[DSRP Evidence](https://dsrpevidence.org/)

# Becoming a “Greeble” Expert: Exploring Mechanisms for Face Recognition

## Details

**Authors** Gauthier and Tarr

**Year** 1997

**Publisher** Vision Research

**Discipline** Cognitive Science

[Read it at the publisher](https://doi.org/10.1016/s0042-6989(96)00286-6) 
10.1016/s0042-6989(96)00286-6

## In authors' words

### Abstract

Sensitivity to configural changes in face processing has been cited as evidence for face-exclusive mechanisms. Alternatively, general mechanisms could be fine-tuned by experience with homogeneous stimuli. We tested sensitivity to configural transformations for novices and experts with nonface stimuli ("Greebles"). Parts of transformed Greebles were identified via forced-choice recognition. Regardless of expertise level, the recognition of parts in the Studied configuration was better than in isolation, suggesting an object advantage. For experts, recognizing Greeble parts in a Transformed configuration was slower than in the Studied configuration, but only at upright. Thus, expertise with visually similar objects, not faces per se, may produce configural sensitivity.

### What they set out to do (purpose)

How does facial recognition work in the brain?

### What they found (results)

The "expert" group (the ones with longer and more productive exposure to the distinctions) had a faster response time and slightly more accurate results on average.

## Commentary

### In short

Purposefully "training" yourself to make distinctions increases the speed at which you can distinguish identities and the accuracy with which you make those distinctions.

### In more detail

Gauthier and Tarr 1997 worked with computer generated things called “greebles.” These were created in order to test different aspects of facial recognition processes in humans. In their experiment, they worked with 32 Yale University undergraduates. They generated 60 of the “greebles”, which are organized into five families and two genders based on the physical configuration of the generated greeble.

Each individual is unique, even though some might look as if they are quite similar. 30 of the generated greebles were used in the expertise training, while 24 of the unused ones were used for both the novice-level and expertise-level test phases. They designated “nonsense words” to call the five families,

two genders, and each individual greeble. 16 of the participants were “experts” and 16 were “novices.” The novice test group were only given the names of the 3 greeble parts: boges, quiff, and dunth.

No extra training was given to the novice group. They then were shown 6 novel greebles with a quick flash of their names (1 second) for 36 trials. Afterwards, the participants were given a forced-choice recognition of the parts. This was done in the form of a greeble’s name and part was shown on the screen (e.g. Pimo’s Boges) and the participants were asked to identify the part. There were three conditions for this test:

1. “Studied-configuration: the two choices were the specified part and a foil part, both in the context of the Greeble specified in the prompt; 2. Transformed-configuration: the two choices were of the specified part and a foil part, both in the context of the Greeble specified in the prompt but with the top parts moved 15 deg towards the front; 3. Isolated-part: the two choices were of the specified part and a foil part, both in isolation on the screen.”

The same procedure was done with inverted greebles afterwards. If the participants were in the “expert” test group, then they were given extensive training to make them experts at greeble recognition. This was done by having them practice greeble recognition at the family, gender, and individual levels with 30 different greebles. They then had to do 60 trials in which they verified their experts status by labelling novel greebles. The participants did 360 randomized trials where their response time and accuracy was collected. On average, those in the expert category responded more accurately and faster than those in the novice groups. Their results indicate that the more exposure one has to fine-tuned and novel distinctions, the better they will be at recognizing them. This is significant for the universal patterns of thought as it implies that the more aware that one is of the Distinctions around them, the more accurate and quick they will be at recognizing and thinking with them.

**Patterns it shows** D

**How to cite this** Gauthier and Tarr (1997). Becoming a “Greeble” Expert: Exploring Mechanisms for Face Recognition. Vision Research.