In the last years, several legislative initiatives are addressing children’s rights protection and promotion in the digital environment. Different regulatory techniques are adopted: from the EU strategy of data (including GDPR, DSA, and the AI Act), we observe the adoption of a risk-based approach aiming to allocate due diligence obligations for each economic operator (i.e. data controllers, digital services providers, developers, and deployers) to continuously monitoring risks. However, methodologies to develop tailored assessments to protect vulnerable categories of users, like children, are not identified by the EU law: appropriate technical and organisation measures shall be identified and implemented to mitigate / prevent harmful consequences for data subjects/users, but compliance activities are not affecting traditional private law mechanisms to protect minors in their digital experience. Conversely, other legal systems introduce age-restricted paradigms for certain digital services that are required to ban access to users based on their age (like the prohibition for under 16 years old users to get an account on certain social media under the Australian Online Safety Amendment -Social Media Minimum Age- Act 2024). Other legal systems foster cooperation between economic operators and parents (or guardians) to ensure parental oversight for an active and continuous care of the child in the digital dimension, by requiring the providers to implement specific technical tools to support effective supervision and intervention tailored to “the age and developmental stage of the child and adolescent” (ECA Digital Statute in Brazil Act. Law N. 15,211/2025).
Considering the diverse policy and technical approaches emerging worldwide to protect minors, alongside the fragmented scientific evidence concerning their effects, it becomes urgent to ask which interventions are effective and which are not. What would “effectiveness” mean in this specific context, and for whom or for what purpose? It is worth considering whether, at times, such legal initiatives may achieve conflicting outcomes, such as improved mental health at the cost of greater social exclusion, or reduced online exposure at the cost of diminished digital literacy among children, or being just not applied (Barnes C, Hall A, Mantach S, Oldmeadow C, Attia J, Backholer K et al. Assessing early effects of Australia’s Social Media Minimum Age Act on adolescents’ social media use: observational study BMJ 2026; 393 :e363695 doi:10.1136/bmj-2026-363695 ). For example, access and use restriction are not necessarily the solution and can give rise to a paradoxical effect. On the one hand, limiting the use of digital technologies may render them even more appealing to children and adolescents; on the other hand, it can leave them ill-equipped for an aware and healthy use of digital tech (Obed Vargas et al., 2025).
In this domain, characterised by common risks and needs for children as vulnerable users, law and policy making, therefore, cannot avoid carrying out a socio-technical appraisal of the enacted or proposed interventions in order to foster harmonised, effective, and efficient solutions for a safer and child-friendly digital environment. Indeed, children’s participative rights enhancement, autonomy, transparency, and capacity of discernment shall be embedded as dimensions to assess even the usability and acceptability of the specific technical safeguard / given policy intervention. In this regard, legal design methodologies are playing an essential role not only to identify the proper legal tool and to translate it into accessible and engaging policy, but also within its early monitoring/assessment in terms of acceptability, usability, and therefore efficacy.
This paper will map the national legal initiatives to foster children’s rights in the digital environment and assess them to analyse the effectiveness of specific provisions. To this end, we will combine interdisciplinary methodologies based on current studies on dark patterns and behavioural changes, aiming to build up multi-layered architecture to be applied within a proposed acceptability and usability stress test of child-friendly interventions and policies.
Figure 1. Intervention matrix against dark patterns, from Bongard-Blanchy et al., 2020
As noticed (Kerstin Bongard-Blanchy, Arianna Rossi, and others, 2023), indeed, a unified strategy to curb the proliferation of dark patterns online was devised, with various levels of parallel interventions that can be implemented (see Figure 1). In that early work, we classified interventions according to their scope (on the vertical axis), namely a) raising awareness of the existence and risks of dark patterns, b) facilitating their detection, c) bolstering resistance towards them, or d) eliminating them from online services. Interventions can further be classified according to the measures that are put in place (horizontal axis), that is, whether they act on the user or on the environment: educational interventions favour users’ agency, regulatory interventions tend to protect them, and technical and design interventions are situated in between. This distinction can serve to identify the actors, who are responsible for implementing the interventions, and to devise appropriate evaluation indicators. This matrix, whose blank spaces are meant to be filled with academic and regulatory experiences, can easily account for addictive design patterns present on social media, videogames, and other digital services, a form of deceptive design pattern.
Figure 2. The behaviour change wheel from the model proposed by Michie et al.
An even more complete framework is offered by the behaviour change wheel (S. Michie et al, 2011), which systematizes concepts and affordances that have proven useful across domains requiring the promotion of target behaviours, such as in healthcare (N. Chiang et al., 2018) and cybersecurity (Alshaikh et al., 2019). The model’s success likely stems from its being simple yet exhaustive and versatile enough both to explain how human behaviour operates and to inform a set of possible interventions capable of encouraging, or discouraging, a given target behaviour. In their seminal work, Michie, van Stralen and West identify three main components of behaviour, which they summarize in the Capability, Opportunity, Motivation Behaviour model (the COM-B model).
Behaviour is thus influenced by capability, understood as individuals’ capacity and comprising both physical capability (skills) and psychological capability (knowledge, skills); by motivation, broadly defined as the brain processes that direct behaviour and encompassing reflective motivation (plans, intentions, evaluations, beliefs) as well as automatic motivation (emotions, desires, impulses); and by opportunity, that is, factors external to the individual, including social opportunities (intrapersonal influences, socio-cultural norms) and physical opportunities (environmental affordances, time, resources, location). All three components are necessary to achieve a target behaviour, except for reflective thinking. Applied to the regulation of digital services for minors, the COM-B model and the matrix of interventions suggest that the European digital rulebook cannot rely on restriction alone, which addresses opportunity at best, while leaving capability and motivation unaddressed.
To reconcile safety, privacy, freedom, participation, and developmental well-being must therefore combine measures that build children’s capability to navigate digital environments, motivate safer and more autonomous behaviour, and reshape the opportunities afforded by design and regulation alike. The interventions shall likely not only target minors, but a range of other stakeholders, such as families and educators.
These are but two possible frameworks for analysing the measures currently in place or that could be put in place to enhance children’s protection online without depriving them of their autonomy.
Essential References:
D. Amram, Non ho l’età, ma… Costruire competenze abilitanti per una società dell’informazione a prova di (in)capacità del minore di età, Milano, 2025, pp. 1 ss.
N. Patti, V. Punzo, R. Romano, Child Vulnerability in the Digital Environment: Comparative Insights and Operational Guidelines, in Opinio Juris in Comparatione, 2025, p. 107 ff.
Obed Vargas-Salas and others, ‘Advances in Policies and Guidelines, and Other Measures Related to Social Networking and Video Game Use in Minors: A Systematic Review of the Literature’ (2025) 15 Sage Open 21582440251393158, 7 <https://doi.org/10.1177/21582440251393158> accessed 2 July 2026.
Kerstin Bongard-Blanchy, Arianna Rossi, and others, ‘”I Am Definitely Manipulated, Even When I Am Aware of It. It’s Ridiculous!” – Dark Patterns from the End-User Perspective’ Designing Interactive Systems Conference 2021 (ACM 2021) <https://doi.org/10.1145/3461778.3462086> accessed 2 July 2026.
Susan Michie, Maartje M van Stralen and Robert West, ‘The Behaviour Change Wheel: A New Method for Characterising and Designing Behaviour Change Interventions’ (2011) 6 Implementation Science 42 <https://doi.org/10.1186/1748-5908-6-42> accessed 2 July 2026.
Nicole Chiang and others, ‘Interactive Two-Way mHealth Interventions for Improving Medication Adherence: An Evaluation Using The Behaviour Change Wheel Framework’ (2018) 6 JMIR mHealth and uHealth e9187 <https://doi.org/10.2196/mhealth.9187> accessed 2 July 2026.
Moneer Alshaikh and others, ‘Toward Sustainable Behaviour Change: An Approach for Cyber Security Education Training and Awareness’ 27th European Conference on Information Systems: Information Systems for a Sharing Society, ECIS 2019 (Association for Information Systems 2020) <https://ksascholar.dri.sa/en/publications/toward-sustainable-behaviour-change-an-approach-for-cyber-securit-2> accessed 2 July 2026.