Probably less than you have heard, and the honest answer is that the evidence is contested. The famous finding that a phone sitting face down on your desk drains cognitive capacity comes from a 2017 study whose direct pre-registered replication failed, and the two meta-analyses that have since pooled the literature reached different conclusions. If you want to know whether having your phone nearby affects focus, the current state is a possible small effect on working memory and null results almost everywhere else, in a literature the meta-analysts themselves describe as underpowered.
This one is worth getting right because the original claim travelled unusually far. It is the study behind a good deal of workplace advice, a category of productivity product, and the confident instruction to put your phone in another room.
What the original study claimed
The 2017 paper, Brain Drain, published in the Journal of the Association for Consumer Research by Ward, Duke, Gneezy and Bos, tested what it called the brain drain hypothesis: that the mere presence of your own smartphone occupies limited-capacity cognitive resources, leaving fewer available for whatever you are actually doing.
Its reported result was striking. Across two experiments, even when people successfully maintained sustained attention and resisted checking their phones, the mere presence of the device reduced available cognitive capacity, with the costs highest among those highest in smartphone dependence.
That is a memorable claim, precisely stated, with an intuitive mechanism. It is easy to see why it spread.
What happened when someone tried to repeat it
In 2022, Ruiz Pardo and Minda published a pre-registered direct replication of the second Ward experiment in Acta Psychologica. Same tasks, same six conditions, crossing phone location (on the desk, in a pocket or bag, or outside the room) with phone power (on or off).
The brain drain effect did not replicate. There was no difference between phone location conditions on either the operation span task or the go/no-go task. The authors’ summary is that the mere presence of one’s smartphone may not be enough to affect cognitive performance.
A single failed replication does not overturn a finding. But a pre-registered direct replication is the strongest form of the test, and it came back null.
What the meta-analyses say, and where they disagree
Two meta-analyses have since tried to settle it, and they did not agree.
The larger one, published in Media Psychology in 2023 by Douglas Parry, ran six meta-analyses across 56 effects and 7,093 participants. Its result: across cognitive functions, only a single statistically significant pooled effect was found. Working memory capacity was negatively affected by the mere presence of a smartphone, and the remaining cognitive functions produced null summary effects. The paper describes what it found as a negative effect smaller than theorised, and adds that there is substantial methodological heterogeneity and generally poor statistical power in the field, such that the extent to which the literature can support or refute the hypothesis is limited.
The other, published in Behavioral Sciences in 2023 by Böttger, Poschik and Zierer, identified 22 studies with 43 relevant effects and reports a significant overall negative effect. Read the sentence carefully though: the effect it reports is of smartphone use and presence together. Pooling active phone use with mere presence answers a different and easier question than the one Ward posed, because nobody doubts that using your phone interrupts your work. The disagreement between the two reviews is partly a disagreement about what was being measured.
Our own reading, not a claim from either paper: when two meta-analyses of overlapping literature reach opposite headlines, and one of them says the field is underpowered and heterogeneous, the appropriate conclusion is that the effect is small and unstable rather than that one side has won.
What actually survives
Strip it back and three things are left standing.
Working memory is the one domain where a pooled negative effect appeared, in the larger meta-analysis, described as smaller than the theory predicted. That is a real if modest finding, and it is worth noting that working memory is exactly the resource you lean on when holding a system’s structure in your head.
Everything else came out null in that analysis. Sustained attention, response inhibition, fluid intelligence: no pooled effect.
And the strongest single test of the original result, a pre-registered direct replication, found nothing.
None of that means your phone is harmless. It means the specific claim, that proximity alone taxes you while you are ignoring it, is much weaker than its reputation. Whether picking the phone up costs you is a separate question with a separate and better-established literature behind it, and it is not the question this research asked.
What to do with a contested finding
There is a version of this article that ends by telling you to put your phone in a drawer anyway, because it cannot hurt. That is worse advice than it looks, because it teaches you to keep acting on claims after the evidence for them has thinned, and that habit generalises badly across everything else you read about attention and productivity.
A better response is proportionality. If moving your phone out of reach is free, do it, since the cost is nothing and there is a possible small working memory effect. But do not build a system around it, do not expect a noticeable difference, and do not conclude that your focus problems are solved because the desk is clear. On the evidence, that is not where the leverage is.
Common questions
Should I put my phone in another room while I work?
If it costs you nothing, it is a reasonable free bet, but do not expect much. The largest meta-analysis on this, covering 56 effects and 7,093 participants, found a single statistically significant pooled effect: a negative one on working memory capacity, described as smaller than theorised. Every other cognitive function it examined produced null summary effects. The same paper notes substantial methodological heterogeneity and generally poor statistical power in the literature, and says the extent to which the field can support or refute the hypothesis is limited. A pre-registered direct replication of the original 2017 study found no difference between having the phone on the desk, in a bag, or outside the room. So the honest expectation is a small effect on one specific capacity, or none at all.
Wasn’t there a famous study proving phones drain your brain?
Yes, and it is the reason the claim is everywhere. The 2017 Brain Drain paper reported that even when people resisted checking their phones, the mere presence of the device reduced available cognitive capacity, with the largest costs among the most phone-dependent. A pre-registered direct replication of its second experiment, published in 2022, found the effect did not replicate, with no difference between phone location conditions on either task used. Two later meta-analyses disagree with each other: one found only working memory affected and everything else null, the other reported a significant overall negative effect but pooled active smartphone use together with mere presence, which is a different question. The original result should be treated as contested, not established.
Does it matter if the phone is switched off or face down?
The evidence does not support a meaningful distinction, largely because it does not clearly support the underlying effect. The 2022 replication specifically crossed phone location with phone power, on or off, across six conditions, and found no difference between location conditions on either the operation span or go/no-go tasks. Since the location effect itself did not appear, there is nothing for power state to modify. Our own reading rather than a finding: any advice you have seen about the precise placement or orientation of the phone is more specific than the evidence underneath it.
If my phone is not the problem, what is?
This research only ever asked about a phone you are not touching. It says nothing about what happens when you pick it up, which is a separate question with its own literature and is not covered by any source here. The meta-analytic picture on mere presence is one small working memory effect and null results elsewhere, in a field the analysts describe as underpowered and heterogeneous. That is a reason to stop treating phone proximity as the explanation for a difficult afternoon, and to look at what is actually fragmenting the work instead, starting with what an interruption actually costs. If your concentration has changed in a way that worries you, or you are struggling to function rather than just to focus, that is worth raising with a doctor rather than solving with desk arrangement.
The more useful thing this episode teaches
The brain drain result is a good case study in how a finding becomes furniture. It was published in a real journal, it described a real mechanism, and it confirmed something people already half-believed. Within a few years it was advice. Within a few more, the replication came back null and almost nobody updated.
What is worth taking from it is not a conclusion about phones. It is a habit: when a claim about your attention is very memorable, very actionable and very widely repeated, those three properties are about how well it travelled, not about how well it held up.
Your phone may or may not be costing you a little working memory while it sits there. The literature genuinely does not know yet, and says so. That is a duller answer than the one you were given in 2017, and it is the one the evidence currently supports.
Leave a Reply