How the attention economy is changing adults and children — and what future it prepares for our ability to think

In the previous article on the topic we followed the story of Tristan Harris, former Google employee and co-founder of the Center for Humane Technology, and his criticism of technologies designed to capture our attention. The starting point was simple: notifications don't just inform us, they often decide when to interrupt our time. The applications call us inside; once we enter, endless feeds, stories, reels, shorts and short videos try to hold us back.

This second article tries to broaden the scope. It's not just about notifications, nor about a single platform. It's about a broader transformation of our relationship with reading, waiting, concentration and complexity. Why do even people who grew up before the digital age, accustomed to books, newspapers and long television programmes, seem to struggle today when faced with extensive texts and prefer increasingly shorter, visual, rapid and immediately gratifying content?

I am not a psychologist, a neuroscientist, or a media sociologist. I deal with other fields and therefore do not pretend to propose specialist conclusions on such a complex subject. However, I decided to delve deeper into the topic because I consider it essential for the future of our societies. The ability to pay attention isn't just about academic performance or screen time. It's about how we understand reality, make decisions, build relationships, participate in democratic life and imagine the future.

To orient myself, I also used artificial intelligence as a tool for research and organization of information. The indications obtained were traced, as far as possible, to scientific studies, systematic reviews, meta-analyses and documents published by institutions such as the World Health Organization, the OECD, UNESCO and the US National Academies. This clarification is also important: artificial intelligence can help identify studies, compare results and build an initial synthesis, but it does not replace the verification of sources nor automatically transform a correlation into a cause.

First of all: what do we really know?

When it comes to social networks, short videos and attention spans, public debate tends to oscillate between two extremes. On the one hand it is argued that TikTok, Instagram and smartphones are destroying the attention, intelligence and mental health of the new generations. On the other hand, the response is that every new technology has produced similar fears and that there is not enough evidence to attribute such serious effects to social media.

The scientific literature suggests a more complex position. There is fairly solid convergence that the amount of information competing for our attention has increased enormously; that interruptions, multitasking and frequent task changes lead to cognitive costs; that problematic or compulsive use of digital media is more clearly associated with negative outcomes than the simple number of hours spent online; and that, for young children, the type of content, the presence of an adult, the time of day and the activities that the screen replaces matter a lot.

On other aspects, however, the evidence is significant but not definitive. The more intense use of short videos is associated with worse performance in attention and inhibitory control, but most of the available studies are observational. This means that we cannot always establish with certainty in which direction the relationship operates. Short videos could contribute to greater fragmentation of attention; at the same time, people who are already more impulsive, tired, anxious or easily distracted could be attracted to it to a greater extent. The relationship can be circular.

Similar caution applies to the mental health of adolescents. Social media can produce benefits and harm. They can offer belonging, creativity, information and support, but also expose you to social comparison, cyberbullying, harmful content, sleep loss and compulsive behaviors. The effects are not identical for everyone and depend on individual vulnerabilities, family context, type of content and characteristics of the platforms.

Finally, there are prospective considerations. No study can demonstrate today how we will think in twenty years. When, in the second part of the article, I try to outline some possible scenarios, I will therefore present them for what they are: inferences formulated starting from available trends, not scientific prophecies.

We have not become goldfish

One of the most repeated claims claims that human attention span has now dropped to eight seconds, becoming less than that of a goldfish. This comparison is not based on a universally recognized scientific measure. Attention does not have a fixed duration. It depends on motivation, interest, difficulty of the task, emotions, fatigue and context. The same person can abandon an article after a few lines and follow a television series, a match, a video game or a discussion that he considers important for hours.

The research also distinguishes between different functions: sustained attention, selective attention, inhibitory control, working memory, interference management and the ability to consciously switch from one task to another. The problem, therefore, cannot be summarized by saying that "the attention threshold has lowered". The most interesting transformation seems to be another: we are increasingly used to abandoning an activity before it has had time to become significant, especially when it requires an initial effort and does not offer an immediate reward.

The point is not that we have become biologically incapable of concentrating. The point is that we live in an environment designed to continually give us a reason not to.

Attention has become a market

Already in 1971 Herbert Simon had observed that a wealth of information inevitably produces a poverty of attention. The more what can be seen, read or heard grows, the poorer the ability necessary to select and understand it becomes (Simon, 1971). That intuition has today become an economic model. For much of the history of media, the consumer purchased the product directly: a book, a newspaper, a record or a movie ticket. On digital platforms, many contents are apparently free because what is monetized is above all the user's time, together with the data generated by their behavior.

Every second spent on a platform offers additional opportunities to display advertising, collect information, refine a profile and propose new content. Platforms don't just compete with each other; they compete with sleep, reading, work, conversation, play, contemplation and even the moments in which we do nothing.

A study published in Nature Communications analyzed the duration of collective attention in different cultural and informational fields. The authors observed that themes peak in popularity and are replaced more and more quickly. The phenomenon is explained through the increase in the quantity of contents that compete for a necessarily limited attention capacity (Lorenz-Spreen et al., 2019). The study concerns collective attention, it does not demonstrate that each individual's brain has become biologically less capable of concentrating. However, it documents an acceleration of the information environment in which that capacity must operate.

Short content is perfectly adapted to this environment: it takes little time, quickly generates data on user behavior and allows the platform to immediately propose an alternative.

Why short videos, stories and Reels work so well

A Reel, a Short or a TikTok video concentrates different psychological and technological mechanisms. The first is continuous novelty: each movement of the finger can open a completely different scene, person, emotion or topic. The second is unpredictability: we don't know if the next video will be useless, funny, surprising, outrageous or moving. Precisely the uncertainty about the quality of the reward can push us to continue.

This mechanism is often summarized, in a simplistic way, using the word "dopamine". Dopamine actually participates in the processes of motivation, learning and anticipation of reward, but it would not be correct to describe it as a substance that automatically makes anyone who watches a short video addicted. The third element is low decision cost. In the feed we don't have to carefully choose a book, a film or an article; just keep scrolling. The fourth is personalization: recommendation systems observe what we watch, interrupt, rewatch or share, record the time spent on content and progressively learn which combinations of themes, images, faces and rhythms are most likely to retain us.

Finally, a natural conclusion point is missing. A book ends, a film reaches the end credits, a newspaper has a last page. The infinite feed never communicates that the experience is complete. The next piece of content is already there before we consciously decide to look for it.

Stories also add a specific element: temporariness. They are presented as contents destined to disappear and this deadline produces a small urgency. We could lose them. They are placed in a visually privileged position, often associated with familiar faces, and require minimal effort. A long post formulates an implicit request: "devote me a few minutes, follow me to the end and keep what you read at the beginning". One story promises something very different: “just look at me for a moment.”

It's not necessarily true that people are no longer interested in ideas. Many times they don't get to the point where an idea might become interesting, because another stimulus has already claimed their attention.

Because it also concerns adults who grew up before digital

Being born before the Internet does not make you immune from these processes. The adult brain retains a certain plasticity and cognitive habits can change through repetition. After years of notifications, messages, open windows and frequent switching between applications, we learn that every wait can be filled and every slight feeling of boredom can be interrupted.

Checking your phone becomes an automatic response to an empty window: a queue, an elevator, a traffic light, the time before an appointment, or even a brief pause in a conversation. Often we do not take the device because we have formulated a specific purpose. We take it because for a few seconds nothing is happening.

Studies on smartphone presence, notifications and digital multitasking do not always produce identical results. The effects vary depending on the task, the person, the type of use and the research method. However, there is a substantial literature on the costs of continuous switching between activities and on the difficulty of filtering interference when attention is repeatedly solicited.

We can then advance a first interpretation, declaring it as such: perhaps what is weakening is not only our ability to concentrate, but also our willingness to tolerate the empty interval. Waiting, getting bored, letting your mind wander, continuing an activity before it becomes rewarding: all these experiences seem less natural in an environment in which any pause can be immediately filled.

What we know about short videos

The specific research on TikTok, Reel and Shorts is more recent than the general research on television, smartphones and social networks. In 2025 Psychological Bulletin, one of the leading international psychology journals, published a systematic review with meta-analysis dedicated to the use of short videos. The analysis collected 71 studies, for a total of 98,299 participants. Overall, greater use of short videos was associated with less favorable cognitive performance, especially in attention and inhibitory control, as well as some indicators of lower psychological well-being (Nguyen et al., 2025).

This is an important source, because it does not represent a single experiment chosen to confirm a thesis: it brings together a large number of studies. However, a fundamental limitation must be considered: a significant part of the research included is correlational. We can therefore state with reasonable caution that the intense or problematic use of short videos is associated with less favorable attentional outcomes. We cannot yet conclude that it necessarily causes a permanent and uniform reduction in attention.

The relationship could be two-way. People who are more impulsive, anxious, tired or easily distracted may resort to short videos more frequently; repeated and fragmented consumption could, in turn, make it more difficult to break the habit and sustain less stimulating activities. Some experimental and neuropsychological studies have observed relationships between problematic use of short videos, self-control, and executive control, but the field remains young and requires more longitudinal and experimental studies (Yan et al., 2024).

The most correct position is therefore not: "short videos are destroying human attention". It is rather: "the first scientific syntheses indicate a significant association between intense use of short videos and difficulties in some cognitive functions, but do not yet allow us to establish a general and permanent causality".

Brevity is not the problem itself

Short content can be clever, useful, and even profound. It can explain a concept, show an experiment, introduce a book, disseminate health information, arouse curiosity or lead a person to delve deeper into a topic they didn't know. Brevity does not necessarily coincide with superficiality, just as length does not guarantee quality.

The problem emerges when the short content no longer represents an entrance to something, but becomes the entire cognitive environment. When each theme must produce a reward in the first few seconds. When what does not immediately capture is abandoned. When the sequence is not consciously chosen and does not have a term.

Five minutes intentionally dedicated to a video is not the same as two hours of automatic scrolling. Curated content is not the same as an infinite feed built to prolong the stay. The decisive difference does not only concern duration. It's about control.

For children and adolescents the problem is different

For an adult it mainly involves the transformation of already formed habits. For a child it also concerns the very formation of cognitive and emotional skills. During childhood and adolescence, impulse control, working memory, self-regulation, the ability to wait, sustained attention and management of frustration develop. The platforms therefore intervene on skills that are still under construction.

Data collected by the World Health Organization in 44 countries and regions shows that the share of adolescents with signs of problematic social media use increased from 7% in 2018 to 11% in 2022 (WHO Regional Office for Europe, 2024). “Problematic use” doesn't simply mean spending a lot of time online; indicates a loss of control, difficulty quitting, the priority given to social media over other activities and negative consequences in daily life. The WHO associates this condition with lower well-being, sleep problems and social difficulties, recognizing however that the relationship can be bidirectional: psychological distress can encourage greater use of digital media and some methods of use can amplify the distress.

The National Academies of Sciences, Engineering, and Medicine reviewed the available literature and came to a cautious conclusion: there is insufficient evidence to support that social media causes a uniform worsening of adolescent health across the entire population. However, there are real possibilities of benefit and harm, distributed unequally (National Academies, 2023). For some kids, social media offers belonging, support, creativity and access to communities that are difficult to reach in everyday life. For others they can accentuate social comparison, anxiety, exclusion, sleep disorders, cyberbullying or exposure to harmful content.

The most scientifically useful question is therefore not: "are social media good or bad for young people?". It is: "to which children, in which conditions, through which contents and through which characteristics of the platforms do they produce benefits or risks?".

In young children, what the screen replaces matters

In the first years of life, the problem is not just what the child watches. It's also about what, while he's watching, he's not doing. Time spent in front of the screen can replace movement, manipulation of objects, symbolic play, dialogue, observation of expressions, exploration of the environment and sleep.

A meta-analysis published in 2024 on JAMA Pediatrics showed that, in children under six years of age, more program viewing and television playing in the background were associated with less favorable cognitive outcomes. Age-inappropriate content and adult device use during interactions were associated with worse psychosocial outcomes. Shared enjoyment with an adult instead showed positive associations with some cognitive outcomes (Mallawaarachchi et al., 2024).

A previous meta-analysis on language development reached similar conclusions: greater quantity and background television were associated with lower language skills, while educational programs and shared viewing were associated with better outcomes (Madigan et al., 2020). A large longitudinal study also found that increased screen time between 12 and 36 months was associated with fewer words heard, child vocalizations, and adult-child conversational exchanges (Brushe et al., 2024).

These findings don't mean that every minute of screen time causes harm. However, they show that speaking generically about “screen time” is insufficient. At least five elements count: quantity, quality of content, age of the child, presence of an adult, experiences that are replaced. A cartoon seen with a parent, commented on and connected to the real world, is not the same as an automatic stream watched alone for hours.

The “screen time” problem taken alone

A meta-analysis published in JAMA Psychiatry, based on 87 studies and over 159,000 children, found statistically significant associations between longer duration of exposure to screens and internal and external behavioral problems. However, correlations were small and varied based on participant characteristics and measurement methods (Eirich et al., 2022). This result is important for two reasons: it does not allow us to completely dismiss the problem, but it encourages us not to transform medium and modest associations into catastrophic statements about all children.

The OECD, in the report How's Life for Children in the Digital Age?, insists precisely on the need to overcome a reading based exclusively on the number of hours. Children's digital experiences include opportunities for learning, creativity and socialisation, but also risks related to content, contact, privacy, commercial exploitation and substitution of essential activities (OECD, 2025). Technology does not operate in a vacuum. It interacts with family, school, economic conditions, the quality of relationships and individual characteristics.

The deepest risk: never going through the difficulty again

At this point we can move from the research results to a more interpretative consideration. Long reading requires a particular form of trust. You have to accept that the meaning doesn't present itself immediately. You need to traverse a slow page, retain some information, build connections, and wait for the conversation to reveal its direction. The same happens in study, in scientific research, in music, in understanding a complex problem and in human relationships.

The short feed implicitly proposes a different rule: if something doesn't immediately generate interest, it can be replaced. This doesn't mean that every TikTok user loses the ability to read or think. It means that dominant daily training often proceeds in the opposite direction of deep attention.

The cultural risk does not only concern the decrease in novels read. It's about the possibility of becoming less open to anything that requires an investment before offering a reward: a mathematical proof, an initially difficult lesson, an articulated political reasoning, a conversation that cannot be reduced to a slogan, a human relationship that goes through a crisis, a project whose results will arrive after years.

This is not the direct conclusion of a meta-analysis. It is a prospective consideration. But it is perhaps the central question of the entire article: will a society that struggles to sustain duration still be able to sustain complexity?

What future does the attention economy prepare?

There is no shared scientific prediction about what human attention will be like in ten or twenty years. The scenarios that follow combine previously documented trends and forward-looking considerations. To avoid misunderstandings, each section distinguishes what is already observable from what we can only infer.

From selection to personalized content generation

The platforms already use recommendation systems capable of selecting content based on individual behavior. Viewing time, interruptions, shares and returns to the same video contribute to building increasingly personalized sequences. With generative artificial intelligence, the next step may not just consist of choosing, among content already produced, the one most suitable for the user. The platforms could generate or modify the duration, rhythm, voice, images and emotional intensity of the content in real time, adapting them to the individual person and the particular moment of the day.

We would thus move from the algorithm that identifies the video capable of holding us to the algorithm that builds it specifically for us. Personalization would offer great educational and informational opportunities. At the same time, it would make it more difficult to distinguish between a service designed to satisfy a need and a system designed to prolong it indefinitely.

Attention could become a new form of inequality

The OECD documents notable differences in children's digital experiences in relation to social conditions, family skills, availability of support and alternative opportunities in the physical world. Not all children have the same access to books, sports, safe spaces, cultural activities, present adults and schools capable of developing critical digital skills.

The new inequality may not just affect access to technology. It could be about the possibility of escaping its intrusiveness. Some families will be able to offer environments in which silence, reading, conversation, movement and device-free time exist. Other children may spend many hours in commercial digital environments designed primarily to maximize engagement.

The ability to focus, wait, and work long hours on a problem could become an unequally distributed form of cultural capital. The division would no longer be only between connected and non-connected, but between those who know how to govern the connection and those who are continually governed by it.

The school will have to choose whether to imitate the platforms or compensate them

UNESCO maintains that technology should only be used in education when it offers clear added value with respect to educational objectives. It also reports that excessive or inadequately guided use can produce distraction and does not automatically guarantee better learning outcomes (UNESCO, 2023). A growing number of school systems have introduced restrictions on the use of smartphones during lessons.

The school will be faced with two possible answers. The first consists in imitating the digital environment: increasingly shorter, spectacular, fast and fragmented lessons, in the hope of maintaining students' interest. The second consists in using digital as an input, while also preserving the capacity for prolonged attention.

A video can introduce a question. An animation can make a phenomenon visible. An artificial intelligence system can help personalize an exercise. Subsequently, however, explanation, writing, comparison, verification, practice and time are needed. If schools simply imitate TikTok, they risk confirming the idea that nothing deserves attention if it doesn't produce instant gratification. If it rejects all technology, it risks distancing itself from the world in which students live. Its task could become more ambitious: to use technology without necessarily adopting its attentional model.

Regulation will increasingly be about design

The National Academies, the WHO, the OECD and other institutions are progressively shifting attention from the sole responsibility of families to the characteristics of the platforms. Among the aspects discussed are recommendation systems, notifications, autoplay, infinite feed, data collection, age verification and protection from harmful content or contacts.

In the future, it may not be enough to recommend that parents monitor the number of hours spent online. It will be necessary to ask why a platform does not have a natural ending point; why notifications must continually interrupt other activities; what data is used to identify the minor's vulnerabilities; which emotions are favored by recommendation systems; how the economic interest of the platform is measured with respect to the well-being of the user.

The problem will no longer be just the content transmitted, but the behavior that the design makes habitual.

A culture of slowness could grow

Alongside rapid consumption, books, long podcasts, audiobooks, newsletters, in-depth courses and complex narrative series continue to exist. Furthermore, practices of voluntary disconnection, limitation of notifications, periods without smartphones and the search for unmediated experiences are spreading.

History shows that societies often react to excesses through compensatory movements. The spread of industrial food has been accompanied by organic and lo slow food. New forms of intentional slowness could respond to permanent connection. Reading, silence and uninterrupted conversation could gain value precisely because they become rarer.

Deep attention would not disappear, but it would cease to be a spontaneous condition. It would become a practice worth protecting.

Long thinking may not disappear, but become a minority

The success of long-form content shows that people have not lost the ability to follow extended experiences at all. When interest, meaning and emotional participation exist, millions of people listen to hour-long podcasts, watch complex series, take courses and read books.

The risk is not the complete disappearance of long thinking. It is its transformation into a minority practice, cultivated above all by those who possess the necessary tools, time and environment. A polarization could occur: on the one hand, increasingly rapid, automatic and personalized consumption; on the other, increasingly intentional and exclusive forms of in-depth analysis.

In such a society, attention would not be just a mental faculty. It would become a privilege.

The decisive question will be who controls attention

In the future it will not be enough to ask how much time we spend in front of a screen. We will have to ask ourselves who chooses what we see, based on which objectives, through which data and favoring which emotions. Psychology can study the effects on individuals. Sociology can analyze the power relations between platforms and users. Economics can explain who gets value from our time. The school can develop tools for understanding and self-control. Politics can establish what limits to place on systems capable of increasingly understanding the vulnerabilities of adults and children.

But each of us will also have to ask ourselves a personal question: when I watch, read or scroll, am I still choosing?

We are not necessarily becoming less intelligent

It would be too simple to conclude that TikTok, Instagram or smartphones are inevitably making us less intelligent. The most authoritative research invites us to avoid both catastrophism and naivety. Digital technologies can broaden access to knowledge, connect distant people, foster creativity, support isolated groups and offer extraordinary educational tools. However, the same environments can fragment attention, disturb sleep, intensify social comparison and exploit self-regulation difficulties.

A technology does not produce effects only through what it contains. It also produces effects through the behaviors it repeats, rewards and makes normal. Maybe we are not losing our intelligence. However, we may lose some of the conditions necessary to use it fully: time, continuity, silence, memory and the willingness to go through what we do not immediately understand.

Conclusion: stopping could become an act of freedom

We have gone from a world in which scarcity was about information to one in which scarcity was about attention. For centuries we have searched for more books, more images, more news and more knowledge. We now have an amount of content that no life could go through. When faced with excess, we don't necessarily choose what's most important. Often what best interrupts us prevails.

The short video did not create this transformation on its own. However, it represents the perfect form: rapid, emotional, personalized and inexhaustible. The challenge of the future will not be to produce content even faster. We already know that. The real challenge will be to preserve places, habits and institutions in which it is still possible to stop long enough for a thought to form.

Many things can be shown in fifteen seconds. Almost nothing truly complex can be understood in fifteen seconds. In a society built to continually hold us back, stopping could become not only a psychological necessity, but an act of freedom.

Image, Room in New York (Edward Hopper, 1932), Sheldon Museum of Art, Lincoln, oil on canvas

Essential sources

Brushe, M. E. et al. (2024), “Screen Time and Parent-Child Talk When Children Are Aged 12 to 36 Months”, JAMA Pediatrics, 178(4), pp. 369-375.

Eirich, R. et al. (2022), “Association of Screen Time With Internalizing and Externalizing Behavior Problems in Children 12 Years or Younger: A Systematic Review and Meta-analysis”, JAMA Psychiatry, 79(5), pp. 393-405.

Lorenz-Spreen, P., Mønsted, B. M., Hövel, P. and Lehmann, S. (2019), “Accelerating Dynamics of Collective Attention”, Nature Communications, 10, article 1759.

Madigan, S. et al. (2020), “Associations Between Screen Use and Child Language Skills: A Systematic Review and Meta-analysis”, JAMA Pediatrics, 174(7), pp. 665-675.

Mallawaarachchi, S. R. et al. (2024), “Early Childhood Screen Use Contexts and Cognitive and Psychosocial Outcomes: A Systematic Review and Meta-analysis”, JAMA Pediatrics, 178(10), pp. 1017-1026.

National Academies of Sciences, Engineering, and Medicine (2023), Social Media and Adolescent Health, Washington, DC, The National Academies Press.

Nguyen, L. et al. (2025), “Feeds, Feelings, and Focus: A Systematic Review and Meta-analysis Examining the Cognitive and Mental Health Correlates of Short-form Video Use”, Psychological Bulletin, 151(9), pp. 1125-1146.

OECD (2025), How's Life for Children in the Digital Age?, Paris, OECD Publishing.

Simon, H. A. (1971), “Designing Organizations for an Information-Rich World”, in M. Greenberger (ed.), Computers, Communications, and the Public Interest, Baltimore, Johns Hopkins Press, pp. 37-72.

UNESCO (2023), Global Education Monitoring Report 2023: Technology in Education — A Tool on Whose Terms?, Paris, UNESCO.

WHO Regional Office for Europe (2024), A Focus on Adolescent Social Media Use and Gaming in Europe, Central Asia and Canada, Health Behavior in School-aged Children International Report.

Yan, T. et al. (2024), “Mobile Phone Short Video Use Negatively Impacts Attention Functions: An EEG Study”, BMC Psychology / open-access literature on short-video addiction and executive control.

Note on the use of artificial intelligence

This article was developed with the support of artificial intelligence tools for research, analysis, organization of sources and writing. The selection of contents, the critical interpretation and the final responsibility of the text remain with the author.