Cox, 2008), but it can also lend insight into why billboards depicting attractive and/or desired content and gateways to pleasurable media experiences such as smartphones and media apps can be strong competitors for our attention. The same can be said for memories for aversive experiences and corresponding attentional biases, with the objects of fear dominating information processing (e.g., Clauss et al., 2022); as will be further explored later in the text, inattention to workplace hazards is frequently defined by a distinct lack of negative consequences and attention to such hazards is likely to increase following an accident (Anderson et al., in press; Kim et al., 2023b). Habits are powerful influences on behavior, and the attention system has been argued to be subject to habit-like prioritization (Anderson, 2016b; Hikosaka et al., 2013; Jiang, 2018). When such memories are sufficiently strong, it might be a tall task for the influence of memories that are brought online volitionally to overshadow the influence of these competing memories.

The idea that different perceptual input compete for attention has been at the heart of theories of attention for decades (e.g., Corbetta & Shulman, 2002; Desimone & Duncan, 1995; Serences & Yantis, 2006). I would argue that, beyond the mere strength of the stimulus input (i.e., physical salience; Itti & Koch, 2001; Theeuwes, 1992, 2010), the focus of attention comes down to a battleground of memory-dependent influences, and volitionally activated memories are only one of multiple such influences. The question concerning distraction then becomes whether a person can “will” a volitionally activated memory to victory and thereby determine what they pay attention to. As the logic goes, if a driver just wants to keep their attention on the roadway badly enough, or a student is sufficiently motivated to learn from their instructor, that is where their attention will remain.

This humanistic prospect of attentional self-empowerment is not without neurobiological resonance. High motivation can strengthen activated memory signals and further bias information processing in favor of the object of motivation (e.g., Etzel et al., 2016; Jimura et al., 2010; Locke & Braver, 2008; Padmala & Pessoa, 2011; Pessoa & Engelmann, 2010). Mind wandering occurs less frequently when individuals are highly motivated to perform a task well (Antrobus et al., 1966; Brosowsky et al., 2020; Unsworth & McMillan, 2013). Such scientific findings square with the intuition that people tend to be better focused on a given task when motivated. The idea that such motivational influences could be strong enough to consistently overshadow the influence of any other memory seems untenable in its strongest form (e.g., Anderson et al., 2011; Anderson & Britton, 2020; Kim & Anderson, 2019a; Esterman et al., 2013, 2014; Fortenbaugh et al., 2017), though, at which point it becomes a matter of the degree and scope of control. However, even if one assumes a dominant role for volitional processes, as will be explored in the next section, the expectation of consistent dominance would place unrealistic burdens on the mental capacities of a person.

Goals and Intentions in Context

If someone would just consistently think about what they wanted to engage with (pay attention to) and were sufficiently motivated, given the issues outlined in the previous section, it would seem that “distractions” would be largely if not completely preventable. As will be explored in a later section, there is good reason to question whether people would actually desire this. But even if we assume that a person genuinely wanted to maintain their attention on something specific, such as the road in front of them while driving or their work task, the question arises as to whether people would be capable of actually doing this. The results of a number of attention studies seem to converge on the idea that such sustained attention is unrealistic.

Identifying task-relevant stimuli can be computationally demanding even with attention when a scene is sufficiently cluttered. When monitoring for pedestrians while driving or for potential workplace hazards, we may direct our attention near to such stimuli and erroneously conclude they were nowhere to be found, or we might actually even attend to them without accumulating enough perceptual evidence to recognize them before shifting attention elsewhere. This tendency to look but fail to see a pertinent stimulus has been referred to as normal blindness (see Wolfe et al., 2022), and it suggests that even when the focus of attention is optimal for a particular task demand and a person is not particularly distracted, important information may be missed. Normal blindness is especially likely to occur under conditions in which relevant stimuli are low in prevalence, as may be the case for rarely experienced roadway hazards (Kosovicheva et al. 2023), as observers typically shift their criterion for identifying a stimulus as relevant in order to facilitate more efficient rejection of irrelevant information (Wolfe et al., 2007). Although failing to recognize a pertinent stimulus clearly constitutes an undesirable situation, heavily scrutinizing task-irrelevant stimuli repeatedly can be inefficient and heighten the risk that actually pertinent stimuli will be insufficiently processed; observers must search in a manner that balances each of these considerations, which is challenging to do.

Even what might be said to constitute optimal attentional control is not without limitations. Allocating attention to stimuli in accordance with task goals is an effortful mental process (Anderson & Lee, 2023), as is maintaining vigilance in a sustained attention task, which is additionally stressful (Warm et al., 2008). As with physical effort, attentional effort is subject to what appears to be fatigue. People wax and wane in their ability to efficiently process task-relevant information over time, to some degree zoning in and out of focus (e.g., Esterman et al., 2013, 2014; Fortenbaugh et al., 2017). The same is true for trying to restrict attention to a searched-for target while avoiding “distraction” by a salient but task-irrelevant stimulus (Leber, 2010). This is to be expected given that the same sort of waxing and waning is evident in the fidelity of working memory representations (e.g., Adam et al., 2015, 2018; deBettencourt, 2019; Murray et al., 2011; Robison & Unsworth, 2019), and maintaining information in working memory reflects a canonically effortful mental task (e.g., Clay et al., 2022; Kool et al., 2010; Vogel et al., 2020; Westbrook et al., 2013, 2020). The strength of the biasing influence of volitionally activated memory on attention is naturally going to fluctuate, and when this biasing influence is at its lower points, an individual will be particularly susceptible to other, more involuntary influences of memory on attention, including broader concerns in the case of mind wandering (see Kopp et al., 2015; McVay & Kane, 2010; Mooneyham & Schooler, 2013; Poerio et al., 2013).

Compounding these fluctuations is the reality of switch costs. In the course of everyday life, we routinely need to switch between tasks and corresponding rules and expectations, or our expectations within a task may change. When these changes occur, different memory representations need to be brought online to guide attention and task performance more broadly. For example, employees need to switch between different work tasks and even sub-components of a given work task (e.g., arguments and ideas within a document they are writing), and students need to switch between attending to their instructor and various course materials, and lectures routinely shift in the topic of focus. There is a wealth of research demonstrating that such switching comes at a cost: the biasing influence of activated memory is initially weak when the contents of activated memory needs updating (e.g., Allport et al., 1994; Leber et al., 2008; Rogers & Monsell, 1995). There are a variety of theories concerning why this is (e.g., Vandierendonck et al., 2010), but the important point here is that the attention system will routinely be more vulnerable to distraction with every shift in task-related demands (see, e.g., Belopolsky et al., 2010; Lien et al., 2014).

Apart from these basic fluctuations in the control of attention that every person faces is the issue of individual differences in the strength with which activated memory can be brought to bear in the control of attention. It is well documented that people vary considerably in the amount and fidelity of information they can maintain in working memory, or their working memory capacity (e.g., Fukuda et al., 2010; Oberauer et al., 2016; Unsworth et al., 2014). As would be expected given the relationship between the control of attention and activated memory, working memory capacity and the ability to restrict attention to a task-relevant stimulus in the face of a potentially distracting task-irrelevant stimulus are correlated (Anderson et al., 2011, 2013; Anderson & Yantis, 2012; Fukuda & Vogel, 2009, 2011; Gaspar et al., 2016). Individuals who have a lower working memory capacity are also more susceptible to mind-wandering during performance of an attention-demanding task (McVay & Kane, 2009, 2012; Unsworth & McMillan, 2013). We all have fundamental limitations in our ability to maintain attention on any one thing, but these limitations are more severe for some people than they are for others in a manner that would be expected given that the control of attention is fundamentally memory-dependent. Individuals with a relatively low working memory capacity will be particularly vulnerable to other, more involuntary memory-dependent influences on attention.

Further Limitations on Attentional Control: Accounting for the Time Course of Bias Signals

Even when the influence of volitionally activated memory is at its strongest, there is another important limitation to the control of attention that must be considered and it exists in the dimension of time. The influence of volitionally activated memory, more frequently referred to in the attention literature in the context of top-down and/or goal-directed attention (Wolfe, 2020, 2021; Wolfe & Horowitz, 2017), is relatively slow compared to the influence of both the physical salience of stimulus input (Donk & van Zoest, 2008; Godijn & Theeuwes, 2002; Theeuwes, 2010; van Zoest et al., 2004) and several other memory systems that exert a more involuntary influence on the control of attention (e.g., Anderson et al., 2011; Anderson & Kim, 2019; Kim & Anderson, 2022). In the case of the latter, this is likely the product of the path that neural communication flows between the perceptual system and the pertinent memory system, which is relatively longer in the case of volitional memory (e.g., Anderson, 2019). The upshot is that there will be a brief period of time in which more involuntary influences dominate the control of attention, before more volitional influences can even be brought online.

Alone, this does not add much to the developing argument. It might serve as a meaningful qualifier, but after the brief period of time that the influence of volitionally activated memory on attention needs to come online (which may be on the order of a fraction of a second; e.g., Donk & van Zoest, 2008; Godijn & Theeuwes, 2002; van Zoest et al., 2004), the story would simply revert to where we left off in our reasoning in the prior section. We might see a greater propensity for periodic little breakthroughs in the current focus of attention, potentially exacerbating the influence of switch costs, but little else. Little else, however, assumes that these fleeting moments of off-task attention or “distraction” are themselves inconsequential. Given the relationship between attention and the information content of the mind that is the core of modern attention theories (Corbetta & Shulman, 2002; Desimone & Duncan, 1995; Serences & Yantis, 2006), it stands to reason that these episodes of fleeting distraction could invite further scrutiny of the attended information. The object of distraction is brought to the forefront of the mind (even though the person may not be aware of the fact that they are explicitly attending to it, given the relationship between attention as a mechanism and awareness) and can then activate further associated memories that might motivate behavioral engagement. Indeed, theories of attentional control link involuntary attentional orienting toward reward-associated stimuli with the activation of approach behavior tendencies (Anderson, 2017a; Anderson et al., 2016; Kim & Anderson, 2019c). With respect to our real-world examples of distraction, perhaps a student’s attention is briefly directed to their phone during class, which reminds them of desirable information that might be newly available for viewing and motivates actual use of the phone. Such a pattern would explain why a distraction can result in 20–30 minutes elapsing before the task from which attention was withdrawn is reengaged (Chen et al., 2020; Dontre, 2021; Gazzaley & Rosen, 2016). Likewise, a driver’s attention may be initially drawn to a billboard in a moment in which their focus on the road was waning a bit, only to linger there once they recognize the desirable information content and it comes to occupy the contents of their activated memory. This leads naturally into the next section, in which I take a broader perspective on what people are motivated to pay attention to.

Reward, Motivation, and What a Person Really Knows and Wants

Distractions are not really distractions if a person either wants to attend to the contents of a “distraction” or is at least quite content to. As simple as this sounds, it has significant explanatory power for a wide array of real-world distractions. Given the central role of memory in the control of attention, we need to think less about how objectively beneficial it is to have attention remain focused on any one thing and more about what a person’s experience tells them is worth paying attention to.

As described earlier, we know that reward learning (e.g., Anderson et al., 2011, 2021; Anderson 2016b) and habit learning (e.g., Jiang et al., 2018) have involuntary influences on attention, and that the contents of mind wandering can be influenced by current concerns (Kopp et al., 2015; McVay & Kane, 2010; Mooneyham & Schooler, 2013; Poerio et al., 2013). If a person has routinely directed their attention to something in the past or has been rewarded for doing so, their attention system develops an involuntary tendency to repeat such patterns of attention allocation. Conversely, if someone has not experienced detrimental consequences for ignoring something, their attention system has little if any reason to prioritize that information and will be apt to prioritize something else with greater interest value; this sort of learned ignoring or deprioritization has been argued to contribute to the ignoring of specific workplace hazards that can be observed with high levels of exposure to such hazards (Daalmans, 2012; Kim et al., 2021b, 2023b, 2023c; Makin & Winder, 2008). Indeed, there is considerable evidence that some degree of mind wandering is intentional (Seli et al., 2016a, 2016b, 2016c), and individuals should be expected to engage in periodic intentional mind wandering when their experience tells them that doing so is relatively inconsequential.

Many drivers have briefly directed their attention away from the road and onto billboards or in-car media hundreds of times without any negative consequences. To the contrary, drivers may have a more mentally stimulating experience when they do attend to such media and are less likely to miss out on an opportunity, such as a social gathering communicated via text message or a good place to grab a bite to eat indicated by a billboard. A driver may know cognitively that they should maintain their attention squarely on the road at all times, but their experience tells them otherwise.

In the classroom and at work, the consequences of distraction on productivity are often only felt after some period of time, especially with respect to concrete indicators of productivity like a performance review or exam grades. And engaging with information content that is likely more interesting and pleasurable than the contents of the task at hand may in the moment be viewed as the more desirable state. In fact, from the standpoint of associative learning, it may be quite difficult for students and employees to meaningfully link summative feedback like grades and a performance review to episodes of distraction, since many other occurrences in their day are probably similarly predictive given the protracted period of time under consideration (see Anderson, 2021d). Certainly from the standpoint of their experiential memory and perhaps from the standpoint of their volition, many distracted individuals were probably quite content to have had their attention directed as it was.

As for employees who tend to ignore specific hazards, they have in many cases been exposed to the same hazard or warning alarm over several hundred instances throughout their career, each time with no consequence to their health or safety. Under such conditions, they have no experiential reason to believe that they are actually in any danger, and the warning alarm or hazard may be quite loud and annoying. Workers will often get their work done a little faster when they simply ignore the alarm/hazard and focus more intently on their work task, and they might easily feel inclined to direct their attention to other things, like what they want to do over the coming weekend (mind wandering), since they can seemingly do so without detrimental consequence. Anecdotal evidence points to a rapid shift in attentional prioritization of a hazard when a person experiences an accident or near-accident, and recent experimental evidence supports this idea using a virtual reality intervention (Kim et al., 2021b, 2023b, 2023c), suggesting that felt consequences can shift attentional priorities in favor of otherwise ignored stimuli.

Not only does a memory-dependent framework for attention lend insight into why a person may in many cases be vulnerable to distraction, but it further suggests that instances of distraction may in fact reflect desirable states. When a person’s experience tells them that ignoring something or periodically attending to something other than the information pertinent to the completion of a particular task produces a more desirable outcome, we should expect them to direct their attention in this way. As the prior sections illustrate, simply intellectually knowing or believing otherwise has significant limitations in its ability to counteract the biasing influence that a person’s broader memory system places on their attention system. Distraction is not really distraction when it is not in conflict with a person’s current goals (Liesefeld et al., 2023), and what exactly constitute a person’s true goals with respect the control of attention is not easily reducible to a single ongoing task.

The Adaptive Nature of Involuntary Attentional Control

Why have an attention system so seemly prone to distraction, and why not have a system in which our current motivations and what we intellectually believe to be true in the here and now could more reliably dictate the focus of attention? In this final argument, I will grapple with the question of whether we would really want to have an attention system that operates any other way than what I have just described in the preceding arguments.

In the interest of brevity, I will only highlight my viewpoints on this issue here, as a more in-depth treatment of the topic has already been published (Anderson, 2021a). The memory-dependent system I am proposing dramatically limits the need for controlled and effortful cognitive processing—much of attentional control is automated (see Anderson, 2018; Theeuwes, 2018). The detection of potential opportunities or threats will be facilitated, which is particularly important in light of limitations in information processing outside of the focus of attention (Anton-Erxleben & Carrasco, 2013; Desimone & Duncan, 1995; Mack & Rock, 1998; Most et al., 2001; Rensink et al., 1997)—we often need to attend to something in order to know what it is and thus how we should respond to it. Through memory-dependent processes, our attention system is wired to selectively process information in a way that downweights inconsequential information and upweights the information pertinent to the quality of the outcomes we experience. Assuming that prior experience is a good predictor of future outcomes, this sort of attention system can be quite efficient, although there will be inherent vulnerabilities compared to a system demanding constant volitional control. For every time that we are “distracted,” it can be easy to miss the remarkable efficiency with which we normally process the abundance of information in the environment.

Summary and Conclusions

What a person directs their attention to is the product of a memory-dependent process that is not itself under volitional control. Although certain influences on the control of attention are under voluntary control, many are not, and these influences compete with each other to determine what a person will direct their attention to. Even when a person is highly motivated to intentionally focus on something specific, several limitations in human information processing make it unrealistic for the person to consistently maintain such focus. In many cases, “distraction” may not be inconsistent with a person’s actual goals and intentions, and it may be quite adaptive for their attentional system to prioritize stimuli on the basis of considerations that extend beyond the goals and intentions tied to the performance of a specific task. For all of these reasons, many cases of “distraction,” in both the real world and in the context of basic laboratory experiments, reflect an adaptive response that can hardly be described as constituting a failure of attentional control.

From the arguments outlined in this paper, I would posit that the study of distraction is not an exercise in how the attention system fails. It is an exercise in understanding how it works. The allocation of attention can be situationally maladaptive, and the key is to understand why in a broader context. In most cases, the attention system does not “fail,” it merely “tries” to operate in a manner that is broadly adaptive but not without limitations.

Trying to tie the occurrence of “distraction” down to a particular moment in time (see Anderson, 2021c), or designating a discrete event as a case of distraction, is unnecessarily arbitrary and unlikely to be productive. More productive would be to ask what people direct their attention to when and why. Understanding why someone directs their attention the way that they do necessitates an understanding of the contents of their memory, which extends beyond what the person might be actively thinking about with respect to their current task.

Expecting the control of attention to be under the constant influence of volitionally activated memory, as reflected in momentary goals, is unrealistic. Effective means of curbing attention should encompass implicit learning, leveraging the more involuntary influences on attentional control (see Anderson et al., in press). It is unlikely that there will be any remedy for how to “handle distraction” in a broad sense, as a propensity towards “distraction” is a fundamental aspect of how the human attention system prioritizes information. What the attention system prioritizes will always have a tendency to be influenced by stimulus-specific (e.g., Anderson, 2017b; Clement et al., 2023) and context-specific memories (e.g., Anderson, 2015a, 2015b; Anderson & Kim, 2018; Gregoire et al., 2021), which any attempt at training someone to better handle distraction should target. Simply expecting someone to “try harder” to restrict their attention to something task-relevant will likely be of limited utility.

Fleeting moments of distraction (e.g., quickly checking a smartphone when the text message chime goes off) and epochs of sustained attention to something off-task (e.g., typing out a text message while driving) are fundamentally different. In the latter case, the current focus of attention is likely sustained by volitionally activated memory—it has simply become what the observer wants to attend to. Calling each an instance of distraction ignores this important distinction.

Involuntarily attending to something (or having attention “captured” by something; e.g., Luck et al., 2021) and being distracted by something are not the same thing. I would advise against calling any involuntary instance of attention allocation a case of distraction. Distraction implies a failure of attentional control that carries a host of unnecessary implications. We can distinguish between volitional and involuntary influences on attention and ask questions about what people direct their attention to when and why without ever really invoking the concept of distraction—so why then invoke it? Care must be taken when defining distraction in terms of whether attending to something results in a measurable decrement in performance (e.g., slowing of response time in target identification), which is common in the attentional control literature (see Liesefeld et al., 2023), since the observer might not be actively trying to optimize performance in this way; this is true both in the real world and in the laboratory.

I would argue that there is no “epidemic of distraction,” really. There is quite possibly an epidemic of people directing their attention to things other than their work, their studies, and perhaps their physical surroundings beyond electronic media. This is merely a shift in priority, and there are a variety of reasons for it that are not reducible to poor or otherwise deficient attentional control. We can debate how bad this shift in priority or focus is for individuals and society, but what we call it should not be predicated on the weight of a value judgment concerning how situationally beneficial the contents of attention appear to be. What we are witnessing in the so-called epidemic of distraction is likely just a reflection of how a functional attentional system operates in modern society and we need to decide how we want to handle that reality.