Soft Fascination: What Attention Restoration Theory Says About the Feed

The video is a minute and a half long and sounds like a lecture. "Directed attention fatigue is a phenomenon where the capacity for voluntary focused attention degrades after sustained use," it begins, "and unlike physical fatigue, it does not recover through passive rest." It recovers through "environments that engage involuntary attention": nature, "unhurried social contact," "unstructured time." This, the narrator says, is "the core finding of attention restoration theory, developed by Rachel and Stephen Kaplan at the University of Michigan." Scrolling during a break consumes attention; a walk without headphones, a meal without a screen, or sitting near a window "watching nothing in particular" restores it. Every traditional culture built such windows into the day "without knowing the mechanism," from the Mediterranean midday break to Japanese forest bathing to "the Islamic prescription of spaces between obligations." Modern productivity culture replaced them with "push through at any cost." The close: "Your attention is not weak, it is depleted."

I saved it because the theory it cites is about where attention goes, which is what my profession designs. By the time I went back to check it, the account could no longer be found: on 6 November the handle returned no profile, and I found no trace of it on the web. The claims are still everywhere, in this form, so I checked them anyway.

The video contradicts the theory it cites on its central point, adds two sources of restoration the founders never proposed, and presents a 1982 government coinage as ancient tradition. The experiments behind the real theory are smaller and less stable than its reputation. But reading the Kaplans in the original turned up something the video missed entirely: a sentence from 1995 that describes a social media feed more precisely than most of what has been written about feeds since. The theory's evidence is thin. Its vocabulary is excellent, and for a designer the vocabulary may be the more useful half.

1. What the video claims, and what holds

Two things hold. The definition of directed attention fatigue is a fair paraphrase of Stephen Kaplan's: directed attention takes effort, is under voluntary control, controls distraction through inhibition, and is susceptible to fatigue (Kaplan 199590001-2), p. 170). And the attribution is right: both Kaplans spent their careers at the University of Michigan.

Most of the rest changes the theory. "It does not recover through passive rest" contradicts the founders, as the next section shows. "Unhurried social contact" and "unstructured time" do not appear as restorative categories in the founding texts; the Kaplans' 1989 book uses a library reading room full of socializing people as an example of a setting that drains attention. The practical claims stretch thin evidence: no study I found tests walks with and without headphones, the research on screens at meals measures how much people eat, and the window studies concern views of trees, not "watching nothing in particular." The scrolling claim has partial support, which I come to in section 4. "The research says that pushing through is exactly what deepens the deficit" is a fair summary of the direction of work-recovery research and an overstatement of its certainty.

The cultural claims fare worst, and I come back to them in section 5. Shinrin-yoku is a modern coinage. The midday rest is old, but a minority of Spaniards nap regularly today. And no recognized Islamic teaching prescribes "spaces between obligations."

2. What the Kaplans actually wrote

The theory has a longer lineage than the video suggests. William James distinguished voluntary from involuntary attention in 1890, describing voluntary attention as a series of repeated efforts to bring a topic back to mind. James did not say it tires. Kaplan credits that idea to Frederick Law Olmsted, the landscape architect, writing about scenery in 1865 (Kaplan 1995, p. 170).

Rachel and Stephen Kaplan's book The Experience of Nature (1989) frames mental fatigue as an overworked capacity for directed attention. It names sleep as a way to rest that capacity, popular with tired people but limited (p. 182), and argues that waking rest needs settings that make few demands on directed attention. The ingredients were never exclusive to nature: the feeling of being in a whole other world, the book notes, can come from plays, shopping malls, zoo design, debugging a computer program or baking a showpiece cake (p. 184).

Stephen Kaplan's 1995 paper is the theory's full statement, and its wording matters, because the video reverses it. "Sleep provides one approach to recovery. While useful, it is insufficient. Certainly for serious cases of directed attention fatigue, insomnia is likely to set in long before full recovery has taken place." What is needed, he continues, "is an alternative mode of attending that would render the use of directed attention temporarily unnecessary" (p. 172). In 2010, Kaplan and Marc Berman described restoration as resting directed attention, listed sleep, meditation and gentle involuntary engagement as the three ways to do it, and compared attentional fatigue explicitly to the overuse of muscles (Kaplan and Berman 2010, pp. 47–48). The video's "unlike physical fatigue, it does not recover through passive rest" is the opposite of what the theory's authors wrote.

The alternative mode of attending is what Kaplan, borrowing from James, called fascination, and he split it in two. Hard fascination grips completely; his example is watching auto racing. Soft fascination holds attention gently and leaves room for reflection; his examples are clouds, sunsets, snow patterns, leaves moving in a breeze. Fascination is necessary for restoration and not sufficient. A restorative setting also needs three more things (p. 173). Being away is a shift of mind more than of place: "Clearly being away involves a conceptual rather than a physical transformation," and "a change in the direction of one's gaze, or even an old environment viewed in a new way can provide the necessary conceptual shift." Compatibility means the setting fits what you want to do. And extent, the component that matters most for this essay:

"The environment must have extent. It must, in other words, be rich enough and coherent enough so that it constitutes a whole other world. An endless stream of stimuli both fascinating and different from the usual would not qualify as a restorative environment for two reasons. First, lacking extent, it does not qualify as an environment, but merely an unrelated collection of impressions. And second, a restorative environment must be of sufficient scope to engage the mind. It must provide enough to see, experience, and think about so that it takes up a substantial portion of the available room in one's head" (p. 173).

He wrote that in 1995. I return to it in section 6.

3. What the experiments show

The theory's most famous study holds up as reported and has never had a large replication. Berman, Jonides and Kaplan (2008) had 38 Michigan students walk for about fifty minutes, once through the Ann Arbor Arboretum and once along a busy downtown street, a week apart, after a task designed to tire their attention (Berman et al. 2008). On backward digit span, the number of digits a person can repeat in reverse order, scores rose 1.5 digits after the nature walk and 0.5 after the downtown walk. Mood improved more after nature, but the mood change did not track the memory change. There was no condition in which people simply rested.

The only test I found that pits nature against passive rest on an attention-type task is older. In Hartig, Mang and Evans (1991), 102 people did fatiguing tasks and then spent forty minutes on a nature walk, an urban walk, or reading magazines with soft music (Hartig et al. 1991). On a proofreading test afterward, the nature group did best and the relaxation group did about as well as the urban group. No baseline was reported. A later study by Hartig and colleagues included a group that was not tired first, and nature helped that group as much as the tired one, which fits a general boost better than recovery from depletion (Hartig et al. 200300109-3)).

The picture-based studies did worse. In 2015, a Melbourne team reported that forty seconds of looking at a photo of a flowering green roof, against a bare concrete one, helped 141 students sustain attention (Lee et al. 2015). In 2022 the same lab ran the procedure twice more, with 68 and 186 people, and found no difference in performance, though participants rated the meadow and ocean images as more restorative (Johnson et al. 2022). An image-based study from 2005 failed its replication in 2021. A preregistered study of nature videos found no restoration of working memory (Hartanto et al. 2023), and in a 2026 study of 456 adults, a five-minute nature video did no better than video of geometric shapes on backward digit span, while both lifted mood more than urban scenes (Trammell et al. 2026).

Pooled, the effect is real and small. A 2016 systematic review of 31 studies found benefits on 3 of 13 measures, most reliably about half a digit on backward digit span (Ohly et al. 2016). A 2018 meta-analysis found low to moderate improvements in working memory, cognitive flexibility and, less reliably, attentional control (Stevenson et al. 2018). Critics add two problems. Soft fascination has never been given an operational definition, and most studies never show that attention was depleted before the nature exposure, so better scores afterward cannot be read as recovery (Joye and Dewitte 2018). And the closest cousin of the depletion idea, ego depletion, which Kaplan and Berman leaned on in 2010, failed two large multi-lab replications, with effects of d = 0.04 across 23 labs and d = 0.06 across 36 (Hagger et al. 2016; Vohs et al. 2021).

4. Feeling restored

Across the recent studies that found no effect on performance, one result keeps appearing: people say nature images and places feel more restorative, and report less fatigue afterward, even when their attention scores do not move. That gap between feeling restored and performing better is the most consistent finding in this literature, and it appears outside nature research too.

Short breaks are a good example. A 2022 meta-analysis of micro-breaks of ten minutes or less, across 2,335 people, found reliable gains in vigor and reductions in fatigue, and no significant overall effect on performance; after cognitively demanding tasks, the performance effect was absent (Albulescu et al. 2022). Breaks reliably make people feel better. Whether they make people work better depends on the work.

The phone evidence has the same shape. In the best-known experiment, 414 students solved anagrams with a break in the middle: a shopping task on a phone, on a computer, on paper, or no break at all (Kang and Kurtzberg 2019). The phone group solved fewer anagrams afterward than the other break groups, and more than the group with no break. The phone task was shopping, not scrolling. In a smaller study, using a laptop while resting in a green space cancelled the attention benefit of the green space (Jiang et al. 2019). And in a month-long trial, blocking mobile internet for two weeks improved sustained attention, partly because people spent more time socializing in person, exercising and outdoors (Castelo et al. 2025). I found no experiment that compares a social media scrolling break with a screen-free one on an attention measure. "Scrolling continues to consume directed attention" is plausible and untested.

5. The traditions, checked

The video's traditions are offered as proof that people knew the cure before science named it. Checked, they show something else: how easily a modern idea can be presented as an old one.

Shinrin-yoku, forest bathing, was coined in 1982 by Japan's Forestry Agency, credited in the agency's own magazine to its then Director-General, on the model of the Japanese words for sea bathing and sunbathing. The aim was to renew appreciation of forests and bring people into national forests. It rested from the start on a claimed mechanism, aromatic compounds released by trees, and funded research began in 2004. The well-known immune-cell studies had 12 or 13 participants each, from one lab. Systematic reviews rate the evidence weak: one found every included trial at high risk of bias, and another found that forest groups already had lower stress hormones before their walks (Oh et al. 2017; Antonelli et al. 2019). Japan's older nature customs, such as shrine groves and seasonal outings to view blossoms or autumn leaves, are the defensible examples of tradition.

The midday rest is genuinely old. "Siesta" comes from the Latin sexta hora, the sixth hour after dawn; the Rule of Benedict has monks rest silently on their beds after the midday meal in summer; and early Muslim reports describe a midday nap after Friday prayer. Today, in a 2024 Spanish poll of 2,000 adults, 24.7% napped daily or habitually. A nap also works through sleep pressure and the circadian dip after lunch, which is a different mechanism from the one the video claims for it.

The third example misdescribes a living religion. In Islamic sources the five daily prayers are the obligations ("prescribed for believers at fixed times," Quran 4:103), and the tradition frames prayer itself as rest; a well-known hadith has the Prophet ask for the call to prayer with the words "give us rest through it." Teachings on moderation are strong and specific: "your body has a right over you." None of this amounts to a prescription of "spaces between obligations." The likeliest source is modern productivity writing that presents the five prayers as daily resets, and the respectful thing is to describe the teachings as the tradition describes them.

Put together, the pattern is an appeal to ancient wisdom confirmed by modern science, which flatters both: the past looks wise and the science looks as if it explains culture. Here it fails on its own terms. One example is forty-four years old, one works through a different mechanism, and one does not exist in the form described.

6. An endless stream of stimuli

Read Kaplan's sentence about extent again with a phone in your hand. "An endless stream of stimuli both fascinating and different from the usual would not qualify as a restorative environment." Lacking extent, "it does not qualify as an environment, but merely an unrelated collection of impressions." Kaplan was arguing against a too-simple reading of his own theory, the reading in which anything that captures involuntary attention restores. He was not describing technology. But it is hard to find a more exact description of a feed: endless, fascinating, different from the usual by design, and made of impressions that have no relation to one another beyond the ranking that placed them next to each other.

The theory's other components describe it as well. A feed offers hard fascination, the kind that grips, more than soft fascination, the kind that leaves room to think. It offers a version of being away, a shift of gaze, while bringing the old thoughts along in the form of messages, work and news. Its compatibility with what the user came to do is often low, since many people open a feed to rest and leave it more tired. And it has no extent. It is never rich and coherent enough to constitute a whole other world, because the next item is chosen to interrupt the last.

In 2010, Kaplan and Berman put a version of this in plain terms: much of what captures involuntary attention in modern life, they wrote, has been taken over by advertisers and media, and often works against people's own goals (p. 48). This is a theoretical prediction, untested on feeds. But it explains a pattern the video gets half right. A walk restores and a feed does not, the theory would say, because the walk has extent and soft fascination and the feed has neither.

7. Designing for extent

I find the theory's vocabulary more useful than its evidence, and for a designer that is a legitimate place to land. The four components are design criteria. They describe what an environment does to attention, and an interface is an environment people spend hours inside.

Extent asks whether a product constitutes a coherent world or a stream of unrelated impressions. A well-made reading app, a long-form essay, a map, a game with a world, a reference library organized by idea: these have extent. An infinite feed sorted by predicted engagement does not, almost by definition. Soft fascination asks whether an interface holds attention gently and leaves room to think, or grips it. Being away asks whether a product offers a real shift of mind, or carries the user's obligations into every corner of it. Compatibility asks whether the product fits what the person came to do.

Designers have had a related vocabulary for thirty years. In a paper dated December 1995, the same year as Kaplan's, Mark Weiser and John Seely Brown at Xerox PARC proposed "calm technology": technology that moves easily between the center and the periphery of attention, informing without demanding (Weiser and Brown 1995). Calm technology and soft fascination describe the same quality from two sides: one from the device, one from the mind. Neither has been the default in consumer software.

My own best ideas come from walking; at my desk I mostly generate tasks. I do not think Kaplan's lab evidence explains that, and I would not claim the research proves it is good for me. What the vocabulary gives me is a way to describe what a walk has that my phone lacks: one continuous world, gentle things to look at, distance from the work, and nothing designed to interrupt the last thing I saw. I have written before about how my own attention works; this is the environmental half of the same question. And in Designing the Slope I argued that the default is still the product. The default of most attention-heavy software today is the endless stream.

8. Against myself

The strongest objection is that I am praising a vocabulary because it is vague. Joye and Dewitte's critique applies to me as much as to the theory: soft fascination has no operational definition, and an undefined term can be stretched to cover whatever a designer already wanted to say. They propose a simpler explanation for the nature effects: fascination as a form of interest that raises persistence on the next task, with no depletable resource involved. If they are right, "extent" may be a poetic name for "interesting enough to hold attention," and a feed is interesting enough by any measure.

The feed's defenders have some evidence too. People use games and entertainment media to recover from stress, and some experiments report recovery benefits from entertaining media. A feed can contain coherent worlds: a long thread, a single creator's archive, a community with its own culture. And the phone-break evidence is mixed: in the best study, a phone break still beat no break.

Finally, the gap between feeling restored and performing better cuts both ways. If nature mostly makes people feel better, the same may be true of a pleasant feed, and feeling better matters. A theory that predicts performance gains and mostly produces mood gains is weaker than its popularizers claim, and that weakness applies to my design reading of it as well.

9. At the desk, and how this could be wrong

For my own work, the conclusion is a short set of questions to ask of anything people will spend long stretches inside. Does it have extent, or is it a stream of unrelated impressions? Does it hold attention gently, or grip it? Does it offer a real shift of mind, or carry the user's obligations everywhere? Does it fit what people came to do? And when we test it, do we measure what people do afterward as well as how they feel, since the research shows those can diverge?

The argument could be wrong in three specific ways. First, a well-powered, preregistered experiment comparing a social media scrolling break with a screen-free break could find no difference on attention measures. That would remove the practical half of this essay. Second, a large multi-lab replication of a real nature walk could fail, which would leave the theory's evidence base close to empty and its vocabulary without support. Third, a study could show that feeds organized for coherence, a single thread or a single topic, restore attention as well as a walk does. That would mean extent can be designed into a feed, which would be the best outcome of all.

Until one of those happens, I read the video as a confident summary of a theory it had not read, and the theory as a modest finding with an unusually good sentence in it. Kaplan wrote that an endless stream of fascinating stimuli is "merely an unrelated collection of impressions." Thirty years later, that is still the most precise thing I know about the feed.

References

• Albulescu, P., Macsinga, I., Rusu, A., Sulea, C., Bodnaru, A., & Tulbure, B. T. (2022). "Give me a break!" A systematic review and meta-analysis on the efficacy of micro-breaks for increasing well-being and performance. PLOS ONE, 17(8), e0272460. https://doi.org/10.1371/journal.pone.0272460
• Antonelli, M., Barbieri, G., & Donelli, D. (2019). Effects of forest bathing (shinrin-yoku) on levels of cortisol as a stress biomarker: A systematic review and meta-analysis. International Journal of Biometeorology, 63(8), 1117–1134. https://doi.org/10.1007/s00484-019-01717-x
• Berman, M. G., Jonides, J., & Kaplan, S. (2008). The cognitive benefits of interacting with nature. Psychological Science, 19(12), 1207–1212. https://doi.org/10.1111/j.1467-9280.2008.02225.x
• Castelo, N., Kushlev, K., Ward, A. F., Esterman, M., & Reiner, P. B. (2025). Blocking mobile internet on smartphones improves sustained attention, mental health, and subjective well-being. PNAS Nexus, 4(2), pgaf017. https://doi.org/10.1093/pnasnexus/pgaf017
• Hagger, M. S., Chatzisarantis, N. L. D., et al. (2016). A multilab preregistered replication of the ego-depletion effect. Perspectives on Psychological Science, 11(4), 546–573. https://doi.org/10.1177/1745691616652873
• Hartanto, A., Teo, N. L. A., Lua, V. Y. Q., et al. (2023). Does watching videos with natural scenery restore attentional resources? Experimental Psychology, 70(2), 96–107. https://doi.org/10.1027/1618-3169/a000578
• Hartig, T., Evans, G. W., Jamner, L. D., Davis, D. S., & Gärling, T. (2003). Tracking restoration in natural and urban field settings. Journal of Environmental Psychology, 23(2), 109–123. https://doi.org/10.1016/S0272-4944(02)00109-3
• Hartig, T., Mang, M., & Evans, G. W. (1991). Restorative effects of natural environment experiences. Environment and Behavior, 23(1), 3–26. https://doi.org/10.1177/0013916591231001
• Jiang, B., Schmillen, R., & Sullivan, W. C. (2019). How to waste a break: Using portable electronic devices substantially counteracts attention enhancement effects of green spaces. Environment and Behavior, 51(9–10), 1133–1160. https://doi.org/10.1177/0013916518788603
• Johnson, K. A., Pontvianne, A., Ly, V., et al. (2022). Water and meadow views both afford perceived but not performance-based attention restoration. Frontiers in Psychology, 13, 809629. https://doi.org/10.3389/fpsyg.2022.809629
• Joye, Y., & Dewitte, S. (2018). Nature's broken path to restoration: A critical look at Attention Restoration Theory. Journal of Environmental Psychology, 59, 1–8. https://doi.org/10.1016/j.jenvp.2018.08.006
• Kang, S., & Kurtzberg, T. R. (2019). Reach for your cell phone at your own risk: The cognitive costs of media choice for breaks. Journal of Behavioral Addictions, 8(3), 395–403. https://doi.org/10.1556/2006.8.2019.21
• Kaplan, R., & Kaplan, S. (1989). The Experience of Nature: A Psychological Perspective. Cambridge University Press.
• Kaplan, S. (1995). The restorative benefits of nature: Toward an integrative framework. Journal of Environmental Psychology, 15(3), 169–182. https://doi.org/10.1016/0272-4944(95)90001-2
• Kaplan, S., & Berman, M. G. (2010). Directed attention as a common resource for executive functioning and self-regulation. Perspectives on Psychological Science, 5(1), 43–57. https://doi.org/10.1177/1745691609356784
• Lee, K. E., Williams, K. J. H., Sargent, L. D., Williams, N. S. G., & Johnson, K. A. (2015). 40-second green roof views sustain attention: The role of micro-breaks in attention restoration. Journal of Environmental Psychology, 42, 182–189. https://doi.org/10.1016/j.jenvp.2015.04.003
• Oh, B., Lee, K. J., Zaslawski, C., et al. (2017). Health and well-being benefits of spending time in forests: Systematic review. Environmental Health and Preventive Medicine, 22, 71. https://doi.org/10.1186/s12199-017-0677-9
• Ohly, H., White, M. P., Wheeler, B. W., et al. (2016). Attention Restoration Theory: A systematic review of the attention restoration potential of exposure to natural environments. Journal of Toxicology and Environmental Health, Part B, 19(7), 305–343. https://doi.org/10.1080/10937404.2016.1196155
• Stevenson, M. P., Schilhab, T., & Bentsen, P. (2018). Attention Restoration Theory II: A systematic review to clarify attention processes affected by exposure to natural environments. Journal of Toxicology and Environmental Health, Part B, 21(4), 227–268. https://doi.org/10.1080/10937404.2018.1505571
• Trammell, J., Kim, M., & Tian, H. (2026). Chronic nature exposure does not moderate affective and attentional effects of acute nature exposure. Frontiers in Psychology, 17, 1731261. https://doi.org/10.3389/fpsyg.2026.1731261
• Vohs, K. D., Schmeichel, B. J., et al. (2021). A multisite preregistered paradigmatic test of the ego-depletion effect. Psychological Science, 32(10), 1566–1581. https://doi.org/10.1177/0956797621989733
• Weiser, M., & Brown, J. S. (1995). Designing calm technology. Xerox PARC (published in PowerGrid Journal, 1996). https://calmtech.com/papers/designing-calm-technology

Source note: sources were checked on 6 and 7 November 2026. Kaplan 1995 is quoted from a scan of the journal pages; Kaplan and Kaplan 1989 and Kaplan and Berman 2010 are paraphrased with page numbers. The Berman, Hartig 2003, Lee, Johnson, Ohly, Joye and Dewitte, Kang and Kurtzberg and Albulescu papers were read in full; Stevenson 2018, Hartanto 2023, Trammell 2026, Jiang 2019 and Castelo 2025 at abstract level; Hartig 1991 through Kaplan 1995 and the Ohly review. The shinrin-yoku history rests on the Forestry Agency's magazine and the Nihon Kokugo Daijiten; the Spanish napping figure on the 40dB poll of January 2024; the Islamic texts on the standard hadith collections and the Quran. The video's account could not be found online after 6 November 2026; its claims are quoted from auto-generated captions.