Case Study 1: The Brain on Smartphones — What the Ward et al. Study Really Found

The Study That Made Everyone Put Their Phone in Another Room

In 2017, a paper appeared in the Journal of the Association for Consumer Research under a title that was almost comically direct: "Brain Drain: The Mere Presence of One's Own Smartphone Reduces Available Cognitive Capacity." Written by Adrian F. Ward, Kristen Duke, Ayelet Gneezy, and Maarten W. Bos at the University of Texas at Austin's McCombs School of Business, the study was the kind of finding that seemed too clean to be real — and then turned out to be real enough to withstand several rounds of scrutiny.

The basic claim was this: having your smartphone nearby, even when it is silent, face-down, and not being used, reduces your available cognitive capacity. Not a little. Enough to measure in a controlled experiment. The brain drain occurs not because you are using the phone but because part of your cognitive machinery is occupied with the effort of not using it.

This case study examines what the Ward et al. study actually found, how it found it, what effect sizes look like, what limitations the study has, how it has fared in subsequent replication and extension work, and why its implications extend well beyond the academic setting.

What the Researchers Were Trying to Understand

The starting point for Ward and colleagues was a deceptively simple observation. People regularly report that their phones are distracting even when those phones are not actively in use. Workers describe difficulty concentrating when their phone is on their desk even if no notifications arrive. Students describe the pull of a silenced device as a constant background friction. Parents describe reaching for phones that aren't there.

The researchers hypothesized that this experience had a measurable cognitive basis. Their theoretical framework drew on two well-established findings in cognitive psychology:

First, working memory capacity is limited (following Miller, 1956; Cowan, 2001, and others) and is occupied by more than just the primary task. Active suppression of competing stimuli draws on the same executive resources as the task itself.

Second, smartphones have become extremely powerful conditioned cues for socially relevant information. For most users, the phone is the primary device through which communication, social monitoring, news, and entertainment arrive. It has acquired the status of a social signal source, and social signals are among the most powerful attentional draws the human brain recognizes.

The researchers' prediction was that the mere presence of the phone — without any use — would trigger a low-level monitoring orientation in the brain, occupying working memory resources through the suppression cost of actively resisting the impulse to check. And that this cost would be measurable on cognitive performance tasks.

Study 1: The Desk Condition

The first study recruited 520 undergraduate participants and randomly assigned them to one of three phone location conditions: phone placed face-down on the desk, phone placed in a bag or pocket, or phone placed in another room. In all conditions, participants were explicitly instructed to silence their phones and to turn them face-down if applicable.

All participants then completed two tasks measuring cognitive capacity. The first was a computer-based working memory task (the Operation Span task, a standard measure of working memory capacity) in which participants had to remember letters while solving math problems. The second was a fluid intelligence task derived from Raven's Progressive Matrices, a standard measure of reasoning capacity.

The results were clear and graduated in the direction the researchers predicted. Participants in the phone-on-desk condition performed significantly worse on both the working memory task and the fluid intelligence task compared to participants in the phone-in-another-room condition. Participants in the phone-in-bag/pocket condition performed in between the other two groups on both measures.

Importantly, the performance differences were found even when the researchers controlled for participants' reported smartphone use during the task (which was near zero across all conditions — participants were not using their phones). The effect was not produced by surreptitious phone use. It was produced by the phone's presence alone.

Study 2: The Smartphone Dependence Moderator

The second study specifically examined whether the size of the phone-presence effect depended on participants' level of smartphone dependence — the degree to which they relied on and were attached to their phones.

The researchers recruited 275 participants and again assigned them to phone-on-desk or phone-in-bag conditions, while additionally measuring smartphone dependence via a validated self-report scale. The cognitive performance measure used was the same Operation Span working memory task.

The finding here was particularly important: the phone-presence effect on working memory was significantly moderated by smartphone dependence. Participants who scored higher on smartphone dependence showed larger performance decrements in the phone-present condition compared to participants with lower dependence. This interaction was statistically significant and of meaningful magnitude.

The moderator finding is theoretically important because it supports the suppression mechanism. If the phone-presence effect were simply due to visual distraction or the general principle that any object on a desk competes for attention, there would be no reason to expect higher-dependence users to be more affected. The fact that higher-dependence users show larger effects supports the interpretation that what varies is the strength of the conditioned cue response — the degree to which the brain has learned to orient toward the phone as a source of important social information — and therefore the suppression cost of resisting it.

Effect Sizes: How Large Was the Brain Drain?

When findings are discussed in media coverage and popular accounts, effect sizes are almost always omitted. This is a serious oversight because effect size is crucial to understanding what a finding means practically.

In the Ward et al. studies, the effect sizes for the phone-presence manipulation were modest to medium by conventional standards. Cohen's d values in the neighborhood of 0.2 to 0.4 were reported across conditions and tasks, meaning that phone presence accounted for a meaningful but not overwhelming proportion of variance in cognitive performance.

To put this in practical terms: a Cohen's d of 0.3 corresponds to roughly a 12-percentage-point difference in performance between conditions at the median. This is not a trivial difference, but it is also not the difference between functional and impaired — it is the difference between performing well and performing somewhat less well.

Two things should be noted about this. First, the effect was produced by phone presence during a single experimental session lasting roughly thirty minutes. Cumulative effects from sustained phone-presence environments over weeks and months would, under reasonable assumptions, compound. Second, these effects were measured under conditions designed to minimize other stressors and distractions — a quiet lab, no other demands. In more typical noisy, demanding environments, the baseline is already higher for extraneous load, and the marginal phone-presence effect might interact with other cognitive demands in complex ways.

Limitations of the Study

The Ward et al. study was well-designed by the standards of experimental cognitive psychology, but like all research, it has important limitations.

The WEIRD population problem. All participants were undergraduate students at a major American university — a population that is young, educated, and tech-familiar. The extent to which the findings generalize to older adults, to populations with different smartphone use histories, or to non-Western contexts is unknown.

Ecological validity. The laboratory context is clean in ways that real-world study and work environments are not. Whether the effect observed in a quiet lab with minimal competing demands reflects what happens in a noisy dorm room or open-plan office is an open question.

Causality and selection. The study was experimental and did randomly assign phone location, which supports causal claims. However, the moderator analysis of smartphone dependence was observational — people high in dependence were not randomly assigned that status. It is possible that some third variable (impulsivity, baseline working memory capacity) both predicts smartphone dependence and predicts performance decrements, confounding the moderation analysis.

Demand characteristics. Participants knew they were in a study, and those in the phone-visible condition may have been more aware of the implicit experimental instructions to not use their phones. Whether this awareness itself contributed to the suppression cost is difficult to determine.

Measurement of the mechanism. The study demonstrated the effect of phone presence but did not directly measure the proposed mechanism — attentional suppression and its associated working memory cost. The mechanistic account is plausible and fits the data, but other mechanisms (general heightened awareness of experimental context, social anxiety about being judged for phone use) are not fully ruled out.

Replication and Extension

The Ward et al. study attracted considerable attention and generated a number of replication and extension attempts in subsequent years.

A 2019 replication by Trifan and colleagues using a similar design found comparable effects with similar effect sizes in a different student population, providing initial replication support. However, a larger preregistered replication attempt by Vanden Abeele and colleagues (2020) found smaller effects than the original, raising questions about whether the original result was somewhat inflated. The replication literature as of the mid-2020s suggests that the basic phenomenon is real — phone presence does have measurable cognitive costs — but that the original effect sizes may have been at the upper end of the realistic range.

Extension studies have examined the phenomenon in work contexts (finding similar but somewhat smaller effects in simulated workplace tasks), in children and adolescents (finding the relationship holds but with some developmental variation), and with different types of cognitive tasks (finding that tasks with high attentional demand show larger phone-presence effects than low-demand tasks, consistent with the working memory suppression account).

One important extension examined whether the effect could be eliminated by actively instructing participants to turn their phones off rather than silencing them. This manipulation did reduce the effect, consistent with the hypothesis that the phone's status as a potential information source is part of what drives the attentional suppression cost. A completely powered-off phone generated smaller performance decrements than a silenced but active phone.

What the Study Does and Does Not Show

It is worth being precise about what the Ward et al. finding implies and what it does not.

It does show that the spatial relationship between a person and their smartphone has measurable consequences for the cognitive resources available for other tasks. The brain does not treat the phone as neutral background furniture. It is processed as a social and informational resource whose potential availability requires ongoing management.

It does not show that smartphones are inherently harmful, that using a smartphone is cognitively damaging, or that the effects observed in a controlled experiment necessarily translate to dramatic real-world functional impairment for the average person. The effect is real and worth taking seriously, but it should be contextualized within what we know about the full spectrum of factors that affect cognitive performance.

It does not show that the effect is permanent or irreversible. The performance decrement observed when the phone is present does not reflect a lasting capacity change — it reflects the transient cost of managing the phone's presence. Removing the phone removes the cost.

Implications for Study Environments and Work Settings

The practical implications of the Ward et al. finding are straightforward but often resisted because they require inconvenient changes to established routines.

For students: the optimal study environment removes the phone from the room, not merely silences it or turns it face-down. The face-down condition was better than face-up, and the pocket/bag condition was better than the desk, but the best performance came from physical separation. This is a simple design choice with measurable expected benefits.

For knowledge workers: open desk environments where phones are routinely visible represent a continuous low-grade cognitive tax on everyone present. Organizations that have begun adopting phone-free meeting rooms and phone-away desk policies have practical neuroscientific justification for these choices.

For parents and educators: the instinct to require phones to be away during class or homework time is not merely disciplinary — it is cognitively well-grounded. A classroom policy that places phones in backpacks or outside the room, rather than merely face-down on desks, is meaningfully different in terms of the cognitive resources available for learning.

For individuals: the finding supports the development of environments deliberately designed around phone absence during cognitively demanding work. This is not about willpower but about environmental design — a form of what behavioral economists call choice architecture applied to attentional management.

The Broader Significance

The Ward et al. study is significant beyond its specific findings because it demonstrates, under controlled conditions, that the smartphone's effect on cognition is not limited to moments of active use. The phone affects the mind through its potential availability — through what it represents as a device and what that representation costs to manage.

This is not primarily a finding about addiction or compulsion. Most of the participants in the study were not pathological users. They were ordinary undergraduates with ordinary smartphone habits. The cognitive cost of phone presence was distributed across the normal range of users, not concentrated among the most problematic.

The implication is that the smartphone's cognitive footprint extends further into our lives than the time we spend actively looking at it. Every hour spent studying or working with a phone nearby — even silently, even face-down — is an hour in which some fraction of the cognitive resources that study and work require is deployed elsewhere. Understanding this is not a call for phone prohibition. It is a call for deliberate environmental design, informed by an accurate picture of what the brain is actually doing when the phone sits on the desk.

Key Takeaways from This Case Study

  • The Ward et al. (2017) study found that smartphone presence — even face-down, silent, and not in use — reduced working memory and fluid intelligence performance in a controlled experiment.

  • The performance decrement followed a gradient: phone in another room > phone in bag/pocket > phone on desk.

  • The effect was largest for participants with high smartphone dependence, supporting the proposed mechanism of attentional suppression (resisting the conditioned cue costs working memory resources).

  • Effect sizes were modest to medium (approximately Cohen's d = 0.2–0.4), real and meaningful but not catastrophic.

  • Replication studies confirm the basic phenomenon, though original effect sizes may represent the upper range.

  • The practical implication is clear: physical separation from phones during cognitively demanding work produces better performance outcomes than proximity, regardless of whether the phone is being actively used.