Skip to content

Ambiguous Expression Confusability Sorter

Run the Ambiguous Expression Confusability Sorter MicroSim Fullscreen
Edit in the p5.js Editor

About This MicroSim

Some robot faces honestly support more than one reading, and a designer has to be able to say exactly why. This sorter loads five faces built from deliberately ambiguous parameter sets and asks you to judge which two named emotions each one could plausibly be. After you submit your pair, the feedback panel always explains which eyebrow, eye, and mouth values the two readings share. Naming that overlap out loud is the skill this chapter is really after.

How to Use

  1. Study the face on the left and the short parameter summary printed under it.
  2. Click two expression-name chips to move them into the answer zone. Clicking a chip again takes it back.
  3. Press Submit once both slots are filled. Your pair is compared with the reference pair, and the result is marked as a full match, a partial match, or a different reading.
  4. Read the justification, which appears whether or not your pair matched.
  5. Press New Ambiguous Face to load the next case. Clear Answer empties both slots before you submit.

Iframe Embed Code

You can add this MicroSim to any web page by adding this to your HTML:

1
2
3
4
<iframe src="sims/ambiguous-expression-confusability-sorter/main.html"
        height="542px"
        width="100%"
        scrolling="no"></iframe>

Lesson Plan

Grade Level

9-12 (High School)

Duration

10-15 minutes

Prerequisites

  • The thirteen named expressions and their parameter recipes from Chapter 10
  • Chapter 10's finding that recognition accuracy is uneven across emotions
  • This chapter's ideas of expression intensity and expression ambiguity
  • The valence-arousal model, which explains why some pairs sit close together

Activities

  1. Exploration (5 min): Work through all five cases. Before you place any chip, write one sentence naming the feature that makes each face unclear.
  2. Guided Practice (5 min): Run the cases again. For every reference pair, name the one parameter you would change to break the tie, and say which of the two emotions your change would push the face toward.
  3. Assessment (5 min): Pick the case you found hardest and write a short critique: the two readings, the shared features, and your recommended fix.

Assessment

  • Places a defensible pair for at least four of the five cases
  • Cites specific parameters, not general impressions, when justifying a pair
  • Explains why afraid and surprised overlap more than happy and sad do
  • Proposes a concrete parameter change that would make one case unambiguous

References

  1. Emotion classification - Wikipedia - How researchers group emotions, including the models this chapter compares.
  2. Facial expression - Wikipedia - Background on how faces signal emotion and where signals overlap.
  3. Human-robot interaction - Wikipedia - The research field that studies how people read a robot's expressions.
  4. p5.js Reference - The drawing functions used to render each ambiguous face.