Eating with a screen does not always increase that meal’s intake. Research points to later snacking, meal memory and important limits to the evidence.
At lunchtime, one hand holds your chopsticks and the other holds your phone. You reply to a message, check the weather and look down to find an empty lunch box. You know you ate. Yet the order in which you ate the food is less vivid than the message that arrived halfway through it. When you open a bag of snacks later, the way you ate lunch may be the last thing on your mind.
Experiments on distracted eating look beyond the plate in front of you. Distraction during a meal does not invariably increase intake at that meal, but it tends to increase intake at a later eating occasion. An updated systematic review and meta-analysis of experiments in adults draws attention to this difference in timing. The screen may be switched off before the outcome researchers are interested in is measured.[1]
Before turning that into “put down your phone and lose weight,” ask what was measured. Food intake, fullness, memory of a meal and long-term body weight are separate outcomes. The companion features on social eating and mindful eating address the influence of other diners and the broader question of weight-loss effects. Here, the focus is narrower: how attention during lunch may carry over into what happens when you next eat.
Separate eating more now from eating more later
The 2026 review selected adult experiments comparing eating with and without distraction, then statistically pooled their intake differences. It included 50 studies: 40 measured intake during the distracted eating occasion, and 10 measured intake later. Calling all of these findings “overeating” without specifying when obscures the distinction at the center of the research.[1]
For intake during the meal itself, the standardized mean difference was 0.123. A standardized mean difference puts results from studies using different measurements onto a common scale. It is not an increase in calories, grams or a percentage of the meal. The lower end of the 95% confidence interval was below −0.01 and the upper end was 0.25. Across the studies, the increase was not statistically significant.[1]
For concurrent intake in this adult analysis, the interval spans estimates on either side of no difference. That matters when deciding how strongly to describe the result. The estimate points toward greater intake, but the overall analysis does not support saying that distraction reliably increased the amount eaten during the meal. A direction in an estimate and a statistically significant result are different statements.[1]
Later intake gave a clearer result. The standardized mean difference was 0.419, with a 95% confidence interval of 0.195–0.642, a statistically significant increase. In everyday terms, the useful question is not only whether you had another helping while looking at your screen. It is also what happened at the afternoon snack or the next meal. An unchanged lunch does not necessarily mark the end of the possible effect.[1]
An earlier analysis reported larger differences. The 2013 meta-analysis included 24 studies and estimated an immediate-intake difference of 0.39, with a 95% confidence interval of 0.25–0.53. Its estimate for later intake was 0.76, with a 95% confidence interval of 0.45–1.07. Relying only on those older numbers would make an increase during the distracted meal look more certain than it does in the updated analysis.[1][2]
Both reviews point in the same direction for later eating. They differ in the strength of the finding for eating at the time of distraction. Additional studies and differences in how evidence is assembled matter when comparing the numbers. The current interpretation is that an effect on subsequent intake is supported, while an immediate increase is not uniform. The newer result refines the earlier picture rather than making every earlier experiment irrelevant.[1][2]
These studies principally measured food intake under experimental conditions. Their standardized differences describe differences in intake, not changes on a bathroom scale. Long-term weight change requires separate investigation. For a reader interested in losing weight, this is a useful boundary: the experiments identify a possible influence on eating, but they do not provide a promised weight-loss outcome from removing a screen at lunch.[1][2]
初日の汗は、予約した人だけが持ち帰れる。
Television and phones are different kinds of distraction
Distracted eating covers several activities: watching a video, reading text and operating a game, for example. The updated adult meta-analysis found that the type of task mattered. Passive distractions, such as watching television, increased intake during the meal, with a standardized mean difference of 0.272 and a 95% confidence interval of 0.128–0.417. Physically demanding and cognitively demanding tasks did not show statistically significant increases.[1]
Passive means mainly receiving something by watching or listening. Cognitive demand involves mental work such as remembering, making decisions or processing information. A phone is a device, not a description of which of these activities someone is doing. Watching a video and reading a work message can involve different demands. Since the research distinguishes tasks, grouping all phone use together loses information that helps explain the results.[1]
The practical implication is to describe the activity before drawing a conclusion about the device. If lunch disappeared while you followed a video, that is a different situation from trying to work through a difficult message between mouthfuls. Neither example establishes what happened to your intake. They illustrate why “there was a phone on the table” is an incomplete description of the condition being compared in a study.[1]
A 2025 meta-analysis specifically examining television gathered 23 studies covering children through adults. Its overall analysis combined 57 effect sizes and found no overall effect, with substantial variation between studies. When the analysis was restricted to television alone, the intake difference was 0.13, with a 95% confidence interval of 0.03–0.24: small, but statistically significant. The paper’s title announces an increase, while the results make that conclusion conditional.[3]
Timing mattered in that television review too. The immediate-intake difference was 0.10, with a 95% confidence interval of −0.01–0.21, and was not statistically significant. At the next meal, the difference was 0.30, with a 95% confidence interval of 0.03–0.57, indicating an increase. Advice to switch off the television therefore needs to consider the next eating occasion as well as the food eaten in front of it.[3]
Evidence specifically about smartphones is less extensive than evidence about television. A 2022 systematic review of screen use included 35 studies, but only 2 involved mobile phones. The authors said that more evidence was needed to determine the effects of using phones and tablets while eating. The television results should be read alongside the limited number of studies that directly examined mobile phones.[5]
That review included experimental research published from 2010–2021. Of the studies, 22 compared television content, 5 compared television with no screen, 4 compared television with games, and 2 examined simultaneous use of multiple screens. These categories show how much of the evidence concerned television and its content rather than the everyday variety of activities now performed on a phone.[5]
In the same review, 27 studies reported increased intake related to advertisements or food cues on screens, while 8 reported no significant difference. The authors discussed limitations in those latter studies involving design, sample size or control of variables. The review thus contained both increases and findings without a significant difference, with limitations affecting the latter comparisons.[5]
There is also a small direct comparison. In a 2024 repeated-measures study, 10 men and 13 women ate lunch under different conditions: no distraction, radio, television and smartphone use. Repeated measures means that the same people’s eating was assessed in multiple conditions. Participants ate more in the smartphone condition than in the other conditions. That is a direct observation about phone use, but it comes from a small study.[7]
Meal duration was 13.74 minutes with the smartphone, compared with 10.36 minutes without distraction. Fullness 30 minutes after eating was 84.23 mm without distraction and 74.70 mm with the smartphone. The millimeters refer to responses marked on a 100 mm line used to rate fullness. They are values on a subjective rating scale, not a physical measurement of food in the stomach.[7]
The longer meal and lower reported fullness occurred together in the smartphone condition. However, the conditions were presented in the order of no distraction, radio, television and smartphone. The fixed order and small number of participants limit interpretation. This study provides a clue worth examining, but its evidence should not be treated as having the same depth as the accumulated television literature.[5][7]
The same lunch, but a different afternoon snack
Meal memory offers a clue linking attention during eating with later intake. In a 2011 experiment, 44 participants ate a fixed lunch at a fixed rate. One group played a computerized card game while eating; the other ate without distracting stimuli. Holding lunch amount constant was important. It meant that a subsequent difference could not be explained simply by one group having eaten a larger lunch.[20]
In a taste test 30 minutes later, the game group ate an average of 52.1 g of biscuits, compared with 27.1 g in the group without distraction. The distracted participants also felt less full after lunch and remembered the order of the 9 lunch items less accurately. The lunch was the same, but reported fullness, memory and the amount eaten next differed between the groups.[20]
This design makes the delayed question unusually concrete. Researchers did not merely ask whether people who habitually use screens also habitually snack. They arranged a meal with and without a distraction, then measured what happened afterward. Nor did they need a larger lunch in the distracted group to produce the observed difference at the taste test. Attention during an otherwise fixed meal was part of the experimental comparison.[20]
The stomach received lunch, but memory did not seem to receive an equally clear account of it. That is a useful way to picture the mismatch. It is not proof that impaired memory was the only cause of greater snack intake. Weaker memory and increased snacking occurred together, but their co-occurrence alone does not establish a single causal pathway. The authors concluded that distraction at one meal can influence subsequent eating.[20]
The 2013 meta-analysis also examined experiments that strengthened memory of food already consumed. Those manipulations were associated with reduced later intake, with a difference of 0.40 and a 95% confidence interval of 0.12–0.68. The effect might vary with participants’ tendency toward disinhibited eating, meaning a tendency for restraint over eating to loosen. The result therefore does not imply an identical response in everyone.[2]
An everyday application is to avoid judging a screen-free lunch only by how you feel immediately afterward. Notice what you want to eat in the afternoon and what you actually eat next. Being able to recall the meal is another observation to consider alongside hunger. The point is to follow the sequence from lunch to the next eating occasion, because that is where these experiments found a difference.[2][20]
This also changes what counts as a useful question. “Did I finish the same lunch?” and “Did I eat the same amount later?” should remain separate. Because lunch was fixed, there was no room for a difference at the meal itself; the difference showed up later. For anyone trying to understand afternoon snacking, the earlier eating occasion is part of the context worth remembering, rather than a reason to assume a particular outcome.[20]
Numbers: distracted eating and later intake
50 studies: the updated adult meta-analysis included 40 measuring intake during the meal and 10 measuring later intake.[1]
0.123: the standardized mean difference in concurrent intake; the overall increase was not statistically significant.[1]
0.419: the standardized mean difference in later intake, with a 95% confidence interval of 0.195–0.642.[1]
52.1 g and 27.1 g: average biscuit intake 30 minutes after a fixed lunch, comparing the game group with the group without distraction.[20]
2 studies: the number involving mobile phones in a review of 35 screen-use studies, leaving limits to smartphone generalization.[5]
初日の汗は、予約した人だけが持ち帰れる。
1
2
