A congruent flanker trial is a row in which every arrow in the display points the same way; an incongruent flanker trial is a row in which the four surrounding arrows point opposite to the outlined centre arrow. In both cases, only the centre arrow is the answer, so the difference between the two trial types lives entirely in the flankers, not in the rule you must follow. The browser game Flanker Test packages that comparison into a single visible-tab route: twelve product-authored rows, six matching and six conflicting, with the answer in every row decided by the centre arrow alone. Because the route is fixed and the two halves are balanced, the final summary can describe what happened on matching rows, what happened on conflicting rows, and how much slower (or faster) the conflicting half was — all from one round of play kept locally in your browser. That structure is what lets the page call its main score an interference index rather than a generic reaction-time average.
What Matching and Conflicting Rows Look Like in the Game
Each row shows five arrows, arranged horizontally, with the middle arrow outlined to mark it as the answer. In a matching row, the centre arrow and the four flankers all point the same direction — for example, every arrow points right, or every arrow points left. In a conflicting row, the four flankers point one way and the outlined centre arrow points the other way — for example, the flankers point left while the centre arrow points right.
The task in both row types is identical: choose only the direction of the outlined centre arrow, using either the two large labelled buttons or the Left and Right Arrow keys. The flankers are visual context for the row, never a second rule that you must infer. The page itself labels the two halves "matching" and "conflicting", and the result panel uses those same labels so the comparison stays readable from the first row to the last.
How the Twelve-Row Route Is Built
The fixed route contains twelve product-authored five-arrow rows: six matching and six conflicting. That 6+6 split is the page's way of balancing the two halves so each average rests on the same number of responses. The rows are not randomised row by row, and they are not pulled from a pool — every run presents the same twelve rows in the same authored order.
This matters for two reasons. First, accuracy and timing describe what happened on that specific route in that specific session, not a re-sampled estimate from a larger deck. Second, because the matching and conflicting halves are equal in size, the two averages and the interference index below them can be quoted side by side without a weighting step or a hidden adjustment. The route is described in the product contract as fixed at twelve rows, with the two groups always present in equal counts.
| Feature | Matching rows | Conflicting rows |
|---|---|---|
| Number of rows in the route | 6 | 6 |
| Direction of the four flankers vs centre arrow | Same direction as the centre arrow | Opposite direction to the centre arrow |
| Which arrow is the answer | The outlined centre arrow | The outlined centre arrow |
| Used in which result-panel figure | Matching-row average response time | Conflicting-row average response time |
| Role in the interference index | Subtracted from the conflicting average | The starting value, with the matching average subtracted from it |
How to Play the Twelve-Row Route
The route runs once per start, with one row visible at a time and one response accepted per row.
- Start the fixed twelve-row route while the tab is visible, and keep the tab in front of you for the duration of the run.
- For each five-arrow row that appears, identify the outlined centre arrow and treat the four surrounding arrows as visual context rather than a separate rule.
- Choose Left or Right using the two large labelled buttons on the page, or use the Left and Right Arrow keys as an optional shortcut.
- Wait for the row to advance — the page accepts one response per visible row, so a repeated key event or an extra click cannot add a second answer to the same row.
- Continue until the twelfth response is registered, then read the result panel that appears in place of the row.
- Use Restart to return to the beginning of the same fixed route if you want another session on the identical deck.
Reading the Result Panel
After the twelfth row, the panel that replaces the play area reports four figures: accuracy as a percentage, the average response time on matching rows, the average response time on conflicting rows, and an interference index.
The interference index is defined plainly as the conflicting-row average minus the matching-row average, both taken from this single fixed route. If the conflicting half is slower, the index is positive; if the conflicting half happens to be faster in a particular session, the index can be zero or negative. A positive or negative number is not compared with a population benchmark, a published reference, an age range, a classroom, a workplace, or a clinical sample. Accuracy is kept visible on the same panel so a faster route with more incorrect selections is not automatically presented as better.
Worked example. Suppose a session finishes with these figures from the result panel:
- Accuracy: 92% (eleven correct out of twelve rows)
- Matching-row average response time: 520 ms
- Conflicting-row average response time: 610 ms
The interference index is then computed as:
Interference index = conflicting average − matching average Interference index = 610 ms − 520 ms Interference index = 90 ms
That 90 ms is the gap between the two halves of this route, not a ranking. The same session can show a smaller or even negative gap on a different device, in a different browser, after a different night's sleep, or with the tab distracted by notifications — the value describes this round of play only.
Why a Hidden Tab Pauses the Row
Browser timing in the game has one practical safeguard. If the page becomes hidden — because you switched to another app, opened a notification, or took a call — the current row pauses instead of registering a hidden-tab response. When you come back to the tab, a Resume this row control starts a fresh visible-tab interval for that same row, so the time you spent away is not folded into the next displayed response.
This safeguard is not a clinical-grade timing protocol, and it does not turn the page into a precise instrument. It only keeps the simple local summary from being polluted by time the row was not actually on screen. Restart returns to the first row of the same fixed route if you want a clean reset on a new session.
What These Numbers Don't Tell You
The route is product-authored and entertainment-focused. Screen size, input method, browser scheduling, interruptions, familiarity with the route, and ordinary day-to-day variation can all change a session's numbers, so the values should be read as transparent feedback from one round of play rather than a claim about attention, intelligence, learning, or working memory. There is no IQ score, no diagnostic cutoff, no norm group, and no classroom or workplace comparison. If you need support for health, learning, work, or accessibility, this is not a substitute for a qualified assessment.
The same shape of comparison — fixed route, balanced halves, within-session summary — is used in other transparent browser games, such as the Stroop Test, which lays out colour-word rows in a fixed order and reports local-only colour and word summaries, and the Go / No-Go Test, which reports clear hit, miss, false-alarm, and correct-rejection counts from a single route. Each of those games describes what happened on its own rows; none of them positions the figure against another player.
For readers who also want to see the same fixed centre-arrow task framed as a target-versus-distractor comparison, the Arrow vs Letter Flanker Task guide walks through the centre-arrow variant next to a letter variant of the same idea.
For a deeper look, see Emotional vs Neutral Go No Go: What Changes and How to Play.