Original Research

Cyberpsychology in the digital age: Addressing social media’s mental well-being effects through digital interventions

Brian Maodza, Anjali S. Morar

Received: 24 Jan. 2026; Accepted: 04 June 2026; Published: 21 Aug. 2026

Copyright: © 2026. The Author(s). Licensee: AOSIS.
This work is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license (https://creativecommons.org/licenses/by/4.0/).

Abstract

Orientation: Rising social media use has intensified concerns about its psychological effects, especially among teenagers and young adults, with prior research linking engagement to anxiety, depression, and reduced emotional well-being.

Research purpose: This study examined how digital mental well-being interventions can mitigate the negative psychological effects of social media use.

Motivation for the study: While extensive literature documents the risks of social media, few studies synthesise solution-oriented digital interventions or examine their ethical implications. This study addressed that gap by integrating psychological theory with emerging digital practices.

Research approach/design and method: A qualitative systematic literature review of secondary data was adopted. Twenty-eight peer-reviewed sources were systematically selected and thematically analysed, guided by social comparison theory, cognitive behavioural therapy (CBT), and self-determination theory.

Main findings: Social media distress was driven by social comparison, fear of missing out, and cyberpsychological pressures. CBT-based applications, mindfulness tools, and online support platforms showed potential to reduce anxiety and depressive symptoms when appropriately designed.

Practical/managerial implications: Ethically designed interventions that prioritise user autonomy, data privacy, and inclusivity are essential. Practitioners, designers, and policymakers should adopt psychologically informed approaches to support sustainable well-being.

Contribution/value-add: This study contributed to cyberpsychology literature by synthesising evidence on the effectiveness and ethical challenges of digital mental well-being interventions, offering a theoretical foundation for future research and responsible innovation.

Keywords: cyberpsychology; social media; digital interventions; mental well-being; qualitative research; secondary data.

Introduction

In an increasingly hyperconnected world, social media platforms such as Instagram, TikTok and X have become deeply embedded in daily life, influencing how individuals communicate, develop personal identities and regulate emotions. While these platforms offer opportunities for self-expression and connectivity, growing evidence links their use to increased anxiety, depression, loneliness and diminished self-esteem, particularly among teenagers and young adults (Keles et al., 2019; Liu et al., 2022). These issues are a core concern within cyberpsychology, a field that examines how digital environments shape human cognition, emotion and behaviour. Although the digital ecosystem continues to evolve rapidly, traditional mental well-being support systems are struggling to keep pace with the psychological pressures created by persistent social media exposure. As a result, researchers have increasingly turned to digital interventions, such as mobile mental well-being apps, online cognitive behavioural therapy (CBT) programmes and artificial intelligence (AI)-supported therapy, as potential scalable solutions (Aboujaoude et al., 2020; Philippe et al., 2022). However, concerns remain about their long-term effectiveness, accessibility, and ethical implementation. This study therefore examines the mental well-being effects associated with social media use and explores how digital interventions can mitigate these effects. Using a qualitative, secondary data approach and thematic analysis of 28 peer-reviewed sources, the study aims to provide conceptual clarity on how ethical, evidence-based digital interventions can support mental well-being among young people.

Research questions

The study was guided by the following research questions:

• What are the primary psychological effects of social media on users’ mental well-being?
• What type of digital mental well-being interventions have been developed or proposed to address social media-induced psychological distress?
• What ethical considerations must be considered when implementing digital interventions for mental well-being?

Research objectives

To address these research questions, the study had the following objectives:

• To explore the primary psychological impact of social media, particularly in relation to self-esteem, anxiety and depression.
• To identify and analyse digital mental well-being interventions that can be implemented to address social media-induced distress.
• To evaluate the ethical considerations and challenges in implementing digital interventions for mental well-being.

Literature review

Literature in cyberpsychology highlights two central issues: the psychological risks of social media use and the potential of digital interventions to mitigate them. Studies consistently link social media engagement to social comparison, fear of missing out (FOMO), compulsive use and reduced emotional stability (Weinstein, 2023). These mechanisms are well explained by social comparison theory (Festinger, 1954) and self-determination theory (Ryan et al., 2021), which remain central to understanding digital behaviour and emotional responses online. While most existing research focuses on documenting risks, fewer studies offer technology-driven solutions. Emerging evidence shows that digital mental well-being tools, such as CBT-based mobile apps, AI chatbots and online peer-support platforms, can offer scalable support (Firth et al., 2017; Naslund et al., 2020). However, challenges remain regarding data privacy, informed consent, and algorithmic transparency (Burr et al., 2020). This study positions itself within this evolving field by examining the remedial potential of digital interventions while addressing their ethical complexities.

Theoretical foundation

Social comparison theory

Social comparison theory (Festinger, 1954) explains how individuals evaluate themselves by comparing their lives with those of others. On image-centric platforms such as Instagram and TikTok, these comparisons are often upward and idealised, intensifying negative self-judgement and emotional vulnerability (Keles et al., 2019; Powdthavee, 2023). This theory clarifies why heavy social media use is linked to lower self-esteem and well-being (Boursier et al., 2020). It suggests the value of digital interventions that reduce comparison pressures and promote healthier online cognition.

Cognitive behavioural therapy

Cognitive behavioural therapy highlights how distorted thought patterns contribute to emotional distress (Stallard, 2021). On social media, users often interpret likes, comments, and follower counts as indicators of personal worth. Cognitive behavioural therapy provides the foundation for many digital interventions, including mobile CBT apps and online restructuring exercises, which help users challenge negative thoughts triggered by online interactions (Vidal et al., 2020). Its widespread application makes it central to evaluating the effectiveness of digital interventions.

Self-determination theory

Self-determination theory (Ryan et al., 2021) proposes that autonomy, competence, and relatedness are core psychological needs. Social media can both support and undermine these needs through addictive design features, superficial interactions and variable reward systems (Shannon et al., 2022). For digital interventions to be ethically and psychologically effective, they must support autonomy and long-term intrinsic motivation, not create dependency. Together, these three frameworks guided the thematic analysis and the interpretation of findings in this study

Details of the problem being investigated

Psychological impact of social media

A substantial body of evidence shows that social media use is associated with anxiety, depression, reduced self-esteem, and emotional exhaustion, especially among adolescents (Keles et al., 2019). Upward comparison and idealised online portrayals intensify vulnerability (Beyens et al., 2020), while constant connectivity pressures increase stress, sleep disruption and burnout (George & Odgers, 2015). Fear of missing out further drives compulsive checking, reinforcing anxiety and low mood (Berryman et al., 2018). These findings emphasise the need for interventions that target the cognitive and behavioural mechanisms that sustain digital distress.

Online interventions for social media-related distress

Digital interventions, including mobile mental health apps, CBT-based chatbots and online therapy platforms, offer accessible solutions to social media-induced distress. Research demonstrates improvements in anxiety and depression when users engage in CBT-based digital tools (Firth et al., 2017; Linardon et al., 2019). Apps such as Woebot and Sanvello provide cognitive restructuring, behavioural activation and mood tracking, aligning well with youth preferences for immediate, mobile-first support (Pavlopoulos et al., 2024). Gamified elements such as badges and streaks can enhance engagement, though they must be designed carefully to avoid replicating addictive platform mechanics (Nishi et al., 2024). Emerging evidence on digital detox approaches also suggests short-term improvements in mood and mindfulness (Liu et al., 2025). Overall, digital tools show promise when grounded in psychological theory and designed around user needs.

Ethical challenges in digital mental well-being interventions

As digital interventions expand, ethical concerns become more pronounced. These include data privacy, user autonomy, algorithmic transparency and the risk of persuasive design (Montag et al., 2019; Palmer & Burrows, 2020). Apps often collect sensitive emotional data, heightening privacy risks. Self-determination theory highlights that interventions must empower, not manipulate, users (Ryan et al., 2021). Accessibility also remains a critical issue, as many interventions are designed for Western, English-speaking and digitally literate populations (Naslund et al., 2020). Ensuring cultural sensitivity, low-cost access and ethical design principles is essential for long-term intervention legitimacy.

Links between the problem and current literature

Although research has identified clear psychological risks and promising digital interventions, major gaps persist. There remains limited longitudinal evidence on the sustainability of digital mental health tools, particularly among young people. Many studies focus broadly on anxiety and depression, overlooking specific challenges such as body-image concerns or social identity issues online. Gamification remains under-regulated, and more work is needed to design motivational features ethically (Nishi et al., 2024). Furthermore, many tools lack user-centred design and fail to incorporate co-design with young people, reducing relevance and engagement. Overall, the literature shows that while digital interventions can help counteract social media’s harmful effects, their effectiveness depends on ethical engagement, inclusivity and long-term usability. This study seeks to address these gaps by examining how ethically designed, theoretically informed interventions can promote mental resilience, user agency and responsible digital innovation.

Research design

Research approach

This study adopted a qualitative, exploratory research design based on an inductive reasoning approach. The research aimed to investigate the psychological effects of social media and explore how digital interventions could be applied to mitigate these negative outcomes, especially anxiety and depression in young adults and teenagers. A qualitative design was chosen to capture subjective experiences and interpret meaning within existing research (Ahmad & Wilkins, 2024). Thematic analysis served as the core analytic technique in this study. This method enabled the systematic identification, organisation and interpretation of themes across qualitative data sources, including peer-reviewed journal articles and theoretical papers. The analysis followed an inductive approach, allowing themes to emerge from the data without imposing a predefined framework. This approach was crucial for capturing how the digital interventions were conceptualised, implemented and perceived in the context of social media-induced distress (Braun & Clarke, 2006). Overall, the approach of this study was fully aligned with the research problem, which sought to understand and respond to the complex interplay between social media and mental well-being. This allowed for a deep, ethically informed analysis of digital interventions and supported the development of a grounded theoretical framework to guide future innovation in mental well-being in the digital space.

Research strategy

The research follows a secondary data analysis strategy, drawing exclusively on peer-reviewed academic literature. This strategy is well-suited to projects involving established bodies of knowledge in which empirical primary data are not essential to answering the research question. By examining existing studies, theories, and evidence, the researcher can derive a comprehensive understanding of the psychological effects of social media and evaluate the effectiveness, design principles and ethical considerations of digital mental health interventions. The strategy allows for methodological efficiency while maintaining scholarly rigour, as it leverages high-quality academic findings already available.

Research method

Research setting

Since this study is based on existing literature, the research setting is non-physical and document-based. The analysis was conducted using journal databases, digital libraries, and peer-reviewed publications within the domains of cyberpsychology, mental well-being, social media research and digital intervention design. The ‘setting’ thus refers to the academic environment in which knowledge is produced and disseminated, rather than an interactional field. This setting ensures access to high-quality, credible scholarship and minimises bias associated with researcher-participant dynamics.

Entrée and establish researcher’s role

In qualitative secondary research, the researcher’s role involves critical interpretation, synthesis and evaluation rather than direct engagement with human participants. Entrée into the field was established through access to academic databases. The researcher assumed the role of an analytical observer, responsible for ensuring the integrity of the data selection process, maintaining objectivity and evaluating methodological and theoretical robustness within the reviewed studies. Ethical conduct was upheld by relying solely on publicly accessible, peer-reviewed academic work and avoiding manipulation or misrepresentation of original findings.

Sampling strategy for literature section

This study is a systematic literature review with thematic synthesis; no human participants were involved. The target population for this research consisted of teenagers and young adults who are active users of social media platforms such as Instagram, X (formerly Twitter) and TikTok. This population was selected as research indicates that younger demographics are the most vulnerable to the negative psychological effects of social media, including increased levels of anxiety, depression, and low self-esteem (Keles et al., 2019). The unit of analysis was therefore the social media users within the published studies themselves, as each article contributed insights into the lived experiences, interventions and measured impacts within diverse social media user groups. In terms of parameters, since the research adopted a secondary data approach, the accessible population was restricted to peer-reviewed academic articles and other secondary data sources that examined young adults and teenagers’ mental well-being in the context of social media usage. These characteristics were chosen as they aligned with the central research problem of understanding how digital interventions could mitigate the psychological risks linked to social media use among vulnerable groups. While no human participants were involved, the ‘participants’ within this study refer to the selected body of literature. A purposive sampling method was used to ensure that only studies directly relevant to the psychological effects of social media and digital mental health interventions were included. This sampling approach ensured that the dataset reflected depth, credibility, and thematic richness.

Data collection methods

The data collection method for this research was designed in direct alignment with the chosen qualitative research design and the research objective of exploring how digital interventions can mitigate the negative psychological effects of social media on mental well-being. Since this study relied on secondary data, no primary data were generated through interviews or surveys. Instead, the process involved a systematic and structured review of peer-reviewed academic literature relevant to the research questions and objectives (Nyimbili & Nyimbili, 2024). To identify relevant sources, a structured search strategy was implemented using databases such as Google Scholar, ScienceDirect, SpringerLink, and PubMed. The following keywords and Boolean combinations were used:

• ‘social media’ AND ‘mental well-being’
• ‘digital interventions’ OR ‘mental well-being apps’
• ‘social media anxiety’ OR ‘depression among young adults’
• ‘psychological effects’ AND ‘social networking sites’
• ‘digital therapy’ OR ‘online counselling’
• ‘young adults’ OR ‘teenagers’ AND ‘coping strategies’
• ‘social comparison’ AND ‘cognitive behavioural therapy (CBT)’.

These search terms were selected based on Braun & Clarke (2006) thematic framework, which supports systematic keyword mapping and clustering around central concepts of mental well-being and digital intervention. To ensure methodological rigour, inclusion criteria were applied to restrict the dataset to sources that were:

• Peer-reviewed and published in English.
• Focused on social media and mental well-being.
• Addressed digital interventions, coping mechanisms, or ethical considerations related to technology-assisted mental health care.
• Published between 2019 and 2025.

Exclusion criteria filtered out studies that were:

• Non-scholarly or opinion-based (e.g., blogs, grey literature).
• Older than 6 years (except for foundational theoretical sources).
• Focused on unrelated topics such as physical health or non-digital interventions.
• Methodologically weak or lacking transparency in data analysis.

This purposive, criterion-based sampling approach ensured that only the most relevant, credible and recent sources were included (Ahmad & Wilkins, 2024). It is widely used in literature-based research where theoretical insights form the basis of analysis rather than empirical data from participants.

Definition and Justification for using Preferred Reporting Items for Systematic Reviews and Meta-Analyses

The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) framework is a standardised method used to enhance the transparency and reproducibility of literature selection and review processes (Page et al., 2021). Preferred Reporting Items for Systematic Reviews and Meta-Analyses provided a structured, step-by-step reporting system that visually demonstrated how studies were identified, screened, and included or excluded. It is valuable in qualitative and mixed-methods systematic reviews, as it ensures a rigorous and traceable pathway from the initial database search to the final data inclusion (Liberati et al., 2009; Moher et al., 2015). In this study, PRISMA was adopted as it aligned with the (Braun and & Clarke, 2006) framework, which emphasised systematic and thematic categorisation of qualitative data. Using PRISMA allowed for a transparent, replicable selection process that minimised researcher bias and improved the credibility of secondary data inclusion. As confirmed by Page et al. (2021), PRISMA enhances both the validity and methodological accountability of literature-based research, making it a suitable instrument for secondary data collection within this qualitative study.

Implementation of Preferred Reporting Items for Systematic Reviews and Meta-Analyses

The primary instrument used for structuring the data collection process was the PRISMA flow diagram (Figure 1), which guided the transparent documentation of article identification, screening, eligibility, and inclusion. The PRISMA instrument enabled a systematic narrowing of articles, beginning with a wide search and ending with a carefully curated final set of studies that were most relevant to the research objectives.

FIGURE 1: Preferred Reporting Items for Systematic Reviews and Meta-Analyses – Flow diagram of data collection process.

The data collection process was implemented in several stages, as shown in Figure 1:

• Identification: Database searches were conducted across Google Scholar, ScienceDirect, SpringerLink and PubMed using keywords and Boolean operators. This initial search yielded over 250 results.
• Screening: Inclusion criteria such as publication year (2019–2025), peer-reviewed status and relevance to the psychological impacts of social media or digital interventions were applied. Exclusion criteria included non-peer-reviewed reports, duplicate studies, articles not published in English, and those outside the research scope. After screening, 38 articles were retained.
• Eligibility: Full-text eligibility checks were then conducted to ensure that each article aligned with at least one of the research questions or objectives.
• Inclusion: After final refinement, 28 articles were included in the dataset for analysis.

Justification of the data collection method

Secondary data collection was selected due to the extensive availability of empirical research in this field, making primary data collection unnecessary and beyond the scope of the study (Ahmad & Wilkins, 2024). This approach enabled the synthesis of findings across multiple studies, strengthening the validity of conclusions through triangulation and aligning with the study’s aim of developing an integrated understanding of digital interventions addressing social media-related distress (Bowen, 2009). The use of PRISMA guidelines and thematic coding ensured a systematic and transparent selection process, with studies purposefully aligned to the three identified themes (Page et al., 2021). This approach supported the qualitative research design by incorporating diverse perspectives while remaining consistent with the research objectives, clearly documenting inclusion and exclusion decisions to minimise bias (Moher et al., 2015).

Data recording

Data were recorded through detailed note-taking, summarisation and categorisation of key ideas extracted from each academic source. Notes were stored digitally in organised folders, with each source documented according to themes, concepts and theoretical contributions. No audio or video recordings were required due to the study’s non-empirical nature. Recording focused on capturing conceptual insights, recurring patterns and contrasting viewpoints across the literature.

Strategies employed to ensure data quality and integrity

To ensure the data quality and integrity of this qualitative study, efforts were made across the four major criteria identified by Korstjens and Moser (2017): credibility, transferability, dependability and confirmability:

• Credibility was enhanced through the systematic application of document analysis and thematic coding, ensuring that all included sources were reputable, peer-reviewed and aligned with the research objectives. The structured data extraction matrix and coding process provided consistency and reduced interpretative bias during analysis.
• Transferability was supported by presenting rich, detailed descriptions of the emergent themes within the findings chapter. This allowed readers to assess the applicability of the findings in different contexts (e.g., age groups, digital platforms, intervention types).
• Dependability was strengthened by maintaining a clearly defined and replicable research process. All stages of data collection and analysis were transparently documented, from database search parameters to coding decisions, thereby creating an audit trail that demonstrated methodological consistency throughout the study.
• Confirmability was achieved by maintaining researcher neutrality through manual thematic analysis. Interpretations were grounded in patterns and evidence directly derived from the reviewed literature rather than personal bias or assumptions. Where relevant, findings were triangulated across multiple sources to enhance objectivity.

In addition to these measures, the trustworthiness of the analysis was further reinforced through strategies aimed at reducing interpretive bias and enhancing transparency. The coding process followed a cyclical and reflective approach, in which early interpretations were continuously compared with source data to ensure alignment with original meanings. Reflexivity played an important role in maintaining analytical integrity. Reflective notes were maintained throughout the coding and synthesis stages to identify any assumptions or predispositions that could influence interpretation, ensuring that findings emerged from the data rather than from preconceived expectations. Triangulation was additionally embedded in the secondary data process by drawing insights from multiple types of peer-reviewed studies, including systematic reviews and case studies. This diversity of evidence strengthened the dependability of conclusions, as patterns identified across differing methodologies increased confidence in the thematic findings. By integrating these methodological safeguards, the study not only meets the qualitative criteria of credibility, dependability, transferability and confirmability (Korstjens and & Moser, 2017), but also demonstrates a commitment to ethical, rigorous and transparent research practice in the interpretation of secondary qualitative data.

Collectively, these measures ensured that the study maintained high qualitative quality and trustworthiness, despite relying exclusively on secondary data. The transparency of the PRISMA-based selection process and the systematic coding framework further reinforced the dependability and confirmability of the findings.

Data analysis

In alignment with the qualitative, secondary data approach adopted for this study, thematic analysis was selected as the primary method for data analysis. This choice was guided by the research design, which sought to explore and synthesise insights from existing peer-reviewed literature on the psychological effects of social media and the role of digital interventions in mitigating these effects. Thematic analysis allows for the identification, coding and categorisation of patterns across diverse sources, making it particularly suitable for examining secondary data derived from multiple studies with varying designs, populations and intervention types (Braun & Clarke, 2006).

Implementation of the analysis proceeded as follows:

• Familiarisation with the data: All 28 articles were carefully reviewed to ensure deep engagement with their findings and discussions.
• Generating initial codes: Using Braun & Clarke’s (2006) framework, relevant segments of text were coded based on recurring ideas related to psychological effects, digital interventions and ethical issues.
• Searching for themes: Codes were grouped into candidate themes aligned with the literature review themes: (1) Psychological impacts (social comparison, FOMO, anxiety, depression, addiction), (2) Digital interventions (CBT apps, chatbots, digital detox, hybrid models), (3) Ethics and design (privacy, algorithmic transparency, accessibility). Themes were refined iteratively to ensure coherence and distinction.
• Reviewing and defining themes: The preliminary themes were refined to ensure they accurately captured the breadth of the coded data while remaining aligned with the research objectives. Each theme was clearly defined and linked to the research questions, ensuring consistency across the analysis.
• Interpretation and synthesis: I linked each theme back to the theoretical lenses (social comparison theory; CBT; self-determination theory) to interpret mechanisms by which social media harms operated and how interventions worked or failed. The final themes were then mapped against the research objectives (see Findings).

Thematic analysis was implemented manually due to the manageable size of the dataset, allowing for refined interpretation and cross-referencing of findings between studies. The coding process was iterative. Initial codes were reviewed and refined to ensure that themes accurately reflected the evidence across multiple sources. This approach ensured the analysis remained grounded in the data while also allowing for the identification of recurring patterns and contrasts, which informed the synthesis of findings.

The use of thematic analysis was directly relevant to the research problem and objectives, as it enabled a structured comparison of how different digital interventions addressed psychological distress linked to social media use. It also facilitated the identification of gaps in the literature, such as underexplored intervention types or populations, which contribute to the study’s anticipated knowledge contribution. By linking data collection and analysis through a systematic coding and theme development process, the study ensured that findings were reliable, valid and closely aligned with the research questions.

Reporting style

The results of the study are presented using a narrative, thematic reporting style, where each major theme is discussed in relation to the research objectives. The reporting emphasises interpretation, synthesis and meaning making rather than numerical representation. Academic language, integrated evidence, and critical discussion guide the presentation of the findings. Figures and tables are included where necessary to enhance clarity and visualise conceptual relationships.

Ethical considerations

Ethical clearance to conduct this study was obtained from the Independent Institute of Education Varsity College School of Computer Science Research Ethics Committee (Ref. No. CSREC/65/2025).

Results

Presentation of findings

Thematic analysis of the selected 28 peer-reviewed articles revealed three main themes that aligned with the research questions and objectives of this study:

• Psychological impacts of social media
• Digital interventions for mental well-being
• Ethical and design considerations.

These themes were consistently identified across the dataset and directly align with the study’s research question and objectives. The organisation of the findings is structured as follows:

• A numerical overview of how the articles were distributed across themes (Figure 2).
• The presentation of Table 1, which summarises the reviewed studies and their corresponding themes.
• A thematic discussion that integrates illustrative evidence from the analysed literature.

FIGURE 2: Counts by theme (N = 28).

Counts by theme (N = 28) – refer to Figure 2:

• Psychological impacts: ~12 articles (systematic reviews, meta-analyses and empirical studies, e.g. Keles et al., 2019; Liu et al., 2022; Orben et al., 2022).
• Digital interventions: ~10 articles (meta-analyses, randomised controlled trials (RCTs), app studies, e.g. Ferrari et al., 2022; Lee et al., 2023).
• Ethics and design considerations: ~6 articles (reviews and thematic articles, e.g., Burr et al., 2020).

The proportions highlighted that while psychological impacts remain a dominant focus in recent research, the exploration of digital interventions and ethical frameworks has gained notable momentum since 2019. This suggested a growing academic shift from describing social media’s harms towards actively developing, evaluating and governing digital tools that mitigate those harms.

Table 1 provides an overview of the 28 articles included in the analysis. Each article is categorised according to its primary focus and the assigned theme. This table provides a visual representation of the dataset and supports the transparency and traceability of the analysis process.

Studies by Aboujaoude et al. (2020) and Philippe et al. (2022) provided integrative perspectives, bridging technological design with psychological health outcomes. The combination of diverse study types enhanced triangulation, allowing a more balanced synthesis of empirical findings and conceptual arguments. This integration was particularly valuable given that mental well-being outcomes in digital contexts are multidimensional, involving behavioural, emotional, cognitive and ethical dimensions.

The three core themes emerging from this synthesis collectively illustrated how social media functions as both a risk factor and a platform for mental well-being interventions as well as how ethical and design frameworks are essential to balancing these dual outcomes. The following sections present the findings under each main theme, followed by interpretation and linkage to the relevant research questions, objectives and theoretical frameworks underpinning this study, which are social comparison theory, cognitive behavioural therapy (CBT) and self-determination theory.

Discussion

Theme 1: The psychological impact of social media

The sub-themes are as follows:

• Social comparison and self-esteem issues linked to excessive use
• Fear of missing out and anxiety
• Cyberbullying and mental distress

Findings

The psychological impact theme was the most consistently evidenced theme across the dataset, with approximately 12 studies converging on two core mechanisms: social comparison and FOMO. The dominant mechanism identified is upward social comparison: repeated exposure to idealised images and comparison-driven environments (e.g. Instagram and TikTok) was identified as a significant trigger for negative affect and self-perception distortions. Orben et al. (2022) found that adolescents’ mental health fluctuates with screen time, highlighting sensitivity to social validation cycles. Similarly, Keles et al. (2019) noted that social media intensifies social comparison (Festinger, 1954), leading to diminished self-worth and emotional fatigue. Cyberbullying and FOMO (Shannon et al., 2022) were recurring mechanisms contributing to distress, particularly among young female users and high-frequency platform users (Liu et al., 2022). Shannon et al. (2022) reported that FOMO-related anxiety correlated with compulsive checking behaviours and emotional exhaustion, while Liu et al. (2022) demonstrated a dose–response relationship between screen exposure and depressive symptoms. Weinstein (2023) and Woodward et al. (2025) further highlighted that cross-platform differences, such as the image-centric nature of Instagram versus the discussion-driven format of Reddit, shape the intensity and type of psychological impact experienced. Cyberbullying, in particular, illustrates how self-determination theory’s (Ryan et al., 2021) need for relatedness can be weaponised: platforms that enable public shaming, pile-on commenting and anonymous harassment undermine both relatedness and autonomy, intensifying exclusion as well as helplessness. This distinction between authentic relatedness-building and manipulative social pressure is critical for intervention design, as tools that nurture genuine peer support differ fundamentally from those that exploit social belonging for engagement metrics.

Interpretation of findings in context

These findings directly addressed Research Question 1 and Objective 1, which aimed to explore the primary psychological impact of social media on mental well-being. The results aligned with social comparison theory (Festinger, 1954), which suggests that individuals evaluate themselves relative to others, leading to emotional consequences when exposed to idealised standards. The consistency across studies reinforced the view that social media’s structure promotes comparative and validation-seeking behaviours that heighten anxiety and depressive tendencies. An additional layer of interpretation emerged when considering developmental sensitivity (Orben et al., 2022), which suggested that digital exposure interacts with age-specific vulnerabilities, such as identity formation and peer validation. This means that not all users are equally affected; the psychological consequences depend on both usage patterns and life stage. In context, these findings confirmed that young adults and teenagers are particularly vulnerable to digital environments that prioritise social feedback, thereby necessitating supportive interventions to promote digital resilience and mental well-being.

Theme 2: The role of digital interventions in mental well-being

The sub-themes are as follows:

• Types of digital interventions (CBT apps, online support communities)
• Effectiveness and limitations of digital interventions

Findings

Ten studies examined digital interventions such as CBT-based applications, mindfulness apps, and online support communities aimed at improving mental health outcomes among social media users. Ferrari et al. (2022) demonstrated that smartphone-based CBT significantly reduced symptoms of anxiety and depression. Similarly, Lee et al. (2023) found that structured digital programmes using gamification and AI-assisted feedback improved emotional regulation and self-awareness. Artificial intelligence-based chatbots such as Woebot (Karkosz et al., 2024) provided immediate, stigma-free mental health support, demonstrating measurable improvements in anxiety and mood regulation. However, several studies (Linardon et al., 2019; Philippe et al., 2022) noted that effectiveness is contingent on sustained engagement, and dropout rates remain a major limitation. Studies such as Linardon et al. (2019) cautioned that while digital interventions hold strong potential, their effectiveness relies on sustained participation and ethical implementation. Boucher and Raiker (2024) identified that personalisation, feedback loops and perceived relevance enhance retention, implying that user experience design is as critical as therapeutic content. A comparative analysis of intervention types reveals meaningful differences in effectiveness profiles. Cognitive behavioural therapy-based applications (e.g., Woebot, Sanvello) demonstrated the strongest and most consistent evidence for reducing anxiety and depressive symptoms, particularly in short- to medium-term trials (Ferrari et al., 2022; Karkosz et al., 2024). Mindfulness-based digital tools showed effectiveness specifically for loneliness and social anxiety (Sun, 2023), but with less impact on depression severity compared to CBT approaches. Artificial intelligence-driven chatbots offered advantages in immediacy and stigma reduction, though their effectiveness was highly dependent on sustained engagement and personalisation features (Boucher & Raiker, 2024; Philippe et al., 2022). Online peer-support communities demonstrated benefits for perceived social support and belonging but posed ethical risks related to unmoderated content and peer influence (Naslund et al., 2020). Structured digital programmes combining CBT with gamified elements showed improvements in emotional regulation (Lee et al., 2023), though gamification features such as streaks and badges risk replicating the variable-ratio reward schedules found in addictive social media platforms (Nishi et al., 2024), and must be designed to support intrinsic motivation rather than compulsive engagement. In summary, no single modality is universally superior; rather, effectiveness is contingent on matching intervention type to user needs, ensuring sustained engagement, and maintaining ethical design principles.

Interpretation of findings in context

The findings correspond to Research Question 2 and Objective 2, which focus on identifying and analysing digital interventions that mitigate social media-induced distress. They align with cognitive behavioural therapy (CBT) (Stallard, 2021) and self-determination theory (Ryan et al., 2021), which suggest that individuals’ autonomy, competence and relatedness drive engagement and behaviour change. In this context, the evidence indicates that well-designed digital interventions can effectively counteract social media’s negative psychological effects by fostering self-regulation, mindfulness and emotional literacy. The integration of therapeutic frameworks into app-based solutions strengthens their potential as scalable, accessible tools for promoting mental well-being in young populations. The synthesis reveals an emerging paradigm: digital interventions are most effective when they integrate behavioural science, data ethics and human-centred design. In other words, technology alone is not therapeutic; it becomes therapeutic when designed to nurture psychological autonomy and user trust.

Theme 3: Bridging cyberpsychology and digital mental well-being solutions

The sub-themes are as follows:

• Theoretical models of behaviour change in digital interventions
• Ethical concerns and digital well-being strategies

Findings

Six studies emphasised the intersection of cyberpsychology and digital ethics. Buda et al. (2022) proposed fairness frameworks to address algorithmic bias and enhance transparency in health-related digital tools, while Olawade et al. (2024) highlighted equity and governance as core prerequisites for ethical AI in mental health contexts. Burr et al. (2020) and Boucher and Raiker (2024) discussed the need for ethical design principles, including data privacy, consent and the minimisation of psychological harm.

Montag et al. (2019) and Palmer and Burrows (2020) cautioned against persuasive interface designs that exploit user vulnerabilities, arguing for human-centred, transparent systems that prioritise psychological safety over engagement metrics. Collectively, these studies bridge the gap between cyberpsychology (the study of how humans interact with technology) and digital ethics, positioning mental well-being technologies within a moral and social accountability framework.

Interpretation of findings in context

These findings address Research Question 3 and Objective 3, which focus on ethical considerations in implementing digital interventions. They are underpinned by Cyberpsychology as a theoretical lens, highlighting the interaction between human cognition and digital environments. The reviewed literature collectively argues that digital mental well-being tools must uphold ethical integrity by ensuring fairness, transparency and inclusivity to protect users’ psychological well-being. Users are more likely to engage with and benefit from systems they perceive as safe, transparent and respectful of their data. The intersection of ethics and psychology thus bridges theory and practice, reinforcing that successful digital interventions must be as psychologically sound as they are technologically responsible. Therefore, mental well-being technologies must balance innovation with empathy, ensuring psychological safety while leveraging data-driven insights.

Findings in the context of the research problem

The literature highlights the effectiveness of digital interventions in mitigating social media-induced psychological distress, particularly through CBT apps and mindfulness tools that reduce anxiety and depression. However, ethical concerns, such as privacy, equity, and long-term efficacy, must be addressed for sustainable integration. Notably, some studies caution against over-reliance on digital interventions, which could paradoxically increase screen dependency, underscoring the need for balanced design. Beyond clinical outcomes, recent evidence suggests that digital interventions also influence users’ digital self-regulation and sense of autonomy. For example, interventions that integrate self-reflective journaling and feedback loops promote healthier online habits, while impersonal or overly automated systems often fail to sustain engagement (Lee et al., 2023). Moreover, interventions embedded directly within social media ecosystems, such as mental health chatbots or algorithmic nudges, show promise for early identification of distress signals, yet they raise ethical questions regarding surveillance and consent (Buda et al., 2022). Importantly, the results underscore a socio-technological gap in the accessibility and cultural adaptability of digital well-being tools. Research shows that most interventions are designed for Western, English-speaking populations, leaving users in low- and middle-income countries underrepresented (Aboujaoude et al., 2020). This raises questions about digital equity and contextual relevance, particularly in diverse youth populations where social media norms differ significantly. As such, while digital interventions provide scalable solutions, their global psychological impact remains uneven (Vidal et al., 2020).

Overall synthesis

Together, these findings provide a comprehensive answer to the main research question: How can digital interventions mitigate the negative psychological effects of social media on mental well-being?

Evidence from the 28 analysed studies shows that while social media can significantly affect users’ self-esteem, anxiety and emotional health, the integration of evidence-based digital interventions, grounded in CBT, self-determination theory and cyberpsychology, can reduce these risks. Furthermore, the inclusion of ethical frameworks ensures that such interventions remain fair, accessible and psychologically safe, making them viable long-term solutions for supporting teenagers’ and young adults’ mental well-being in the digital age. The synthesis also highlights that no single approach is universally effective. Interventions are most successful when they combine personalisation, theoretical grounding and user co-design (Liu et al., 2022). A purely technological or clinical lens is insufficient, and effective interventions must account for social context, user motivation and emotional literacy. This suggests that hybrid, interdisciplinary models integrating psychology, data science and digital ethics hold the most potential for future well-being innovation.

Practical implications

The implications of this research are twofold. Theoretically, it contributed to digital mental well-being literature by integrating effectiveness data with ethical considerations, thereby offering a more refined understanding of technology’s role. In practice, the results suggested that interventions must be context-sensitive, ethically designed and tailored to user needs to ensure sustainable mental well-being outcomes. This positions digital well-being not as a universal solution, but rather as a conditional tool whose effectiveness depends on thoughtful implementation and governance. Additionally, the research underscores the importance of digital literacy and mental well-being education in amplifying intervention outcomes. Users who understand how algorithms influence their emotions or how to critically manage social comparison are more likely to benefit from digital mental health tools (Orben et al., 2022). Therefore, educational institutions and policymakers should consider integrating digital well-being curricula that complement intervention-based strategies. Ethically, the study reinforces that privacy-preserving mechanisms, informed consent and inclusivity are not merely procedural but foundational for trust and user engagement (Buda et al., 2022). Without these safeguards, even the most technologically sophisticated interventions risk exacerbating the very distress they seek to alleviate

Limitations and recommendations

Despite achieving its objectives and offering meaningful insights into how digital interventions mitigate the psychological effects of social media, this study had several limitations. A primary limitation was the reliance on secondary data, which prevented the inclusion of first-hand perspectives from users or developers of digital mental health tools. This constrained the depth of experiential insight. However, the limitation was mitigated by drawing on a broad range of peer-reviewed sources, enabling cross-study comparison and analytical breadth (Tripathy, 2013). Publication bias presented a further limitation, as positive intervention outcomes are more frequently reported than neutral or unsuccessful ones. To address this, the study intentionally incorporated literature that critically examined challenges, mixed outcomes, and intervention limitations, supporting a more balanced interpretation of existing evidence (Tripathy, 2013). The rapidly evolving nature of digital technologies also affects the timeliness of data, as earlier studies may not fully reflect current developments. This limitation was minimised by prioritising research published between 2019 and 2025, with older foundational sources included only where conceptually necessary (Tripathy, 2013). Generalisation of findings was limited by contextual factors such as age, cultural and regional differences in social media use and mental health interventions. The focus on young adults and teenagers, identified as particularly vulnerable to social media-related distress, provided analytical clarity but reduced applicability to other populations. Future research should therefore incorporate mixed-method designs and primary data to validate and extend these findings. Although the study relied exclusively on publicly available secondary data, ethical considerations remained central, particularly when the analysed studies involved qualitative narratives or case descriptions. All data were discussed in a generalised and respectful manner to protect privacy and dignity (Tripathy, 2013). As the study was classified as low risk and involved no human participants, no consent procedures were required. Overall, the research adhered to institutional and international ethical standards for secondary qualitative research, demonstrating respect for intellectual property, confidentiality in the original studies, and academic transparency, in line with principles of honesty, accountability and respect (Tripathy, 2013).

Conclusion

This study explored how digital interventions can mitigate the negative psychological effects of social media on mental well-being, with a focus on secondary qualitative data. Drawing from 28 peer-reviewed academic sources, the research examined the main psychological effects of social media use, such as anxiety, depression, loneliness and social comparison as well as how digital tools like mobile mental well-being apps, online therapy platforms and AI-driven chatbots can provide effective interventions. The study was grounded in social comparison theory, CBT and self-determination theory, which provided a theoretical foundation for understanding user behaviour, motivation and emotional regulation in online contexts. The literature was reviewed thematically, revealing that while social media use contributes to psychological distress, digital interventions show significant promise in reducing these effects by promoting self-awareness, emotional resilience and healthier online habits. Overall, this study achieved its purpose by demonstrating that digital interventions have substantial potential to mitigate the psychological impacts of social media, provided they are developed and implemented responsibly. By integrating psychological theory, ethical design, and technological innovation, digital mental well-being tools can play a transformative role in promoting healthier social media engagement and improved mental well-being. This research contributes to the growing field of cyberpsychology and offers a foundation for both academic inquiry and practical application. Ultimately, it reaffirms that technology, when ethically and thoughtfully applied, can be part of the solution to the very mental well-being challenges it once seemed to amplify.

Acknowledgements

This article is based on research originally conducted as part of Anjali S. Morar’s Honours dissertation titled ‘Cyberpsychology in the Digital Age: Addressing Social Media’s Mental Well-being Effects Through Digital Interventions’, submitted to the Emeris University in 2025. The thesis was supervised by Brian Maodza. The thesis was reworked, revised, and adapted into a journal article for publication. The original thesis is available at: https://drive.google.com/file/d/1CYimQ4KYh0WawY8lfoRxmzS6oW2SqTxB/view?usp=sharing.

The authors declare that they have no financial or personal relationships that may have inappropriately influenced them in writing this article.

Brian Maodza: Conceptualisation, Supervision and Writing – review & editing. Anjali S. Morar: Conceptualisation, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualisation, Writing – original draft and Writing – review & editing. All authors reviewed the article, contributed to the discussion of results, approved the final version for submission and publication, and take responsibility for the integrity of its findings.

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

The data that support the findings of this study are available from the corresponding author, Brian Maodza, upon reasonable request.

The views and opinions expressed in this article are those of the authors and are the product of professional research. They do not necessarily reflect the official policy or position of any affiliated institution, funder, agency, or publisher. The authors are responsible for this article’s results, findings, and content.

References