pattern playground

Cognitive forcing functions

Added

Interventions that add analytical friction at the moment of accepting a system recommendation, so users engage their own judgement instead of rubber-stamping output.

The term comes from human-factors engineering’s forcing function tradition and was applied to AI-assisted decision-making by Buçinca et al. (2021). The mechanism is the same across all variants: interpose a moment of active cognition between system output and user acceptance.

Over-reliance runs through several pathways: anchoring on the system’s answer, automation bias that treats machine output as authoritative, cognitive offloading where evaluating feels costlier than accepting. Cognitive forcing functions interrupt all of them by requiring the user to act (wait, decide first, articulate a reason, complete a gap) before accepting. The friction is load-bearing: without it, users accept wrong answers at higher rates.

When to apply

Cognitive forcing functions earn their cost when:

  • Stakes are high — the consequence of accepting a wrong answer matters (medical, legal, financial, safety-critical)
  • The user has domain competence — they can evaluate the output if given the chance; forcing functions don’t help when the user has no basis for judgement
  • The task supports verification — thought can land on a checkable answer. Fok & Weld (2023) argue many decision tasks fundamentally resist verification regardless of intervention; in those cases no amount of forcing rescues reliance
  • The decision is discrete — there’s a clear moment of accept/reject. Ashktorab et al. (2024) show that in open-ended generation, users bypass the structure by appending AI output alongside their own correct answers — the intervention assumed a seam the task didn’t have

They’re not warranted when:

  • Stakes are low — a wrong autocomplete in casual writing doesn’t need a checklist
  • The user lacks the competence to evaluate — forcing them to think doesn’t help if they can’t tell right from wrong; see learnability
  • Speed is the primary value — in time-critical workflows, friction can cause worse outcomes than the over-reliance it prevents

Susceptibility to over-reliance varies with need for cognition (the tendency to engage in effortful thinking), and this produces a tension in who forcing functions actually help. Low-NfC users are the population of concern: they default to cognitive offloading and accept AI output most readily. But Buçinca’s own data shows the intervention benefits high-NfC users more: the friction only works on people already motivated to engage with it. One reading, from Vasconcelos et al. (2022), is that engagement is a rational cost-benefit choice rather than a cognitive bias: users evaluate when verification is cheaper than acceptance.

The family

Each member of the family uses a different mechanism to interpose judgement. They vary in friction intensity, anchoring resistance, and implementation cost:

  • Wait before reveal — delay the recommendation for a fixed interval.
  • On-demand reveal — hide the recommendation behind an explicit user action.
  • Update after initial judgement — user records their answer first, then sees the system’s and can revise.
  • Checklist — structured review of considerations before accepting.
  • Partial explanations — show only part of the reasoning; user must complete the logic.
  • Rationale logging — user records why they accepted or rejected.
  • Human goes first — user performs the task unaided; system compares and teaches.

Resources & references

Consequences

  • the actor has had to do something — wait, commit an answer first, name a reason, finish an incomplete argument — before accepting, so the judgement on record is theirs; where that something was writing down why, the reasons accumulate into a trail somebody else can read
  • the friction is real and paid on every decision, so it holds only where the stakes keep justifying it; where they don't, actors learn the shortest path through it
  • the help lands unevenly — it works best on actors already inclined to think, and least on the ones most likely to defer, who are the reason for it

Related patterns

Precedes

  • Activity log — the actor has had to do something — wait, commit an answer first, name a reason, finish an incomplete argument — before accepting, so the judgement on record is theirs; where that something was writing down why, the reasons accumulate into a trail somebody else can read

Enacts

  • Agency — inserts a beat of thought between output and acceptance, so the judgement is the actor's own
  • Learnability — keeps the actor's own judgement in use — and only helps those who can already evaluate the output
  • Temporality — the wait variants make a deliberate delay the friction, interrupting automatic deferral

Complements

  • Approval gate — deliberate friction that keeps approving a judgement rather than a reflex
  • Generated content — long-form output especially prone to over-reliance because verification cost is high

Related

  • Checklist — structured review of considerations before accepting.
  • Human goes first — user performs the task unaided; system compares and teaches.

Preceded by

  • AI completion — inline generation makes accepting cheaper than evaluating, so over-reliance accrues by default — deliberate friction is the countermove
  • Explanation — a complete and plausible explanation can make deferring easier rather than harder: the actor now holds a justification they didn't have to think through, which is the state a forcing function exists to interrupt
  • Expressed uncertainty — confidence is part of the output's surface — hedged phrasing, ranges, marks on the doubtful spans — so the actor's scrutiny can land where it is actually needed
  • Next-best action — where the stakes are high, the recommendation being easy to accept is the problem rather than the feature — the accept needs a beat of deliberate thought built into it
  • Suggestion — the actor's option space now contains a concrete proposal, which is the help and also the anchor: shown before they have formed their own idea, it narrows what they go on to consider, and the alternatives they never reached aren't visible to them as a loss
  • Transparent reasoning — visible reasoning anchors as readily as it informs: a trail that looks thorough invites the actor to stop reading it and start relying on it, which is the state a deliberate interruption exists for