Serial Position Effect.

01Overview

The serial position effect describes how position in a sequence determines memorability. Items at the start of a list are remembered well — the primacy effect, because they had time to transfer into long-term memory. Items at the end are remembered well — the recency effect, because they're still in working memory. Items in the middle are the most likely to be forgotten.

This U-shaped recall curve applies to navigation menus, feature lists, onboarding sequences, email campaigns, and anywhere else that items compete for limited attention.

02Detailed explanation

Murdock's 1962 free-recall experiments showed the classic curve: when participants recalled a list of words, recall was highest for the first few and last few words, with a valley in the middle. The mechanisms are distinct: primacy relies on rehearsal and long-term memory encoding; recency relies on working memory freshness.

• The primacy effect is disrupted by distraction between learning and recall — working memory gets cleared, and the recency advantage disappears.
• The recency effect is disrupted by any significant delay before recall — the last few items fade as they are no longer in working memory.
• In navigation: tab bars with Home and Profile at position 1 and 5 see higher engagement than functionally equivalent items in positions 2–4, even when controlling for the importance of the function.

03Why it exists

Working memory is limited to roughly 4–7 chunks at a time. The brain defaults to the most recent data (recency) and the data with the most rehearsal time (primacy). Middle items compete with both ends for limited storage and lose.

04Effects on users

• Navigation items in the middle of a long nav bar are clicked less frequently than those at the start and end — even when they're higher-priority by function.
• In a feature comparison table, the features at the top and bottom shape users' overall evaluation more than middle features.
• Onboarding step sequences: users remember their first and most recent impressions; mid-flow steps need to be especially clear or they'll be forgotten by the time users need to apply them.
• Email campaigns: the first and last messages in a nurture sequence have higher read rates; mid-sequence messages routinely underperform regardless of content quality.

05Effects on designers & teams

• Meeting agendas: important decisions are better placed at the start or end — not buried in "other business" in the middle of a long agenda.
• Critique structure: feedback given in the middle of a long session is the most likely to be forgotten or conflated with other points.
• Roadmap prioritisation: when reviewing a long feature list, middle items are systematically underweighted. Present priorities in a format that doesn't rely on middle position.

06Practical takeaways

• Put your most important navigation items first or last — never in the middle of a long list.
• In feature comparison tables, lead and close with your strongest differentiators.
• Audit your onboarding sequence: what's in the middle? Is it the least important step, or did it end up there by accident?
• For long content, use structural breaks and headings to create multiple "start and end" positions rather than one long undifferentiated middle.
• In user testing, rotate the order of presented concepts to control for serial position effects on evaluation.

07Design examples

08Ethical risks

The serial position effect is used manipulatively when negative or important information is deliberately buried in the middle of a list — terms and conditions sequenced so the significant liabilities appear after users' attention has drifted, or pricing structures that put the expensive-but-important items in middle positions to reduce their salience.

Structural fairness means important information goes where it will be noticed — regardless of whether it's positive or negative.

09Related biases