Visual match / non-match decisions
Compare one or two target shapes with a row of original geometric candidates, then choose Present or Absent. Practice, timed, fixed-question, and custom modes report accuracy by trial type, median correct response time, net throughput, and first-half versus second-half pace.
Decide whether either target shape appears in the candidate row.
Decide whether any target symbol appears in the candidate row.
Target symbols
Candidate symbols
.
Net throughput = (correct − incorrect) ÷ elapsed minutes. It describes pace in this page only; it is not an IQ or standardized processing-speed score.
Attempted
0
Correct
0
Incorrect
0
Missed / timeout
0
Present-trial accuracy
0%
Absent-trial accuracy
0%
Median correct RT
—
Net throughput
0/min
First-half accuracy
0%
Second-half accuracy
0%
No validity flags were detected
This symbol search test is a visual scanning and same/different decision game built from original SVG geometry. It does not copy a standardized IQ test, reproduce Wechsler materials, use age norms, or calculate an intelligence score.
The result describes how many page-specific questions you answered accurately and how quickly correct answers were made under the chosen target count, candidate count, symbol similarity, and endpoint.
Each question shows one or two targets above five to nine candidate symbols. Select Present when at least one target appears anywhere in the candidate row; select Absent when neither target appears. Keyboard shortcuts P and A provide the same actions as the touch-friendly buttons.
Practice contains ten questions and gives short correctness feedback. Thirty- and sixty-second modes continue until time expires. Fixed mode contains thirty questions for repeat comparison. Custom mode can end by a bounded question count or a bounded time limit.
Symbols are composed from project-owned SVG primitives: polygons, curved outlines, internal lines, crosses, arcs, and center marks. They render as vector paths rather than font glyphs, so their structure does not depend on a device font. Decisions never depend on color.
Higher similarity draws targets and distractors from a smaller pool of related forms. That increases shared features without making symbols identical or unreadable. Candidate order and present/absent status use seeded question generation so scoring logic can be reproduced in automated tests.
Attempted is the number of answered questions. Correct and incorrect divide those answers. A timed session can report one missed current question when time expires. Present-trial accuracy uses only questions where a target was present; absent-trial accuracy uses only questions where no target appeared.
Median correct response time includes only correct answers and reduces the influence of one very slow question. First-half and second-half accuracy split answered questions near the midpoint. These figures describe the current symbol set and interaction, not a standardized processing-speed construct.
Net throughput is calculated as correct answers minus incorrect answers, divided by elapsed minutes. The subtraction prevents random rapid guessing from looking equivalent to accurate scanning. A negative value is possible when incorrect answers outnumber correct answers.
Throughput should be read with present and absent accuracy. Two sessions can share a throughput value while using different combinations of speed and error. For comparison, keep the endpoint, target count, candidate count, similarity, present probability, display, and input method fixed.
A present trial can end as soon as a convincing target match is found. An absent trial generally requires checking enough candidates to decide that no target appears. The row length, position of a present match, number of targets, and visual similarity can therefore change response time.
This page reports accuracy separately but does not claim that one person's present-versus-absent difference has a diagnostic interpretation. It is a normal consequence of the decision rule and the generated row.
The site's Visual Search Test asks you to locate one odd item in a larger field and records board completion. Symbol Search presents small question rows and requires a binary present/absent judgment against explicit targets. The input, stopping rule, and error types are different.
Use Visual Search when you want a spatial target-finding board. Use Symbol Search when you want repeated matching decisions with balanced present and absent questions.
Screen size changes symbol spacing, while touch and keyboard controls have different action times. Browser load, display refresh, familiarity, and motor input also affect response timing. Keep the same setup for a useful personal comparison.
This game is not an IQ test, clinical assessment, school placement tool, age percentile, or formal processing-speed evaluation. It has no standardized norms or professional cutoffs and must not be presented as equivalent to copyrighted assessment material.
Completed sessions save settings, aggregate counts, timing summaries, and validity flags in the current browser. They do not store the complete generated SVG question sequence or send answers to a score server. The history can be cleared from this page.
General site analytics described by the privacy policy remain separate from local task history. Switching browsers or clearing site storage removes access to the local record.
These questions stay tied to the visible content on this page.
Continue with tools that pair naturally with this page.
Find one target in symbol, letter, or number boards and review completion time, errors, difficulty, and local history.
Start
Wait for a visual signal across five rounds, then review average, best response, false starts, and local history.
Start
Test visual memory by recalling highlighted grid positions, then review accuracy, level, mistakes, and browser-only session history.
Start