Related Experiment Video
Updated: Jun 21, 2026
07:31
Investigating the Effect of Visual Imagery and Learning Shape-Audio Regularities on Bouba and Kiki
Published on: September 13, 2019
10.1K
Better than Goodenough? Evaluating new computational techniques for finding diagnostic structure in children's
Clint A Jensen1, Timothy T Rogers2, Karl S Rosengren3
1Department of Psychology, University of Wisconsin-Madison, Madison, WI, USA. cjensen5@wisc.edu.
Memory & Cognition
|April 26, 2024
Summary
Computational analysis of historical drawings reveals rich cognitive insights. Modern methods extract age, gender, and mental ability data from 100-year-old human-figure drawings, offering new perspectives on cognitive evaluation.
Area of Science:
- Cognitive Science
- Developmental Psychology
- Computational Intelligence
Background:
- Florence Goodenough's 1926 work pioneered quantitative analysis of children's drawings for cognitive evaluation.
- Checklist-based scoring methods, derived from Goodenough's work, remain widely used in clinical practice.
- Recent advancements in computational cognitive science suggest drawings contain more structural information than previously captured.
Purpose of the Study:
- To apply contemporary computational tools to analyze the structure of drawings from Goodenough's original study.
- To assess if this extracted structure correlates with demographic and cognitive characteristics of the participants.
- To re-evaluate the relationship between human-figure drawings and mental ability.
Main Methods:
- Utilized computational innovations from cognitive science to analyze image structure.
- Applied methods to characterize structural elements in drawings from Goodenough's 1926 study.
- Assessed the correlation between drawing structure and participant demographics (age, gender) and cognitive abilities.
Main Results:
- Contemporary computational methods reliably extracted information on participant age, gender, and mental faculties from historical drawings.
- Extraction of data was achieved with minimal human effort and without requiring expert training.
- The study identified a different relationship between drawing characteristics and mental ability compared to Goodenough's original approach.
Conclusions:
- Human-figure drawings contain significant, extractable information about cognitive and demographic characteristics, even from historical data.
- Computational analysis offers a powerful, objective method for cognitive evaluation using drawings, surpassing traditional checklist approaches.
- These findings have significant implications for the future use of drawing-based assessments in cognitive evaluation.
