So ’ve been thinking a lot lately about the concept of “digital phenotyping”. The term itself feels a little detached, almost clinical. But at its core, it’s about something deeply human: understanding who we are through the patterns we create.

This isn’t a new idea. Many of us have been working on it, turning the tools of technology toward capturing not just personal health, but something larger—something shared. And while digital phenotyping has come a long way, I think we’re at a moment where we need to shift our focus.

The Individualistic Origin Story

Digital phenotyping started as a way to measure individuals. You. Me. Our smartphones tracking how far we walk, how we interact with apps, how often we sleep. It was part of the quantified-self movement, promising us personal insight and optimization (Torous et al., 2016).

But it’s always felt limited to me. It’s hard to understand the full picture of health—physical, mental, or social—when we isolate the individual from the world they’re a part of. So much of who we are exists in the spaces between us, in the relationships and communities we build.

Digital phenotyping, in its current form, struggles to capture that. It silos the data. It tries to tell the story of an individual heartbeat without acknowledging the broader ecosystem of heartbeats it belongs to (Vayena, 2021).

Fractals, Ecosystems, and a Bigger Picture

There’s a lesson here, one I’ve seen reflected in nature. Fractals—those self-repeating patterns that show up everywhere from tree branches to galaxies—remind us that the micro and the macro are deeply connected. You don’t need to see the entire picture to understand the whole; each part reflects the structure of the larger system.

What if we thought of digital phenotyping the same way? What if, instead of trying to understand individuals in isolation, we looked at them in the context of the larger ecosystems they’re a part of? What if we focused less on measuring the self and more on measuring the connections?

This isn’t a groundbreaking insight. It’s something I see many of us moving toward, a gradual shift in how we use tools like generative AI and machine learning. These technologies are increasingly capable of recognizing the kinds of patterns that define ecosystems, not just individuals (Pan et al., 2022).

Disconnection & Disease

It’s worth noting how much of our understanding of health points back to connection. Studies have shown that disconnection—whether at the cellular level, in social structures, or in ecosystems—often underlies disease (Schmeller et al., 2020). A cell that becomes isolated from its network is more likely to become cancerous. A community that loses its connections is more likely to experience stress and decline. “There is no community without communing”.

In digital phenotyping, focusing on isolated individuals risks missing the forest for the trees. The more we lean into an ecosystem perspective, the more potential we have to understand and address health at its roots.

A Call to Collaborate

None of this is new. These ideas—of connection, of fractals, of ecosystems—have been with us for a long time. And many of us have already been exploring how to apply them to digital phenotyping. I’m not claiming to have a new framework or a definitive path forward. Instead, I’m simply saying: the technology is here. The tools are in our hands. It’s time to use them to focus more on the de-siloing nature of digital phenotyping and to prioritize the ecosystem level.

So, If you’re someone working on these kinds of problems, I’d love to connect. Let’s share ideas, build tools, and refine our approaches together. This isn’t about any one of us; it’s about what we can do collectively. The potential is enormous—not to create something entirely new, but to collaborate on something deeply necessary.

As Jamie Wheal puts it, “Stay awake. Build stuff. Help out”.

yours,

Bradley C & GPT

References

• Torous, J., Kiang, M. V., Lorme, J., & Onnela, J. P. (2016). New tools for new research in psychiatry: A scalable and customizable platform to empower data-driven smartphone research.
  JMIR Mental Health, 3(2), e16. Link
• Vayena, E. (2021). Digital phenotyping: An epistemic and methodological analysis.
  Philosophy & Technology, 34, 339–358. Link
• Pan, S., et al. (2022). Digital phenotyping in health using machine learning approaches: A systematic review.
  JMIR Bioinformatics and Biotechnology, 3(1), e39618. Link
• Schmeller, D. S., et al. (2020). Biodiversity loss, emerging pathogens, and human health risks.
  Biodiversity and Conservation, 29, 3095–3102. Link