The Science Behind FlowStates

Research on streak-based tracking reveals a troubling pattern: roughly 78% of users abandon their habits entirely after a single streak break. This is known as the abstinence violation effect — the psychological tendency to treat a single lapse as total failure, triggering abandonment of the entire behavior.

However, Lally and colleagues demonstrated that missing a single day of practice does not materially affect the trajectory of habit formation. Their longitudinal study found that 80% adherence produces nearly identical outcomes to 100%. What matters is the overall consistency rate over time, not an unbroken chain. This is why FlowStates shows you consistency rates rather than streak counters — it reflects what the evidence actually says about behavior change.

A meta-review spanning 30 systematic reviews and 18,347 participants found a consistent positive effect of exercise on cognitive function, with a standardized mean difference of SMD = 0.33. The mechanism is well-understood: exercise increases levels of brain-derived neurotrophic factor (BDNF), a protein essential for neuroplasticity, with a measured effect size of g = 0.46.

Research suggests that the timing of exercise relative to learning matters. Van Dongen and colleagues found that exercise performed four hours after learning significantly improved retention two days later, compared to immediate exercise or no exercise. This points to two distinct windows: a priming window (exercise before learning enhances acquisition) and a consolidation window (exercise hours after learning strengthens memory encoding).

Interleaving — the practice of alternating between different types of activities within a study or practice session — produces dramatically better long-term learning than blocked practice. On delayed tests, interleaved practice groups scored 63% compared to 20% for blocked practice groups. A meta-analysis of 42 studies found a transfer advantage of SMD = 0.55.

The underlying mechanism is called contextual interference. When you switch between domains, your brain must repeatedly retrieve and reconstruct strategies, strengthening the underlying neural pathways. Interleaving feels harder in the moment — performance during practice is often worse — but this difficulty is what Bjork termed a desirable difficulty, one that produces superior retention and transfer.

The spacing effect is one of the most robust findings in all of cognitive psychology. Cepeda and colleagues reviewed 271 studies spanning over a century and found that distributed practice was superior to massed practice in 259 of them — a consistency rate rarely seen in behavioral science. Spaced practice typically yields a 10-30% improvement in long-term retention.

The principle dates back to Ebbinghaus in 1885, who documented the forgetting curve — the exponential decay of memory over time without review. His findings were replicated in 2015 with modern methodology. Spaced repetition systems exploit this by scheduling reviews at expanding intervals, catching memories just before they fade. The implication for multi-domain practice is clear: shorter, more frequent sessions across domains outperform long, infrequent marathons.

Deci, Koestner, and Ryan conducted a landmark meta-analysis of 68 experiments and confirmed that tangible rewards — badges, points, leaderboards — consistently diminish intrinsic motivation. This effect is known as motivation crowding: when an external reward is attached to an activity someone already enjoys, the activity begins to feel like work rather than play.

The closely related overjustification effect describes how people who receive rewards for inherently enjoyable activities begin to attribute their motivation to the reward rather than the activity. When the reward is removed, motivation drops below the original baseline. Self-Determination Theory (SDT) explains this through three fundamental needs — autonomy, competence, and relatedness. Gamification mechanics undermine autonomy by shifting the locus of control from the person to the reward system. FlowStates is deliberately designed without badges, points, or streaks for this reason.

Research suggests that cognitive decline accelerates when cognitive load exceeds sustainable thresholds — one study observed a 40% increase in cognitive load leading to measurable performance degradation within three weeks. Engaging in varied skill domains provides natural recovery: the neural networks used for music practice, for example, are largely distinct from those used during intense programming, allowing one system to rest while another is active.

A 2019 study published in Nature found that skill variety is protective against cognitive decline, building what neuroscientists call cognitive reserve — a buffer of neural flexibility that helps maintain performance under stress. Systematic reviews consistently show that cross-domain engagement correlates with reduced burnout risk, as varied activity prevents the overuse of any single cognitive system.

Lally and colleagues found that the average time to form a new habit is 66 days, with a range of 18 to 254. The key measure is automaticity — how reflexive the behavior becomes — not streak length. The Self-Report Behavioural Automaticity Index (SRBAI) is a better predictor of long-term habit maintenance than any streak counter.

Implementation intentions — specific plans of the form “when X happens, I will do Y” — are one of the most effective habit-formation strategies, with an effect size of d = 0.65 across multiple meta-analyses. Fogg's behavioral model (B = MAP) frames behavior as a function of Motivation, Ability, and a Prompt. The research consistently suggests that reducing friction (increasing Ability) and establishing contextual cues (Prompts) are more effective than relying on motivation, which fluctuates naturally.

Root-Bernstein and colleagues analyzed the extracurricular activities of Nobel laureates and found striking patterns: laureates were 22x more likely to perform or act, 12x more likely to write fiction, and 7x more likely to paint than average scientists. This suggests that breadth of engagement is not a distraction from depth — it may be a catalyst.

Araki's framework for multi-domain development describes three dimensions: Breadth (the number of domains), Depth (mastery within each), and Integration (connections between them). Career shapes are often described as I-shaped (deep in one area), T-shaped (one deep plus broad), Pi-shaped (two deep), or Comb-shaped (multiple areas of depth). Epstein's research on sampling periods suggests that early breadth — trying many domains before specializing — correlates with higher eventual performance and career satisfaction. The evidence points to a model where both breadth and depth matter, and the real power comes from integration across domains.