JH
Jonathan Haber

Default to assuming you’ve been misunderstood

After any important message, assume it landed differently than you intended — then check.

Key takeaways

  • What it is: After any important message, assume it landed differently than you intended — then check.
  • Why it works: The bias operates through overconfidence: we feel understood because the relationship is close, regardless of whether the message was actually received. Deliberately defaulting to the assumption of misunderstanding inverts this overconfidence and replaces it with a verification habit. Cognitive research shows that actively considering alternative interpretations reduces egocentric anchoring — the tendency to assume our own perspective is shared.
  • Evidence: Backed by randomized trials / meta-analyses.
  • Avoid: Asking "Did you understand?" instead of "What did you understand?" — the first invites a yes/no that confirms nothing; the second reveals the actual mental model.

Why it works

The bias operates through overconfidence: we feel understood because the relationship is close, regardless of whether the message was actually received. Deliberately defaulting to the assumption of misunderstanding inverts this overconfidence and replaces it with a verification habit. Cognitive research shows that actively considering alternative interpretations reduces egocentric anchoring — the tendency to assume our own perspective is shared.

How to do it

  1. 1After communicating something important, pause and ask yourself: "How many ways could that be heard differently?"
  2. 2Invite your partner to reflect back what they understood — not "Did you get that?" but "What did you hear me saying?"
  3. 3If their reflection differs, treat it as information rather than a failure on either side.
  4. 4Make this a norm in both directions: neither partner is expected to read minds.

What the evidence says

RCT / meta-analysis

Keysar’s experimental research found that people with close others significantly overestimated communication accuracy, and that this overconfidence was not reduced by actual shared knowledge.

Honest caveat: The research is experimental (lab studies); generalizing to naturalistic relationship communication is reasonable given the robustness across contexts, but field evidence is more limited.

References
  • — Keysar & Henly (2002), "Speakers’ Overestimation of Their Effectiveness," Psychological Science
  • — Savitsky, Keysar, Epley, Carter & Swanson (2011), closeness-communication bias, Journal of Experimental Social Psychology

Common mistake

Asking "Did you understand?" instead of "What did you understand?" — the first invites a yes/no that confirms nothing; the second reveals the actual mental model.

IX Coach builds reflection-checking into structured conversations: after any significant exchange, it prompts each person to state what they heard before either responds further.

Practice this with IX Coach →

Related practices