E-Motions

What you see

A session is a series of sets. Each set is a run of faces, and the set is what gets scored.

1. You press Start (or Space). A 3, 2, 1 countdown runs.
2. A face appears on the screen and is held for a fixed exposure time. Then the screen blanks to a faint dashed outline.
3. Below the screen are the options, each an emotion button with its number key. They stay available after the face has gone, so you can answer during the exposure or after it. There is no separate answer deadline.
4. Choose an option (or press its number key). It is highlighted for about a quarter of a second, then the next face comes up. Each face takes exactly one answer and you cannot change it.
5. After the last face the set is recorded and the line under the board reads, for example, "10 of 10 faces answered". Press Check answer (or Enter) for the verdict.

The faces

The faces are line drawings with only three moving parts: the tilt of the brows, how open the eyes are, and the curve of the mouth. Each emotion has its own setting of those three, and everything is scaled by an intensity value. At 100% you see the full expression; at 30% the same expression is drawn at less than a third of its strength, so every face drifts towards neutral.

Two design details matter for strategy. The answer options on every face always include the true emotion and its usual confusion partner: fear with surprise, surprise with fear, anger with disgust, disgust with anger, sadness with neutral, neutral with sadness, and happiness with surprise. The rest of the options are filled at random. And the seven emotions are spread evenly across a set, so no label appears more than once more often than any other.

What changes as levels rise

Each practice session starts at level 3. A flawless set (every face correct) moves you up one level, a set with no correct answers moves you down one, and anything in between keeps you where you are.

Three things get harder at once: faces get fainter, they stay up for less time, and from level 7 every one of the seven labels is among the options.

What it measures

E-Motions is a facial emotion recognition task: the perceiving branch of ability emotional intelligence in Mayer and Salovey's model, and the entry point of Joseph and Newman's cascading model, where perceiving an emotion comes before understanding and regulating it. In our taxonomy it is the only task behind the Workplace dimension's single facet, emotion recognition.

The format descends from the forced-choice labelling studies that launched the universality debate. Ekman and Friesen asked people to match faces to short emotion stories; many later studies asked participants to choose a label for each face from a short list. Russell pointed out that this forced-choice format, with its short list of options, itself pushes agreement upwards, and Barrett and colleagues argued that posed, context-free faces overstate how readable real faces are. Our task inherits both limits deliberately: it measures how accurately and quickly you decode conventional expressions, which is exactly what commercial face tasks also measure.

Emotion recognition is a genuine ability. It correlates modestly with intelligence (r = .19 across 133 samples), and in field studies better recognisers sold more and showed more transformational leadership. Its link with job performance is clearest in jobs high in emotional labour.

The hardest discrimination, fear against surprise, has its own literature. The two share most of their facial movements, and recognition of fear suffers when the distinctive features are few. In our drawings the same is true: fear and surprise have identical eyes and almost identical brows, and differ mainly at the mouth.

How it is scored here

Each set produces an accuracy: faces named correctly divided by faces shown. The report also shows:

• Expressions read correctly, the headline accuracy across the session.
• Misreads that follow a pattern, the share of your errors that picked the face's confusion partner, plus your single commonest misread (for example "fearful read as surprised").
• Expressions left unanswered, which count as wrong.

Your 0 to 100 score blends accuracy with a smaller speed component (accuracy carries roughly four fifths of the weight, time about one fifth), and compares the result with what the difficulty you worked at predicts. Difficulty is measured from the faces you actually saw: fainter faces and shorter exposures both count as harder. On our scale, 50 means you performed as expected for that difficulty; above 50 you did better than the level predicts. Because the level adapts, a score is comparable whether you spent the session at level 3 or level 8.

Where you will meet it

Face-labelling tasks are less common in hiring than personality questionnaires, but they do appear:

• Game-based batteries. Candidates report that pymetrics (now part of Harver) includes a Faces game in which you identify the emotion shown on faces, sometimes with a caption that gives context. That caption is a twist our task does not have: you must pick an emotion that fits both the face and the situation.
• Ability EI tests. The MSCEIT's perceiving branch includes judging emotions in faces, and expert agreement on its answer key was especially high for this kind of item.
• Research tests such as the Reading the Mind in the Eyes test and the Geneva Emotion Recognition Test, which you may read about but are unlikely to meet in recruitment.

Our task is a close cousin of these formats, not a copy: it uses drawn faces rather than photographs, and it adds explicit time pressure and adaptive difficulty.

The method

1. Fix your eyes on the centre of the face before it appears

   The face is small and the exposure can be under a second. Look at the bridge of the nose during the countdown so brows, eyes and mouth all fall inside one glance. Do not chase features one by one.

2. Read the mouth: up, flat or down

   A clear upward curve is happy, and nothing else in the set curves up. A strong downturn points to sad or disgusted; a moderate one to angry; a slight one to fearful; almost none to surprised or neutral.

3. Read the eyes: wide, normal or narrow

   Widest eyes mean fearful or surprised. Narrowed eyes mean disgusted or sad. Angry sits slightly above normal.

4. Break the tie with the brows

   Steep tilt with inner ends down: angry (disgusted is the same direction, but gentler). Tilt with inner ends up: surprised, fearful and, gently, sad. Level brows: neutral.

5. Check the confusion partner, then commit

   Before pressing, name the partner that is certainly among the options and the one feature that separates them. Then press. You get no second chance on a face, so a half-second check is worth it.

Worked example 1: the classic pair

A level 4 face at about 70% intensity. The options are Surprised, Happy, Fearful, Neutral and Angry.

Show the reasoning

Fearful. The eyes are wide and the brows are raised at the inner ends, which puts the face in the fear and surprise family and rules out Happy, Neutral and Angry. The tie-breaker is the mouth: it turns down slightly. A surprised face at this intensity has an almost perfectly flat mouth (and slightly steeper brows). Fear's small downturn is the only thing separating them, so train yourself to look for it.

Worked example 2: anger or disgust

A level 5 face at about 55% intensity, held for 1.5 seconds. The options are Angry, Sad, Neutral, Disgusted, Happy and Fearful.

Show the reasoning

Disgusted. The mouth turns down clearly, so Happy, Neutral and (almost certainly) Fearful are out. The brows tilt with their inner ends down, which points to the angry and disgusted pair rather than sad. Now compare the two: an angry face has the steepest downward brow tilt in the set and eyes a little more open than normal, a glare. Here the tilt is gentle and the eyes are narrowed. That combination is disgust.

Worked example 3: the faintest faces

A level 9 face at about 35% intensity, shown for 0.7 seconds, with all seven options among the options.

Show the reasoning

Sad. Nothing is strong, which tempts you to press Neutral. But the mouth is not flat: it has a small downturn, and the eyes are a touch narrower than a neutral face's. A neutral face has perfectly level brows and a straight mouth at any intensity, so any consistent departure from flat is evidence of an emotion. At faint levels, trust small but consistent signals, and remember that Neutral is no more common in a set than any other label.

Tips and traps

Levelling up

At higher levels the task stops being about knowing the expressions and becomes about catching small differences in a short glance. Adapt in three ways:

• Shift your weight to the mouth and brows. At low intensity, eye opening changes least visibly between emotions, while mouth curve and brow tilt keep more of their difference.
• Commit to the family first. Decide up, flat or down from the mouth in the first instant; everything else is a tie-break inside that family.
• Expect seven options. From level 7 every label is present, so you cannot rule anything out by its absence. The confusion partner is always there.

1. Week 1
   • Three sessions; after each, write down your commonest misread
   • Say the scan order (mouth, eyes, brows) to yourself on every face
   • Study the seven-face chart above until you can sketch each face from memory
2. Week 2
   • Three sessions focusing on your worst confusion pair
   • Before each session, describe aloud the one feature that separates that pair
   • Aim for a flawless set to move up a level
3. Week 3
   • Three sessions; push past level 6 and note where exposure starts to hurt
   • Practise the centre-fixation habit during every countdown
   • Check that no label, especially Neutral, dominates your answers
4. Week 4
   • Two sessions at different times of day to see how fatigue affects accuracy
   • In two real conversations, silently name one emotion and the cue you used
   • Compare your first and latest session reports: accuracy, misread pattern, level reached

Check yourself

5 practice questions

1. In E-Motions, what happens when the exposure time ends?
2. If the true emotion on a face is Fearful, which other label is guaranteed to be among the options?
3. What moves you up a level during a session?

and 2 more

Create a free account to answerSign in

Sources

1. Mayer, J. D., Salovey, P., Caruso, D. R., & Sitarenios, G. (2003). Measuring emotional intelligence with the MSCEIT V2.0. Emotion, 3(1), 97–105. doi.org/10.1037/1528-3542.3.1.97
2. Joseph, D. L., & Newman, D. A. (2010). Emotional intelligence: An integrative meta-analysis and cascading model. Journal of Applied Psychology, 95(1), 54–78. doi.org/10.1037/a0017286
3. Ekman, P., & Friesen, W. V. (1971). Constants across cultures in the face and emotion. Journal of Personality and Social Psychology, 17(2), 124–129. doi.org/10.1037/h0030377
4. Russell, J. A. (1994). Is there universal recognition of emotion from facial expression? A review of the cross-cultural studies. Psychological Bulletin, 115(1), 102–141. doi.org/10.1037/0033-2909.115.1.102
5. Barrett, L. F., Adolphs, R., Marsella, S., Martinez, A. M., & Pollak, S. D. (2019). Emotional expressions reconsidered. Psychological Science in the Public Interest, 20(1), 1–68. doi.org/10.1177/1529100619832930
6. Schlegel, K., Palese, T., Schmid Mast, M., Rammsayer, T. H., Hall, J. A., & Murphy, N. A. (2020). A meta-analysis of the relationship between emotion recognition ability and intelligence. Cognition and Emotion, 34(2), 329–351. doi.org/10.1080/02699931.2019.1632801
7. Byron, K., Terranova, S., & Nowicki, S. (2007). Nonverbal emotion recognition and salespersons. Journal of Applied Social Psychology, 37(11), 2600–2619. doi.org/10.1111/j.1559-1816.2007.00272.x
8. Rubin, R. S., Munz, D. C., & Bommer, W. H. (2005). Leading from within. Academy of Management Journal, 48(5), 845–858. doi.org/10.5465/amj.2005.18803926
9. Roy-Charland, A., Perron, M., Beaudry, O., & Eady, K. (2014). Confusion of fear and surprise: A test of the perceptual-attentional limitation hypothesis with eye movement monitoring. Cognition and Emotion, 28(7), 1214–1222. doi.org/10.1080/02699931.2013.878687
10. GraduatesFirst. Harver/pymetrics assessments and practice games (candidate-reported). graduatesfirst.com
11. Baron-Cohen, S., et al. (2001). The "Reading the Mind in the Eyes" test revised version. Journal of Child Psychology and Psychiatry, 42(2), 241–251. doi.org/10.1111/1469-7610.00715
12. Schlegel, K., Grandjean, D., & Scherer, K. R. (2014). Introducing the Geneva Emotion Recognition Test. Psychological Assessment, 26(2), 666–672. doi.org/10.1037/a0035246

Put it into practice: E-MotionsSign in to try it