// Where this came from

The research behind RetroPals.

RetroPals started as a master's research project, not a product idea. This page explains the question it was asking, what the literature said, and how those findings turned into design decisions you can point at in the app.

// The question

How can we design a digital wellbeing tool that fosters autonomous, meaningful, and contextually sustainable self-reflection?

Autonomous, meaning motivated from inside rather than pressured from outside. Meaningful, meaning connected to what the person actually cares about. Sustainable, meaning it still works in month eight.

// The problem

What the literature says about habit apps

Engagement in self-tracking apps spikes on novelty and then falls off a cliff. The evidence is consistent across studies: most wearable users stop within five to seven months, and in one study only 16% were still tracking after around 320 days (Maher et al., 2017; Hermsen et al., 2017; Shin et al., 2019).

The more interesting finding is why. Self-tracking has been shown to increase how much of an activity people do while simultaneously reducing how much they enjoy it (Etkin, 2016). The measurement itself changes the relationship with the behaviour.

Self-Determination Theory offers the mechanism. Streaks, badges, leaderboards, points and rigid daily targets push people toward controlled motivation: acting from guilt, external pressure, or fear of breaking something. Controlled motivation produces compliance while the pressure lasts. Autonomous motivation, grounded in a person's own reasons, is what persists after it stops.

This hits students particularly hard. Student wellbeing is shaped by unstable routines, exam cycles, fluctuating academic workload and seasonal change, which is exactly the context that a fixed daily target handles worst.

// From findings to requirements

Seven design requirements

Each is a constraint the app is built to satisfy, and each maps to something you can see when you use it.

Reflective value: interpret patterns and personal context, not raw metrics

The companion writes a plain-language reflection connecting ratings, notes and attached context

Low extrinsic rewards: no points, XP, badges or competitive leagues

None by default. The EXP layer is opt-in, beta, and warns you before enabling

No punitive mechanics: no broken streaks, penalties or visible failure history

There are no streaks at all, and a missed week records nothing

Low performance pressure: no rigid daily targets or nagging prompts

The cadence is weekly, every step is skippable, reminders are gentle and configurable

Context sensitivity: adapt to workload, stress, sleep, weather and seasonal shifts

Optional attachments for sleep, steps, exercise and weather, fed into the reflection

Autonomy support: users control goals, cadence, style and their own definition of progress

Reflection steps, frameworks, AI, tracking and sync are all individually configurable

Student accessibility: works without expensive wearables or costly subscriptions

Fully functional free plan, no wearable required, biometrics entirely optional

// Honesty

What we can and cannot claim

RetroPals is the applied artifact of that research. The design decisions are literature-derived, and they can be traced.

What has not happened yet is a formal evaluation of whether the finished app achieves the outcome. The intended next step is to evaluate it with validated psychological measures (the Intrinsic Motivation Inventory and the User Motivation Inventory) to test whether AI-assisted reflection genuinely fosters sustainable, self-determined motivation.

So the comparisons elsewhere on this site are best read as what they are: hypotheses the app was purpose-built to test, not results it has already proven. We would rather say that plainly than dress a design rationale up as evidence.

// Status

Is there a study running?

No.

There is no research programme running at this time, no study is open, and no research data is being collected from anyone using RetroPals.

If a study opens in future it will require its own explicit consent document, describing that specific study: what it collects, how long it runs, and how to withdraw at any point. Even then, the design principle stands: free text is never transmitted. A study of this app would work from anonymised numeric ratings, never from the words you wrote.

// Sources

References

  • Etkin, J. (2016). The hidden cost of personal quantification. Journal of Consumer Research, 42(6), 967 to 984.
  • Hermsen, S., Moons, J., Kerkhof, P., Wiekens, C., & De Groot, M. (2017). Determinants for sustained use of an activity tracker: observational study. JMIR mHealth and uHealth, 5(10), e164.
  • Maher, C., Ryan, J., Ambrosi, C., & Edney, S. (2017). Users' experiences of wearable activity trackers: a cross-sectional study. BMC Public Health, 17(1), 880.
  • Ryan, R. M., & Deci, E. L. (2000). Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. American Psychologist, 55(1), 68 to 78.
  • Shin, G., Jarrahi, M. H., Fei, Y., Karami, A., Gafinowitz, N., Byun, A., & Lu, X. (2019). Wearable activity trackers, accuracy, adoption, acceptance and health impact: a systematic literature review. Journal of Biomedical Informatics, 93, 103153.

Note: these citations are being checked against the original thesis bibliography. Volume, issue and page numbers should be verified before this page is treated as a citable source.

A calmer way to keep going.

One reflection a week. No streaks to break, no scores to chase, no pressure to perform.