pymetrics Arrow Game: Rules, Timing and What It Measures
Luca from Candidate Falcon
Editorial Team
The pymetrics arrow game, called Arrows in the battery, shows five arrows in a row and asks you to press left or right. The catch is colour: it tells you which arrows to follow, and it can change on any set. Arrows is one of the 12 core pymetrics games, which Harver now runs as its gamified assessments after buying pymetrics in August 2022. Prep sites usually report it as the ninth game.
Facts below are labelled official (pymetrics or Harver material, including pymetrics' patent), prep sites (commercial descriptions) or not public.
The Arrows game at a glance
| What is known | Source | |
|---|---|---|
| Task | Five arrows in a row; a colour cue decides which ones count | Official (pymetrics patent) |
| Rule in the live game | Blue or black arrows: answer the middle arrow. Red arrows: answer the side arrows | Prep sites |
| Controls | Left and right arrow keys | Prep sites |
| Length | About 3 minutes; 135 sets on most prep sites, 90 on one | Prep sites |
| Time per set, feedback, scoring | Not published | Not public |
What you see and do
- Instructions. They explain the colour rule. Read them carefully: the patent and prep sites describe different rules, and your screen is what counts.
- Five arrows appear (official). Sometimes all five point the same way (congruent); sometimes the middle one points against the rest (incongruent). One prep site also reports neutral sets.
- Check the colour (prep sites). Blue or black: press the direction of the middle arrow. Red: press the direction of the four side arrows.
- Press left or right (prep sites). The next set then appears.
- The rule switches without warning (prep sites). Colours change unpredictably, so you cannot plan for the next set.
pymetrics' patent uses a different colour mapping: a red centre among red arrows means follow the centre, and a red centre among blue arrows means follow the blue side arrows.
The psychology behind it: the flanker task and task switching
Arrows combines two classic paradigms.
The Eriksen flanker task. Eriksen and Eriksen (1974) asked people to identify a target letter flanked by "noise" letters. When the flankers called for the opposite response, people were slower, even though they knew to ignore them. Later versions swapped letters for arrows; the Attention Network Test (Fan and colleagues, 2002) is a well-known example. Blue sets are a straight flanker task: ignore the sides, follow the middle.
Task switching. Red sets flip the rule: the side arrows become the target and the middle one the distraction. Research on task switching, such as Rogers and Monsell (1995), shows that people are slower and less accurate just after the task changes, even when they know a change is coming. That extra time is called the switch cost. Because the colour changes without warning, switch costs are built into the game.
What it is thought to measure
Prep sites and the patent point to selective attention (ignoring arrows that point the wrong way), cognitive flexibility (adjusting when the rule changes) and learning. Attention and focus are among the nine trait categories Harver lists. pymetrics does not publish which measures it takes, or how it weighs speed against accuracy.
Overall, pymetrics compares your play across all the games with people who do well in the role and gives the employer a fit score, not a pass mark (official). See pymetrics results and the trait report.
Is there a right answer?
Yes, for each set. Unlike the money games, every Arrows set has one correct key, so accuracy and reaction time can be measured. pymetrics still says there is "no right or wrong answer" across its games, because the result is a profile matched to roles rather than a mark: the same speed and accuracy could fit one role better than another. More in pymetrics games answers.
Timing and length
- The whole set (official): about 25 minutes for the 12 games, 1 to 3 minutes each, with no pausing mid-game.
- Arrows (prep sites): about 3 minutes. 135 sets in 3 minutes is roughly 1.3 seconds per set, gaps included (our arithmetic). Whether each set has its own time limit is not public.
How to approach it
pymetrics' advice is to be yourself. In Arrows, that means a pace you can sustain with full attention.
- Learn the rule before you start. Say it once in your head: blue, middle; red, sides.
- Colour first, arrows second. The colour tells you where to look.
- Use a keyboard. Prep sites advise a computer; rest two fingers on the arrow keys.
- Let mistakes go. A wrong key on one set says nothing about the next one.
- Ask about colour adjustments. pymetrics' research paper says the games include accommodations for colour blindness. If red and blue are hard for you to tell apart, ask the employer before you start.
Common mistakes
- Following the middle arrow out of habit on a red set.
- Remembering the rule backwards: "red means middle".
- Answering on the first arrow you notice rather than the one the colour points to.
- Trying to predict when the colour will change.
- Playing on a phone when a computer is available.
A worked example
Here are five sets we made up, using the rule prep sites report. Arrows are listed left to right.
| Set | Colour | Arrows | Rule | Answer |
|---|---|---|---|---|
| 1 | Blue | → → ← → → | Middle | Left |
| 2 | Blue | ← ← ← ← ← | Middle | Left |
| 3 | Red | ← ← → ← ← | Sides | Left |
| 4 | Red | → → → → → | Sides | Right |
| 5 | Blue | ← ← → ← ← | Middle | Right |
Sets 1, 3 and 5 are incongruent: the middle points against the sides. Sets 3 and 5 also come straight after a colour change, so research would predict the slowest answers there. Set 3 is the classic trap: the middle arrow points right, but red says follow the sides.
How Candidate Falcon's Arrows practice works
Candidate Falcon's independent pymetrics Arrows practice game follows the rule prep sites report, with original screens. It is not the real game.
- 90 sets per run, half blue and half red, half congruent and half incongruent, shuffled so colour changes come without warning.
- Up to 2 seconds per set, answered with the arrow keys or on-screen buttons on a phone.
- No feedback during the run, then your accuracy and median reaction time by rule, by arrow set and after a colour change (your switch cost).
The 90 sets and the 2-second window are our choices. Practice results are not pymetrics or Harver scores. The pymetrics games come with every plan that includes Harver; see pricing. The free pymetrics guide covers the whole battery.
FAQs
What are the rules of the pymetrics arrow game?
Prep sites report: with blue or black arrows, press the direction of the middle arrow; with red arrows, press the direction of the side arrows. pymetrics' patent describes a different colour rule, so follow your on-screen instructions.
How many rounds are in the pymetrics Arrows game?
pymetrics does not publish a count. Most prep sites report 135 sets in about 3 minutes; one reports 90.
What is a switch cost?
It is the extra time, or the extra errors, on the first set after the rule changes, compared with a set where the rule stayed the same.
Is it better to be fast or accurate?
pymetrics does not say how it weighs the two. Accuracy at a pace you can keep up is a sensible aim, since a fast wrong answer adds nothing.
Sources
- pymetrics patent WO2015200880A1: five arrows and the patent's colour rule.
- Wilson et al. (2021), Building and Auditing Fair Algorithms, with pymetrics co-authors: devices, accommodations, the fit model.
- pymetrics candidate FAQ, employer-hosted copy (2018): timing, no pausing, no right or wrong answer.
- Harver: gamified assessments and Harver acquires pymetrics (read October 2026).
- Prep-site descriptions (GraduatesFirst, JobTestPrep, MBB Prep, Psychometric Success): the live colour rule, controls, set counts.
- Eriksen, B. A., and Eriksen, C. W. (1974). Effects of noise letters upon the identification of a target letter in a nonsearch task. Perception & Psychophysics, 16(1), 143–149.
- Fan, J., et al. (2002). Testing the efficiency and independence of attentional networks. Journal of Cognitive Neuroscience, 14(3), 340–347.
- Rogers, R. D., and Monsell, S. (1995). Costs of a predictable switch between simple cognitive tasks. Journal of Experimental Psychology: General, 124(2), 207–231.
Candidate Falcon is independent and not affiliated with pymetrics, Harver or any employer.
