Upright and inverted unfamiliar face-matching tasks - everything correlates everywhere all at once

Area of Science:

• Cognitive Psychology
• Perception

Background:

• Previous research suggested unfamiliar face matching tasks may not reflect real-world face processing.
• A key study found high performance correlations between upright and inverted unfamiliar face matching, implying image-specific factors drive performance.

Purpose of the Study:

• To replicate and extend findings on unfamiliar face matching using various paradigms and judgment types.
• To investigate if these limitations apply to both identity and emotion judgments.
• To assess construct validity of unfamiliar face matching tasks for general face processing.

Main Methods:

• Replicated and extended previous findings on unfamiliar face matching.
• Examined within-subject performance for upright/inverted unfamiliar face matching.
• Utilized sequential/simultaneous presentation and sorting paradigms.
• Assessed both identity and emotion judgment types.
• Measured accuracy and reaction time.

Main Results:

• High correlations between upright and inverted conditions were consistently observed for both accuracy and reaction times.
• Factor analyses revealed that upright and inverted conditions loaded together into task-specific latent variables.
• These findings were consistent across identity and emotion matching tasks.

Conclusions:

• Unfamiliar face matching tasks, for both identity and emotion, exhibit low construct validity for measuring general face processing.
• The results support the original conclusion that such tasks are "not faces" and are driven by image-specific factors.
• Researchers should ensure alignment between their chosen face-processing tasks and the theoretical constructs they aim to measure for accurate interpretations.