Confirmation bias

확증 편향

An initial belief can draw attention toward matching reviews.

···
html
<div class="confirm"><div class="confirmHead">FIRST BELIEF <button id="belief">LIKELY GOOD</button></div><div class="reviewList"><div class="positive">GOOD FIT <small>Easy to set up</small></div><div class="mixed">MIXED <small>Learning curve</small></div><div class="critical">NOT FOR ALL <small>Limited options</small></div></div><div class="confirmFoot">SAME REVIEWS / DIFFERENT FIRST LOOK</div></div>
css
@media(min-width:700px){body>div{transform:scale(1.45)}}.confirm{width:min(88%,570px);font:800 clamp(10px,max(1.7vmin,1vw),14px)/1.1 sans-serif}.confirmHead{display:flex;justify-content:space-between;align-items:center;gap:5px}.confirmHead button{background:var(--accent);color:white;border:0;padding:clamp(6px,2vmin,12px);font:inherit;cursor:pointer}.reviewList{display:flex;gap:clamp(5px,2vmin,14px);margin:clamp(12px,4vmin,28px) 0}.reviewList div{flex:1;min-width:0;min-height:clamp(70px,20vmin,145px);background:var(--surface);border:2px solid var(--line);padding:clamp(6px,2vmin,15px);opacity:.58;transition:.4s}.reviewList div.spot{opacity:1;border-color:var(--accent);transform:translateY(-4px)}.reviewList small{display:block;margin-top:clamp(7px,2vmin,15px);font-size:clamp(10px,max(1.5vmin,1vw),12px);font-weight:600}.confirmFoot{color:var(--accent);letter-spacing:.06em}
js
const cards=[...document.querySelectorAll('.reviewList div')];const button=document.querySelector('#belief');let positive=true;function show(){button.textContent=positive?'LIKELY GOOD':'LIKELY BAD';cards.forEach((card,i)=>card.classList.toggle('spot',i===(positive?0:2)))}let manual=false;show();setTimeout(()=>{if(!manual){positive=false;show()}},500);const timer=setInterval(()=>{positive=!positive;show()},2400);button.addEventListener('click',()=>{manual=true;clearInterval(timer);positive=!positive;show()});

Confirmation bias is a tendency to seek or accept information that supports an existing belief. With the same set of reviews, an initial expectation can change the order in which evidence gets attention.

The demo holds three fictional reviews constant and changes only the initial expectation, highlighting different cards. It does not measure real clicks or memory.

Do not exploit the bias by sorting only favorable reviews to the top. Show positive and critical feedback with context, and review disconfirming evidence within the team as well.

When to use

Use it to check whether reviews and research expose disconfirming evidence.

Open as page ↗