The Accidental Curriculum
- 3 days ago
- 3 min read
Week 6 · Blog Post 11 · Web 2.0-Based Learning and Performance · Summer 2026
Every time someone scrolls past a wellness reel, a "things my therapist told me" carousel, or a stranger's three-paragraph caption about their anxiety, they're learning something. Not because they meant to. Because the feed teaches whether or not anyone designed it to.
I read four pieces recently that, stacked together, reshaped how I think about building a mental-health platform — and made me a little less smug about the whole project.
We're Already Enrolled
Hussenoeder (2022) makes an argument I can't unsee: social network sites are learning environments, full stop. Their communicative infrastructure — profiles, comments, likes, the whole apparatus — runs two of the oldest learning mechanisms we know. We learn from feedback (a like is a reward; silence is a punishment) and we learn from observation (we watch what others post and quietly absorb the norms). Mostly this happens implicitly, as a side effect of doing something else.
The trouble is what gets taught. He links negative online feedback to depression and following influencers to disordered eating. The feed is biased toward everyone's highlight reel, so observational learning curdles into comparison. If a platform is a classroom, the default curriculum is often a bad one.
For me — someone building a space explicitly for mental-health learning — that's both the warning and the brief. I'm not introducing learning into a neutral environment. I'm competing with one that's already running.
Where People Actually Go
So where does this learning happen, and how? Kanchan and Gaidhane (2023) lay out the landscape: Facebook and YouTube dominate by sheer users, and Facebook, Twitter, and Instagram are the top three platforms health researchers actually use — to recruit participants, mine public posts, and broaden the reach of findings. They also surface the idea I keep circling: "peer-to-peer health care," where people swap what they know about symptoms, treatments, and where to get help. That's not patients being passive. That's a community teaching itself.
Ghahramani et al. (2022) add a useful wrinkle: platforms get used differently. In their review of health-promotion campaigns, Facebook and YouTube skewed toward intervention (pushing content to change behavior), while Twitter and Instagram skewed toward observation (watching what people already do). Same toolbox, different jobs. If I'm designing distribution, that distinction matters — a YouTube explainer and an Instagram listening post are not the same instrument.
Awareness Is the Floor
Here's the line from Ghahramani that I've been quoting at myself: most campaigns are good at creating awareness and bad at everything after. Raising awareness is cheap to measure — reach, impressions, likes — and easy to mistake for impact. But likes aren't behavior. The review found that engagement metrics rarely map onto real-life change.
What separated the more credible interventions was theory. The majority leaned on Social Cognitive Theory — the idea that people adopt behaviors they watch others model, and that self-efficacy (believing you can do the thing) drives whether they try. A few used the Transtheoretical Model, designing content to match where someone sits in the stages of change rather than blasting the same message at everyone.
This lands directly on a principle I've been building my platform around: self-efficacy comes before skill. You can't scaffold a grounding technique for someone who doesn't yet believe they're the kind of person who can self-regulate. Theory isn't academic garnish here — it's the difference between a campaign that informs and one that actually moves someone a stage forward.
Design for Agency, Not Attention
The piece that tied it together was Lee, Ellison, and Hancock (2023) on agentic social media use — using these tools intentionally, with the beliefs, knowledge, and practices to bend them toward your own goals. Their finding that stuck: someone's mindset about social media predicts their wellbeing better than the hours they spend on it. People who feel in control fare better than people who feel used.
They break agency into three buildable parts — an agentic mindset, social-media literacy, and concrete practices (curating a feed, refining what the algorithm serves, muting and unfollowing without guilt). And underneath all three sits self-efficacy again — Bandura's, the same engine Ghahramani's campaigns relied on.
That reframes my whole job. The goal isn't to capture attention the way the default feed does. It's to hand people enough literacy and self-belief that they can use any feed — mine included — on their own terms. A mental-health platform that makes its users more dependent on it has, by this logic, failed.
So I'll put it to you: if every feed is already teaching something, what's the most honest thing a platform can teach its users about how to use it — even when that means teaching them, eventually, how to need it less?
Comments