Why do I change correct answers?
Few test experiences are more frustrating than discovering you had it right and changed it. That memory is vivid, which is exactly why it misleads.
The numbers say change
Research on answer changing has repeatedly found the opposite of the folk advice. In a study of more than 1,500 students' exam papers, Kruger, Wirtz and Miller found that 51 percent of answer changes went from wrong to right, 25 percent from right to wrong and 23 percent from one wrong answer to another. Across students, 54 percent of those who changed answers improved their score and 19 percent lowered it. The belief that you should stick with your first instinct is what they called the first instinct fallacy.
Why the first-instinct belief persists
| Mechanism | What happens |
|---|---|
| Regret asymmetry | Changing to a wrong answer feels worse than staying wrong |
| Memory bias | Painful changes are remembered more vividly |
| Advice culture | Teachers and peers repeat "trust your first instinct" |
| Overlooked gains | Wrong-to-right changes feel unremarkable |
Why the myth outlives the data
The same researchers tested why the myth survives despite the numbers. Their answer was counterfactual thinking: after changing from right to wrong, it is easy to imagine "if only I hadn't changed," which intensifies regret and makes the episode memorable. Sticking with a wrong answer produces much less of that thought, so those cases fade. In their studies, people's memories of their own answer changes were skewed toward the painful cases, even when their actual papers showed the opposite pattern.
When to change and when to keep
| Situation | Better choice |
|---|---|
| You noticed a rule or detail you missed | Change |
| You misread the question the first time | Change |
| You're anxious but have no new reason | Keep |
| Two options still seem equally good | Consider time and move on |
| You're rushing at the end of a timed test | Keep unless clearly wrong |
A short scenario
A test taker flags five matrix items for review. On three, she spots a missed rotation rule and changes her answer; two of those turn out correct. On two others, she changes out of vague doubt and loses one point. The pattern shows the value of changing with evidence rather than feeling.
A quick review routine
- Flag uncertain items as you go.
- On review, ask what new information you've noticed.
- Change only when you can state the reason.
Answer changing questions
Is your first answer usually right?+−
Not reliably. In the study above, changes went wrong-to-right twice as often as right-to-wrong.
Why do I regret changing answers so much?+−
Right-to-wrong changes create strong regret and are remembered more vividly.
Should I review answers in a timed test?+−
If time allows, review flagged items with a specific focus.
What is the first instinct fallacy?+−
The mistaken belief that sticking with your first answer is generally better.
Continue exploring
Explore IQ and reasoning explained
Return to the topic hub to browse all guides, questions, comparisons and reflection resources.
Start with the main guide
This guide explains the broader topic before you explore individual questions and concepts.
Understand the key ideas
Explore the concepts connected to this guide.
Fluid reasoning explained
Fluid intelligence solves new problems; crystallized intelligence uses what you've learned. How the two differ, how they change with age, how tests measure them.
Read guideHow visual pattern recognition works
How the brain turns shapes and lines into recognizable patterns: feature detection, grouping principles, top-down expectations.
Read guideInductive vs deductive reasoning
Deductive reasoning moves from rules to guaranteed conclusions; inductive reasoning moves from examples to probable rules. Clear definitions, examples, common errors.
Read guideRelated guides
Continue with practical questions and focused explanations.
Analogical reasoning
Analogical reasoning maps a relationship from one situation to another. How it works cognitively, the common analogy types in tests and why it matters for learning.
Read guideAre IQ tests mostly pattern questions?
IQ-style tests go beyond pattern puzzles. The main question types found in reasoning tests and professional assessments, what each measures.
Read guideAre visual IQ questions easier for some people?
Some people see visual patterns instantly while others prefer words or numbers. Why cognitive profiles differ, the role of experience and strategy.
Read guideCan you learn pattern recognition?
Pattern recognition improves with practice, but mostly within the domain you practice. How it develops, what training can and can't do.
Read guideExplore in daily life
See how these patterns can appear in everyday settings.
Pattern recognition at work
Spotting trends in data, recurring customer issues and early warning signs is pattern recognition in action. Where it creates value at work, where it misleads.
Read guidePattern recognition in learning
Learning is largely noticing structure: recurring problem types, principles behind examples and connections between topics. How pattern recognition speeds learning.
Read guideCompare related patterns
Explore where related experiences may overlap or differ.
IQ test vs aptitude test
Aptitude tests measure readiness for specific tasks or training; IQ tests estimate general reasoning. How they differ in purpose, content and use.
Read guideIQ vs EQ
Emotional intelligence and cognitive ability are different constructs with different evidence bases. What each measures, how EQ is assessed, what research supports.
Read guideTake the next step with structured self-reflection
Use the test as a starting point for noticing everyday patterns and deciding what you may want to explore further.