[Paper Review] God(s) Know(s): Developmental and Cross-Cultural Patterns in Children Drawings

Ksenia Konyushkova, Nikolaos Arvanitopoulos+3 more|arXiv (Cornell University)|Nov 11, 2015

Animal and Plant Science Education27 references3 citations

TL;DR

This study introduces a computer vision-based framework to objectively analyze developmental and cross-cultural patterns in children’s drawings of God(s). Using image metrics like gravity center, color intensity, palette diversity, and visual complexity, it reveals that children across cultures tend to draw God above the page center, favor yellow and gray, and increase drawing complexity until around age 13 before simplifying again.

ABSTRACT

This paper introduces a novel approach to data analysis designed for the needs of specialists in psychology of religion. We detect developmental and cross-cultural patterns in children's drawings of God(s) and other supernatural agents. We develop methods to objectively evaluate our empirical observations of the drawings with respect to: (1) the gravity center, (2) the average intensities of the colors \emph{green} and \emph{yellow}, (3) the use of different colors (palette) and (4) the visual complexity of the drawings. We find statistically significant differences across ages and countries in the gravity centers and in the average intensities of colors. These findings support the hypotheses of the experts and raise new questions for further investigation.

Motivation & Objective

• To develop automated, culturally unbiased methods for analyzing children’s drawings of supernatural agents, addressing limitations of manual, expert-based coding in psychology of religion.
• To detect developmental trends in children’s drawings—particularly shifts in composition, color use, and complexity—across different age groups.
• To identify cross-cultural differences in drawing patterns, such as color preferences and compositional strategies, across diverse religious and cultural contexts.
• To provide objective, quantitative tools that support hypothesis testing and generate new research questions in cognitive developmental psychology and religious cognition.
• To enable large-scale, reproducible analysis of growing datasets of children’s drawings, overcoming the infeasibility of manual evaluation at scale.

Proposed method

• Applied computer vision techniques to analyze bitmapped scans of children’s drawings, focusing on four key metrics: gravity center, average color intensities (green and yellow), palette diversity, and visual complexity.
• Used Harris corner detection to quantify visual complexity by counting salient features in each drawing, serving as a proxy for detail and intricacy.
• Calculated palette variability by measuring the average variance in RGB channels of dominant colors, indicating richness and diversity of color usage.
• Employed image processing pipelines to extract and normalize features across diverse drawing styles and cultural contexts, ensuring cross-cultural comparability.
• Grouped children by age (1–7, 7–11, >11) to test developmental trends, aligning with Piagetian developmental stages.
• Validated findings through statistical analysis of differences in feature distributions across age groups and nationalities, using non-parametric tests to assess significance.

Experimental results

Research questions

• RQ1 How does the gravity center of children’s drawings of God(s) vary with age and across different cultural contexts?
• RQ2 To what extent do color preferences—particularly yellow and green—reflect developmental or cultural influences in children’s representations of supernatural agents?
• RQ3 How does the visual complexity of children’s drawings, measured via corner detection and palette variance, change with age, and does this follow a non-linear developmental trajectory?
• RQ4 Are there statistically significant differences in drawing features (color palette, composition, complexity) between children from different countries and religious backgrounds?
• RQ5 Can automated, culture-independent image analysis methods reliably detect developmental and cross-cultural patterns in children’s drawings of God(s) without expert bias?

Key findings

• Children consistently draw God(s) above the center of the page, with a statistically significant gravity center shift toward the upper half, suggesting a cultural or cognitive belief in God’s transcendence.
• Children from Switzerland predominantly use basic colors like yellow, red, blue, and black, while those from the USA use more complex and varied color palettes, indicating cultural differences in artistic expression.
• The average intensity of yellow is significantly higher than green across all countries, supporting the hypothesis that yellow symbolizes divine or celestial presence.
• Green is predominantly used at the bottom of drawings, correlating with natural or earthly elements, reinforcing its symbolic role in religious cognition.
• Visual complexity, measured by Harris corner count and palette variance, increases from ages 1–7 to 7–11, peaking around age 11–13, then decreases in older children, indicating a non-linear developmental trend.
• Gray is the most common color in drawings with minimal coloring, suggesting that many children prioritize sketching over coloring, and that grayscale drawings are a distinct category in the dataset.