Posted on by rkoh029

Scroll. Like. Repeat.

Scroll. Like. Repeat.

The Hidden Cost of Social Media on Young Minds

Insights from an AI-based interviewing study of 583 youth and parents across Singapore and Australia

by Gemma Calvert, James Breeze and Ella Carnegie-Brown

Executive Summary

As policymakers worldwide debate the future of platforms like TikTok, this study delivers urgent, first-person insight into how social media is shaping the brains, behaviours and emotional lives of the next generation. In a joint project by Nanyang Technological University (NTU Singapore), Research Network, and ListenLabs.ai, 583 young people aged 13–25 and their parents from Singapore and Australia were interviewed using an innovative AI-based interviewing platform. Instead of surveys or Likert scales, participants spoke freely and candidly in voice-recorded interviews, revealing how social media affects their attention, emotions, mental health, relationships and sense of self.

The findings are clear and confronting:

    •  68% of youth report reduced attention spans, struggling to focus on content longer than a minute
    • 33% describe their inability to stop scrolling as “a kind of addiction” with many opening apps for “just a second” but losing hours
    • Teens liken their brains to being “monkey-minded” or “popcorn thinkers,” bouncing constantly from one thought to the next.
    • Emotional whiplash is common, with users reporting a rollercoaster of gratification, comparison, anxiety, and guilt, often within seconds
    • 65% of youth and 60% of parents worry about long-term impacts on school, work, and wellbeing

“It’s like my brain is wired to scroll now. I can’t stop.”

– Teen, Australia

These subjective experiences are not simply anecdotal. They align with recent neuroscientific evidence, including fMRI studies showing that social media “likes” activates the brain’s nucleus accumbens, the same dopamine-rich area involved in drug and gambling addiction (Sherman et al, 2016).

Yet amid the concern, some positive insights emerge. Many teens cite online communities as sources of emotional support, creativity and skill-building. Parents, too, are shifting from strict restriction to more collaborative approaches. But both generations agree: left unchecked, social media is shaping minds in ways that demand attention.

“We’re not just raising digital natives, we’re shaping their neural wiring in real time. The data is clear: social media is rewiring attention, emotion regulation, and selfworth. The question isn’t if this matters, it’s what we do about it.”

– Prof Gemma Calvert, Lead Investigator

“Social media isn’t going away. But if we want to protect the next generation’s mental and cognitive health, platforms must redesign for awareness, not addiction, and governments must act to make that the norm. It’s time for the industry to stop monetising distraction and start restoring attention.”

– James Breeze, CEO, Research Network

“I’m Gen Z. I grew up online. But even I can see something’s not right. This study gave us a voice and we’re saying we need help.”

– Ella Carnegie-Brown, 19, Research Network Intern

This white paper details the study’s findings and implications for education, policy, parenting and platform design. With attention now a precious commodity, the time for a collective, cross-sector response is now.

1. Introduction

Why this study, why now?

In the wake of rising global concern over the psychological effects of social media, particularly on young people, this study offers a timely, in-depth look into how digital habits are reshaping minds, moods and lives. While governments in the United States and Europe push forward with age-verification systems and mental health warning labels on platforms like TikTok and Instagram, there has been little direct consultation with those most affected: young people themselves. This research, focused on Australia and Singapore, fills that gap, bringing the voices of young users to the forefront and highlighting cultural differences in how these challenges are experienced and tackled.

This white paper presents findings from a groundbreaking, AI-based qualitative interview study conducted across Singapore and Australia. By capturing spontaneous, spoken reflections from 583 young people (aged 13–25) and their parents, the project reveals how attention spans, emotional stability, social behaviours and even identity development are being quietly but powerfully reshaped by algorithm-driven content consumption.

The study is timely. In July 2025, the European Commission launched age verification pilots across five countries to protect minors online (Reuters, 2025), while the U.S. state of Minnesota passed legislation mandating mental health warnings on social media platforms (MPR News, 2025). At the same time, China has proposed sweeping regulations to limit screen time to under 2 hours a day for users under 18 (CNN, 2023). Singapore’s Ministry of Education has already restricted mobile phone use during school hours (Straits Times, 2024), but elsewhere, such measures remain rare.

Amid this regulatory flux, this report takes a different approach. Rather than measuring app usage alone, we sought to understand what it feels like growing up immersed in social media and what the long-term implications might be for cognition, emotion, education, relationships and career readiness. By listening to their lived experiences, not just their screen time, we begin to see how digital environments are shaping not only behaviours but beliefs about self, learning, success, and connection. These insights matter not just to policymakers and parents, but to educators, mental health professionals, and the platforms themselves. As attention becomes a scarce resource and emotional regulation a growing challenge, understanding the subjective and neural impact of digital environments is no longer optional. It’s crucial. Media companies must now shift from monetising attention to restoring presence, not as a courtesy, but as a responsibility.

“Social media platforms have mastered how to capture attention. Now it’s time to take responsibility for the neurobiological and social consequences of their design.”

– Prof Gemma Calvert

2. Methodology

This study explored the views of 583 young people aged 13–25 and their parents across Singapore and Australia using a novel AI-powered interviewing platform, Listen Labs.

Participants were selected based on age, social media usage and location, and took part in voice-based, semi-structured interviews, where questions were presented on their mobile or desktop screen as text. The AI adapted in real-time to each respondent’s answers, enabling deep, personalised conversations about attention, wellbeing and digital life. This level of insight into beliefs, emotions, thoughts, and behaviours has never been possible at scale before. AI has transformed how we collect and analyse human experience.

The resulting transcripts were analysed using a combination of machine learning and thematic coding to identify patterns, themes, and demographic differences. Full methodological details are provided in Appendix A.

3. Findings

The results of our study paint a confronting picture of how social media is reshaping young minds. Instead of feeling connected and in control, many teens described being caught in a compulsive loop, scrolling not for joy but out of habit, often at the expense of focus, sleep and self-worth. Emotional spikes and crashes were common, with some feeling anxious within seconds of gratification. Attention spans are collapsing and even simple tasks now compete with the instant rewards of digital life. Beneath the surface, a deeper concern emerged: a growing fear that the ability to think clearly, feel steadily, or work meaningfully is slipping away. And while some teens are learning to carve out healthier paths through content creation, supportive communities, or boundaries, the dominant story is one of disruption, not balance.

3.1 Shrinking Attention Spans: Trained for Distraction

“TikTok has made my attention span so low that I can’t even watch a one-minute video.”

– Teen, 17, Australia

68% of our respondents reported trouble focusing, with 52% admitting they frequently get distracted during class by social media use. Many blamed the short-form nature of platforms like TikTok and Reels. The study also found that 15% habitually consume videos at 2x speed, training their brains to expect constant novelty and immediate payoff. One respondent called this state of mind “monkey-minded”, while another described themselves as a “popcorn thinker”, constantly jumping from thought to thought without sustained focus.

“Social media is conditioning the human brain to expect constant novelty and instant rewards… mirroring patterns seen in addiction. The consequences have far-reaching implications for every industry and educational sector”

– Prof Gemma Calvert

The cognitive impact is also structural. Research shows that high media multitasking is associated with reduced grey matter density in the anterior cingulate cortex, a region key for focus and decision-making (Loh & Kanai, 2014). In a digital environment optimised for distraction, adolescents are learning to skim, skip and switch and at the expense of deeper learning and comprehension.

3.2 Emotional Fallout: Guilt, Anxiety and the Rollercoaster Effect

“It’s like getting dopamine hits one second and shame the next.”

– Teen, 16, Singapore

Almost half (45%) of respondents described mixed or negative emotional reactions after using social media. Many spoke of guilt, emptiness, and anxiety, particularly after long scrolling sessions. Others highlighted “comparison anxiety,” triggered by curated images of beauty, success or happiness that made them feel inadequate.

“I start off laughing at memes and end up crying over someone’s perfect life.”

– Teen, 18, Australia

“When I scroll through social media and see others’ highlight reels… it can make me feel inadequate.”

– Teen, 17, Singapore

This emotional volatility reflects the constant highs and lows of algorithmically curated content and what one respondent called “emotional whiplash.” The brain is flooded with stimulation in rapid succession, often without time to process or reflect. Research backs this up: platforms amplify peer comparison and idealised self-presentation, which are especially potent during adolescence, a time of identity formation and heightened social sensitivity (Fardouly et al., 2015). While 10% of respondents did find community or comfort online, the dominant narrative was one of instability and exhaustion.

3.3 Academic Strain and Future Readiness

Many teens expressed worry that their current digital habits would undermine their ability to succeed in higher education or the workplace. 65% said they believed social media use was harming their learning, with many admitting to difficulties in finishing homework without checking their phones or losing track during lectures.

“During a lecture I was bored and used my phone and it made me lose track of some chapters the teacher was teaching.”

– Teen, 15, Singapore

“If I’m used to skimming info, I might struggle with detailed research or critical thinking.”

– Teen, 19, Australia

This aligns with recent cognitive research showing that accelerated content consumption (e.g., speed-watching) reduces processing depth, comprehension, and memory retention (Chiossi et al., 2023). Students are being conditioned to process information at surface level, rather than engaging deeply, a worrying trend for future job performance in knowledgebased economies. Some parents echoed this concern, describing their children as “mentally absent” during family time, and worrying that current school assessments don’t account for long-term attention erosion. Yet there was also a degree of resignation: “We cannot stop them from using social media… we just hope they learn to manage it,” said one parent in Singapore.

3.4 From Users to Creators: A Minority Harness the Upside

Although the majority of teens described negative or passive experiences, a small but vocal group (8–10%) reported benefits when they actively created content rather than simply consumed it. These young people described building confidence, learning new skills (e.g. editing, scripting, or gaming reviews), and finding supportive online communities.

“Posting videos helped me find my voice. It’s where I learned to speak up.”

– Teen, 17, Australia

“I follow creators who talk about climate and mental health. It feels real.”

– Teen, 16, Singapore

This group exhibited more self-awareness and purpose in their social media use. They also reported lower levels of comparison anxiety and likely because they were focused on creation and community, not simply consumption. However, these benefits were not universal. Many respondents avoided posting due to fear of judgment, pressure to appear perfect or privacy concerns. Others spoke of the exhausting perfectionism required to maintain a public image online. As one put it, “I take 50 photos just to find one that looks OK… then I edit for 10 minutes.”

4. Policy Implications: What Needs to Change, Now

Our findings expose a widening gap between the pace of digital innovation and society’s ability to mitigate its effects. Across two countries, we heard from youth that compulsive scrolling, emotional fragility, and attention collapse are now part of their daily online experience. If we continue to respond reactively, patching over symptoms rather than addressing the root causes, we risk entrenching these patterns. The time has come for proactive and systemic action that moves beyond cosmetic solutions.

4.1 For Policymakers: Stop Patching, Start Protecting

Too many so-called safeguards, like voluntary screen time alerts, are easily overridden and fail to account for how adolescent brains work. As behavioural researcher James Breeze notes, we must “redesign the digital environment, not just ask kids to self-regulate in a system engineered to hijack their focus”.

What’s needed are default-on protections for minors that shift the burden away from the user. Policy recommendations include:

    • Age-appropriate algorithm design, with reduced novelty cycling and intentional friction
    • Mandatory attention audits for high-usage platforms that are used by youth
    • Mental health labelling on social media interfaces, as pioneered in Minnesota
    • Public investment in longitudinal research to track neurocognitive and emotional outcomes

Singapore offers a compelling precedent for effective policy. The Ministry of Education has actively encouraged schools to limit phone access during school hours. Many schools now require students to hand in their phones at the start of the day, a policy that youth themselves report helps them focus. As one Singaporean youth stated, “I actually think rules help. If we didn’t have them, I’d be scrolling all day”. This demonstrates that structural changes, not just individual responsibility, are key to mitigating harm.

4.2 For Educators: Teach Attention Like a Muscle

If attention is the new literacy, then it deserves the same curriculum time as reading or numeracy. Teachers are already working miracles with overstimulated, device-saturated classrooms but they should not be expected to do it without tools. Simple interventions can help. Recommendations include:

      • Embedding metacognitive strategies into curricula (e.g., attention tracking, mindfulness breaks)
      • Chunking lessons into shorter, interactive segments to accommodate reduced stamina
      • Using digital tools with purpose, not as distractions, and being clear about when they genuinely support learning versus when they’re a convenient distraction

“Our brains aren’t designed for this kind of input. We need to be taught how to focus again.”

– 17-year-old, Singapore

4.3 For Youth: Reclaiming Agency and Focus

The insights from the youth participants themselves were a critical component of this study. They reveal a cohort that is acutely aware of the challenges social media presents to their concentration and well-being. As one participant articulated:

“Looking ahead, I think that social media habits could make it harder to stay focused on long-term tasks, like writing reports, studying for exams or working on big projects… it might be tough to engage with something complex or boring for a long time.”

To address these concerns, we propose the following evidence-based strategies, reinforced by the voices of the youth themselves:

    • Practice Mindful Content Consumption: Actively transition from passive scrolling to mindful engagement. A key reflective question should be, “Is this content contributing positively to my well-being and goals?”
    • Implement Structured Focus Techniques: To counteract fragmented attention, adopt structured methods like the Pomodoro technique (25 minutes of uninterrupted focus followed by a 5-minute break) (Biwer et al, 2023). This helps rebuild the capacity for deep concentration
    • Establish Clear Digital Boundaries: Designate specific times and physical spaces (e.g., study areas, the dinner table) as “tech-free zones”. This practice is shown to improve both focus during tasks and the quality of real-world social engagement
    • Curate a Supportive Digital Environment: Actively seek out and participate in online communities that align with personal interests, hobbies, or future goals
    • Transition from Consumption to Creation: Shift the balance of online activity from passive consumption to purposeful creation
    • Enhance Digital Literacy: Develop a critical understanding of platform mechanics, including algorithmic content delivery and data privacy
    • Conduct Regular Digital Well-being Audits: Periodically track and reflect on the correlation between social media use, mood, and productivity

4.4 For Tech Platforms: From Profit to Responsibility

The infrastructure to build healthier digital spaces already exists. What’s lacking is the will to deploy it. The challenge is not technical; it is a question of leadership and business model. The companies that built the attention economy now face a choice: continue to monetize distraction or take responsibility for redesigning their platforms with user wellbeing at the core.

Current efforts, such as optional screen-time alerts or voluntary app limits, are cosmetic and easily bypassed. This approach places the burden of self-regulation on users who are already struggling. As one 15-year-old from Singapore admitted, “I know I need to stop scrolling all night, but it’s hard”. A fundamental shift is required, moving from features users must opt into, to safeguards that are on by default.

Suggested industry actions:

    • Implement “Default-On” Ethical Design: Redesign youth-facing features by disabling auto-play, replacing infinite scroll with intentional friction, overhauling notification systems to eliminate non-essential interruptions by default, and reducing the prominence of ‘like’ counts
    • Proactive Collaboration with Experts: Move beyond token gestures and form deep, ongoing partnerships with educators and neuroscientists to allow independent research to inform platform mechanics and feature development
    • Radical Transparency and Accountability: Commit to transparent reporting on youth engagement data, including emotional impact metrics and the results of mandatory, third-party “attention audits”

4.5 For Parents & Families: Shift from Control to Connection

Rules alone are not enough. Our findings show that the most effective parental strategies were those built on trust, shared reflection and role modelling. Best practices include:

    • Starting early with explained boundaries, not imposed bans
    • Creating device-free spaces and shared screen rituals
    • Encouraging offline hobbies and guilt-free downtime

5. Conclusion

What’s at stake and what next?

This study makes one thing clear: we’re not just facing a tech problem, we’re facing a generational challenge. Social media is reshaping how young people think, feel and connect. And for many, it’s doing so faster than families, schools or societies can respond. Across Singapore and Australia, young people described a digital world that keeps them hooked but leaves them emotionally drained, cognitively scattered, and quietly worried about the future. Their parents shared those concerns, often with a sense of resignation, and many expressing concerns for themselves too!

“We can’t stop them using it. We just hope they learn to use it well.” 

– Parent respondent, Australia

But there is hope. This study showed not only what’s broken, but what can work. Teens who curated their feeds, parents who watched with their kids, schools that adapted their teaching all point toward a more conscious digital future.

“It’s not about banning everything. It’s about learning how to live with it and still be human.”

– Ella Carnegie-Brown, 19, Research Network

We call on governments, educators, industry leaders and parents to act together and now. Because attention, connection, and emotional resilience are not optional for the next generation. They’re foundational.

Appendix A: Methodology

Screening and Participant Selection

The study employed a comprehensive screening process to ensure a diverse and representative sample of participants. Respondents were selected based on specific demographic criteria, including age groups ranging from 13–25 years old and parents of children within this age range. The study focused on participants from two primary locations: Singapore and Australia. Participants were also screened based on their social media usage, with a preference for those who actively use popular platforms such as Instagram, YouTube, Facebook, TikTok, and LinkedIn. This careful selection process allowed researchers to gather insights from a wide spectrum of social media users, including both young adults navigating the digital landscape and parents observing its impact on their children.

Interview Method and Respondent Tasks

The study employed an innovative interview method utilizing the Listen Labs AI system. This advanced AI was programmed with a script supplied by the researchers, allowing it to generate text-based questions tailored to the study’s objectives. Participants engaged in a voice-based interaction, responding to the AI’s questions using speech alone. The AI demonstrated remarkable adaptability, generating follow-up questions based on each respondent’s unique voiced responses, ensuring a personalized and in-depth exploration of topics. This semi-structured approach combined predetermined themes with the flexibility to delve into emerging areas of interest. Participants were asked open-ended questions about their social media experiences, habits, and perceptions, covering topics such as the impact on focus and concentration, personal and professional development, and in-person social interactions. Respondents were encouraged to provide specific real-life examples, like describing how social media affects their ability to focus on tasks or changes in communication patterns with friends. Parents shared observations about their children’s social media use and its perceived effects on behaviour and wellbeing. The AI also prompted participants to consider future implications of their social media habits on academic performance and career prospects. This AI-driven interview method allowed for consistent data collection while accommodating the individual narratives of each participant, resulting in rich, qualitative data that offers deep insights into the multifaceted role of social media in the lives of young adults and families.

Data Analysis Process

The analysis of the interview data employed a rigorous, multi-stage approach to ensure comprehensive insights. Initially, all interview transcripts were processed through an advanced natural language processing (NLP) system to identify key themes and patterns. This system utilized a combination of machine learning algorithms and pre-defined coding schemes based on the research objectives. The analysis was structured around specific aspects, which corresponded to either questions asked during the interviews or properties of the responses. Within each aspect, a set of themes was developed to categorize and quantify the responses. For instance, under the aspect “Impact on Focus and Concentration,” themes ranged from “Severely Disrupted” to “No Impact.” Each response was then tagged with relevant themes, allowing for both qualitative interpretation and quantitative analysis. The system calculated the frequency of each theme, providing percentages of responses that matched specific themes within each aspect. This approach allowed for a nuanced understanding of prevalent trends while preserving the richness of individual responses. To ensure accuracy, a team of human researchers reviewed a subset of the AI-generated themes and tags, making adjustments where necessary. The analysis also incorporated demographic data, enabling comparisons between different segments such as age groups, nationalities, and parental status. This segmentation allowed for deeper insights into how social media impacts varied across different groups. Finally, the researchers conducted a cross-aspect analysis to identify correlations and patterns across different themes and segments, providing a holistic view of the social media landscape among young adults and parents in Singapore and Australia.

Analytic Foundations (Academic Addendum)

The analysis of the interview data followed a systematic, multi-stage approach grounded in established qualitative research methods. Transcripts from the AI-conducted interviews were processed using thematic analysis, a widely recognized method in qualitative research (Braun & Clarke, 2006). This approach involved identifying, analyzing, and reporting patterns within the data. The process began with familiarization with the data, followed by generating initial codes, searching for themes, reviewing themes, defining and naming themes, and producing the report. To enhance reliability and validity, the study employed a mixed-methods 12 approach, combining qualitative thematic analysis with quantitative content analysis (Krippendorff, 2018). This allowed for the quantification of qualitative data, providing frequency counts and percentages for each identified theme. The coding process was initially performed by an AI system trained on a subset of manually coded responses, with human researchers reviewing and validating the AI-generated codes to ensure accuracy and consistency (Saldaña, 2021). The analysis incorporated demographic segmentation, enabling comparisons between different groups such as age ranges, nationalities, and parental status. This approach aligns with best practices in cross-cultural research, allowing for the exploration of cultural and generational differences in social media use and perceptions (Okazaki & Mueller, 2007). Throughout the analysis process, measures were taken to ensure trustworthiness, including peer debriefing and member checking, as recommended by Lincoln and Guba (1985) for enhancing the credibility of qualitative research findings. The full coding approach was based on Braun & Clarke (2006); content analysis by Krippendorff (2018); validation reviewed using Saldaña (2021). Cross-cultural reliability supported by Okazaki & Mueller (2007); trustworthiness grounded in Lincoln & Guba (1985).

Appendix B: References

    • Biwer F, Wiradhany W, Oude Egbrink MGA, de Bruin ABH. (2023). Understanding effort regulation: Comparing ‘Pomodoro’ breaks and self-regulated breaks. Br J Educ Psychol. Aug;93 Suppl 2:353-367. doi: 10.1111/bjep.12593.
    • Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77–101. https://doi.org/10.1191/1478088706qp063oa
    • Carr, N. (2011). The Shallows: What the Internet Is Doing to Our Brains. W. W. Norton & Company.
    • Chiossi, F., Haliburton, L., Changkun, O., Butz, A & Schmidt, A. (2023). Short-Form Videos Degrade Our Capacity to Retain Intentions: Effect of Context Switching On Prospective Memory. DOI:10.1145/3544548.3580778
    • Fardouly, J., Diedrichs, P. C., Vartanian, L. R., & Halliwell, E. (2015). Social comparisons on social media: The impact of Facebook on young women’s body image concerns and mood. Body Image, 13, 38-45.
    • Krippendorff, K. (2018). Content Analysis: An Introduction to Its Methodology. Sage.
    • Lincoln, Y. S., & Guba, E. G. (1985). Naturalistic Inquiry. Beverly Hills, CA: Sage Publications, Inc. ● Loh, K. K., & Kanai, R. (2014). Higher media multi-tasking activity is associated with smaller graymatter density in the anterior cingulate cortex. PLoS One. 2014 Sep 24;9(9):e106698. doi: 10.1371/journal.pone.0106698
    • Okazaki, S., & Mueller, B. (2007). Cross-cultural advertising research: where we have been and where we need to go. International Marketing Review, Vol. 24 No. 5, pp. 499- 518. https://doi.org/10.1108/02651330710827960
    • Sherman LE, Payton AA, Hernandez LM, Greenfield PM, Dapretto M. The Power of the Like in Adolescence: Effects of Peer Influence on Neural and Behavioral Responses to Social Media. Psychol Sci. 2016 Jul;27(7):1027-35. doi: 10.1177/0956797616645673.
    • Saldaña, J. (2021). The Coding Manual for Qualitative Researchers. Sage.
    • Twenge, J. M., Joiner, T. E., Rogers, M. L., & Martin, G. N. (2017). Increases in Depressive Symptoms, Suicide-Related Outcomes, and Suicide Rates Among U.S. Adolescents After 2010 and 13 Links to Increased New Media Screen Time. Clinical Psychological Science, 6(1), 3- 17. https://doi.org/10.1177/2167702617723376

News & Policy Sources

 

Key Findings Table

Theme Finding
Attention & Focus

68% of youth say social media has hurt their ability to focus

52% report using phones during class; 15% watch videos at 2x speed

Common terms: “popcorn thinking,” “monkey-minded,” “dopamine hits”
Emotional Wellbeing

45% report feeling worse after use; 16% feel anxious from social comparison Teens describe “emotional rollercoaster” within seconds of scrolling

10% found comfort in niche support communities
Social Relationships

48% often reference online content in real-world conversations

27% use social media while socialising; depth of interactions declining

Parents report reduced family time and emotional presence
Learning & Work Readiness

Teens fear they’re “losing the ability to focus or think deeply”

Homework disrupted by notifications; multitasking perceived as “normal”

Parents worry about future workforce unpreparedness
Generational Divide

Parents worry about future workforce unpreparedness

Teens see benefits and harms; parents focus more on risks

Creators report more confidence than passive users

SG teens appreciate school phone limits; AU teens ask for more guidance
About the Authors

Gemma Calvert
Director of Industry Outreach and Thought Leadership Nanyang Centre for Marketing Technologies, Nanyang Business School, NTU, Singapore

James Breeze
CEO, Research Network

Ella Carnegie-Brown
Intern, Research Network

Leave a Reply

Your email address will not be published. Required fields are marked *