Open Eyeballing Jewell and McCourt, 2000

By eye.

Six designer tasks in six rounds: the middle of a line, the centre of a circle, a right angle, the missing corner of a square, a parallel and the centroid of a triangle. See how far your eye lands from the ruler, and whether you shift the middle of a line to the left or the right.

3 minto take part 1responses Anonymousno sign-up needed

EXP. 047

Without a ruler, set square or compass, how accurately can you find the middle of a line, the centre of a circle or an exact right angle?

Loading the experiment… It needs JavaScript to run; you can still read the science box and the article below.

Play again, climb higher

Leaderboard.

No one is on the leaderboard yet. Be the first.

Only volunteers appear on the leaderboard. Your anonymous plays are linked to your account when you volunteer.Volunteer eyeballing · each person’s best play counts · “Today” resets every midnight (Istanbul time)

Take part first.

Reading the science could sway your answer. Everything unlocks once you take part, but you can read it now if you prefer.

Back to the experiment

Science box

The eye is a measuring instrument that comes before all other measuring instruments: designers, craftspeople and painters trust their eyes first when they centre or align something. But the eye has its own biases; the best known is that most healthy people put the middle of a line slightly to the left.

What we measure

We measure how accurate your eye is in six geometry tasks. All tasks are drawn in a square drawing area on a paper-coloured background, so the geometry is the same on every screen. In round 1, four horizontal lines come one at a time: their lengths are 55 to 82 per cent of the area's width and their midpoints are shifted slightly from the centre of the screen, so the screen's centre gives no clue. The orange dot starts on the left of the line (at 20 per cent) for two lines and on the right (at 80 per cent) for two, so the starting side does not push your left/right bias one way. The other rounds have one task each: the centre of a circle, a right angle to a given arm, the fourth corner of a square with three given corners, a parallel to a tilted line through a given point, and the centroid of a triangle. In the right-angle and parallel tasks the given lines are deliberately tilted (at least 15 degrees from the nearest horizontal or vertical), because horizontal and vertical directions would make the task easier. With a mouse you grab and move the dot directly; on a touchscreen, sliding your finger anywhere moves the dot by the same amount, and a short tap puts the dot there; the arrow keys move it by one pixel (ten with Shift). In the position tasks the error is measured relative to the size of the shape: line length for the midpoint, radius for the centre, side length for the square, and the vertices' average distance from the centroid for the centroid. In the angle tasks the error is in degrees. Each task scores 10 / (1 + (error / k)^1.6); k is 2.5 per cent for the midpoint, 5 per cent for the centre and the square, 6 per cent for the centroid and 2.5 degrees for the angles. Round 1's score is the average over the four lines; the six rounds add up to at most 60 points. From your four midpoint answers we also calculate your personal left/right bias.

What the research says

The tendency of healthy people to put the middle of a line slightly to the left is called pseudoneglect, by analogy with neglect, in which half of space is ignored after brain damage; Bowers and Heilman first described it in 1980, in a task in which participants tried to find the middle of a thin wooden rod by touch. In 2000 Jewell and McCourt reviewed the studies in this field and combined data from 2,191 people in 73 studies in a meta-analysis: healthy people showed a significant leftward bias, with an effect size between −0.37 and −0.44 depending on the method. The effect was sensitive to many factors: scanning from left to right enlarged the leftward bias, scanning from right to left produced rightward errors, and older participants erred further to the right than younger ones. Learmonth and colleagues (2015) found that manual line bisection gives consistent results when measured on two separate days. For angles, Gray and Regan (1996) showed that when people set the angle of a 'V', they set a 90-degree angle with less variability than other angles. For position, Whitaker and colleagues (1996) found that the perceived location of blurred blobs is well explained by their centroid.

Why it happens

The most common explanation of the leftward bias is that spatial attention is processed mainly in the right half of the brain: when the right hemisphere gives a little more attention to the left visual field, the left half of the line looks a little bigger and the midpoint shifts left. This explanation fits real neglect, in which the left half of space is ignored after right-hemisphere damage; pseudoneglect is a small reflection of the same balance in a healthy brain. But the bias is small and varies from person to person: some people err to the right. Learmonth and colleagues (2015) found that different pseudoneglect tasks did not correlate strongly with each other; so a 'leftward bias' may be made of several separate components rather than one trait. It also changes with age: Friedrich and colleagues (2018) reviewed inconsistent findings above the age of 60, with the leftward bias growing in some studies and shrinking or turning rightward in others. In the other tasks, errors mostly come from the geometry itself: we read the direction of tilted lines less sharply than horizontal and vertical ones (the oblique effect; Appelle, 1972), and we have no single intuition for the 'middle' of a triangle.

Limitations

Four midpoint measurements are too few to show a personal left/right bias conclusively; the bias is small and a single careless answer can change the result. We are measuring a tendency here, not making a diagnosis. Mouse and touchscreen require different movements; on a touchscreen the dot is drawn slightly below so that it is not hidden under your finger. The edges of the screen offer a horizontal and vertical reference; we recommend holding your phone upright and level. The starting side of the dot (left for two lines, right for two) is balanced, but the direction your hand or finger comes from may also affect the result; Jewell and McCourt (2000) found that the hand used and the scanning direction change the bias. The general picture will come from the crowd's data.

effect size −0.37 to −0.44Leftward midpoint bias in healthy people (73 studies, 2,191 people)Jewell and McCourt, 2000

enlarges the leftward biasScanning from left to rightJewell and McCourt, 2000

Bowers and Heilman, 1980First description (bisecting a rod by touch)Bowers and Heilman, 1980

90°Angle set with the least variabilityGray and Regan, 1996

centroidPerceived location of blurred blobsWhitaker et al., 1996

Jewell, G., & McCourt, M. E. (2000). Pseudoneglect: A review and meta-analysis of performance factors in line bisection tasks. Neuropsychologia, 38(1), 93–110. View source ↗
Bowers, D., & Heilman, K. M. (1980). Pseudoneglect: Effects of hemispace on a tactile line bisection task. Neuropsychologia, 18(4–5), 491–498. View source ↗
Learmonth, G., Gallagher, A., Gibson, J., Thut, G., & Harvey, M. (2015). Intra- and inter-task reliability of spatial attention measures in pseudoneglect. PLoS ONE, 10(9), e0138379. View source ↗
Friedrich, T. E., Hunter, P. V., & Elias, L. J. (2018). The trajectory of pseudoneglect in adults: A systematic review. Neuropsychology Review, 28(4), 436–452. View source ↗
Gray, R., & Regan, D. (1996). Accuracy of reproducing angles: Is a right angle special? Perception, 25(5), 531–542. View source ↗
Whitaker, D., McGraw, P. V., Pacey, I., & Barrett, B. T. (1996). Centroid analysis predicts visual localization of first- and second-order stimuli. Vision Research, 36(18), 2957–2970. View source ↗
Appelle, S. (1972). Perception and discrimination as a function of stimulus orientation: The "oblique effect" in man and animals. Psychological Bulletin, 78(4), 266–278. View source ↗

The eye's ruler

When we centre a headline on a poster, hang a frame on a wall or align two strokes in a logo, most of us trust our eyes first. This skill, which designers call 'eyeballing', is surprisingly good, but not perfect, and its imperfections are not random.

There are six classic geometry tasks in this experiment: the middle of a line, the centre of a circle, a right angle, the missing corner of a square, a parallel and the centroid of a triangle. In each, we measure your error relative to the size of the shape; a 5-pixel error on a large circle counts as less than 5 pixels on a small one.

Left of centre

Asked to mark the middle of a line, most healthy people put the mark slightly to the left of the true middle. This small bias is called pseudoneglect, by analogy with neglect, the condition seen after brain damage. Bowers and Heilman first described it in 1980, in a study in which participants tried to find the middle of a thin wooden rod by touch.

In 2000 Jewell and McCourt reviewed the studies on the subject and combined them in a meta-analysis: in data from 2,191 people in 73 studies there was a significant leftward bias, with an effect size between −0.37 and −0.44. The effect was sensitive to many details; scanning from left to right enlarged the leftward bias, and scanning from right to left reversed it. That is why in this experiment we balanced the starting side of the dot: it starts from the left on two lines and from the right on two.

The right hemisphere's attention

The most common explanation of the leftward bias is that spatial attention is processed mainly in the right half of the brain. When the right hemisphere gives slightly more attention to the left visual field, the left half of the line looks slightly bigger, and when you try to find the middle, the mark shifts left. After right-hemisphere damage, the left half of space can be almost entirely ignored; that is the severe form of the same imbalance.

But pseudoneglect is not a single, fixed trait. In 2015 Learmonth and colleagues found that manual line bisection gave consistent results on two separate days, but that different pseudoneglect tasks did not correlate strongly with each other. In their 2018 review, Friedrich and colleagues showed that findings above the age of 60 are inconsistent: in some studies the leftward bias grows, in others it shrinks or turns rightward.

Is a right angle special?

Right angles are everywhere in the man-made world: doors, pages, screens, bricks. In 1996 Gray and Regan asked participants to set the angle of a 'V' to 90, 45 and three other angles. To stop the direction of the arms from being a reliable cue, they turned the arms in different directions. Even with feedback, the 90-degree angle was set with less variability than the other angles.

In this experiment we deliberately tilt the given lines in the right-angle and parallel tasks. With a horizontal or vertical arm, the edges of the screen and the eye's superior sharpness for horizontal and vertical (the oblique effect; Appelle, 1972) would make the task easier. Building a right angle on a tilted arm requires the eye to really construct the angle.

Centre, square, centroid

In the position tasks the eye has to summarise a whole shape. In 1996 Whitaker and colleagues measured the perceived location of blurred, asymmetric blobs and found that location was well explained by the blob's centroid: the visual system seems to read a shape's position with a kind of centre-of-mass computation. That makes finding the centre of a circle relatively easy.

The centroid of a triangle is harder. It is where the medians, which join each vertex to the middle of the opposite side, cross, and it divides each median in a 2-to-1 ratio from the vertex. In an equilateral triangle the centre of the inscribed circle, the centre of the circumscribed circle and the centroid are the same point; as the triangle gets more pointed, these points separate and the intuition of its 'middle' can mislead. In the square task, too, the missing corner is where lines from the ends of the two sides, parallel to the other sides, cross.

Your score and your screen

Each task's score is calculated from the error measured relative to the size of the shape (or in degrees): 10 / (1 + (error / k)^1.6). An error of 2.5 per cent of the line length at the midpoint, 5 per cent for the centre and the square, 6 per cent for the centroid and 2.5 degrees for the angles earns 5 points. Round 1's score is the average of the four lines; the six rounds add up to at most 60 points.

All tasks are drawn in a square area; the geometry is the same on a phone and a computer. On a touchscreen you only need to slide your finger anywhere so that the dot isn't hidden under it; the dot moves by the same amount. At the end we calculate your left/right bias from your four midpoint answers: the share of the crowd that shifts left will be a small dataset that can be compared with Jewell and McCourt's meta-analysis.

FAQ

What is pseudoneglect?

It is the tendency of healthy people to put the middle of a line slightly to the left. It is named after neglect, in which half of space is ignored after brain damage; but it is very small and not a sign of illness. Jewell and McCourt's (2000) meta-analysis found a significant leftward bias across 73 studies.

How is my score calculated?

In each task the error is measured relative to the size of the shape (line length for the midpoint, radius for the centre, side length for the square, the vertices' average distance from the centroid for the centroid) or in degrees. The score is 10 / (1 + (error / k)^1.6); k is 2.5% for the midpoint, 5% for the centre and the square, 6% for the centroid and 2.5° for the angles. The total for six rounds is at most 60.

Is something wrong if I shifted to the right?

No. The leftward bias is a small tendency seen in the crowd's average; many healthy people shift to the right. Four measurements are not enough to show a personal tendency conclusively. This experiment is not a diagnostic tool.

How do I move the dot?

With a mouse, grab the dot and drag it, or click where you want it. On a touchscreen, slide your finger anywhere on the screen; the dot moves by the same amount as your finger, so your finger doesn't cover it. A short tap puts the dot where you tapped. The arrow keys move it by one pixel, or ten with Shift.

Why are the lines always tilted in the right-angle and parallel tasks?

Horizontal and vertical lines make the task easier: the edges of the screen offer a ready reference, and the eye sees these directions more sharply than oblique ones (the oblique effect). With a tilted line we can measure whether the eye really constructs the angle.

What are the daily series and the challenge?

In free play you get new shapes every time. In the daily series everyone sees the same shapes in the same order that day; the day changes according to Istanbul time. In a challenge you play exactly your friend's shapes via their link; in a race room 2–6 people play the same shapes at the same time. Only your first game is kept in the scientific data.

Discussion 0 comments

Join the discussionReading is open to everyone. Volunteer to comment and vote; it takes 20 seconds.

No comments yet.Be the first to write; a good question gets the discussion going.

Threads about this experiment

Start a new thread →

There’s no separate thread for this experiment yet. Start one to critique the method, share a paper or ask a new question.

Up next.

Next · EXP. 048

Number line

You see a number and a line: 0 at one end, 100 or 1000 at the other. You press where the number belongs and let go. After twenty numbers your estimates are plotted: do they look more like a straight line, or like a logarithmic curve that inflates the small numbers?

2 min · 0 responsesTake part

Previous · EXP. 046

The square in the shadow

In six scenes you match the grey of a dotted square by adjusting an empty square on a neutral background: in a shadow, on dark and light backgrounds, between black and white stripes. At the end we remove the context: you see with your own eyes that two squares that look different are exactly the same grey.

3 min · 0 responsesTake part

Tip: use the ← → keys, or swipe the header left and right on your phone, to move between experiments. All experiments Run a session with this experiment