Geometric forms are among the most fundamental components of a child’s educational environment. They appear in furniture, wall decorations, toys, window openings, floor tiles, and playground equipment, and a long tradition of environmental psychology has suggested that the shapes surrounding young children may influence their emotional comfort, spatial perception, and even learning behavior. Curved forms are often associated with warmth and approachability, while angular forms can read as more stimulating or assertive. Yet despite decades of interest in how built environments affect children, the authors note that empirical evidence about how shape preferences vary across cultures has been strikingly limited. Most existing work has been conducted in Western settings, and few studies have asked children directly, using methods suited to their age and verbal abilities.

To address that gap, the researchers adopted a field-based descriptive-analytical design centered on a child-friendly visual interview technique. Thirty children were recruited from each country, producing a total sample of sixty youngsters between the ages of four and six, comprising 26 girls and 34 boys. Each child completed four forced-choice trials in which they were asked to choose between the four most basic geometric forms: the circle, the square, the rectangle, and the triangle. Crucially, the stimuli varied in how they were presented. In one trial children saw colored three-dimensional versions of the shapes; in another, uncolored three-dimensional versions; in a third, colored two-dimensional versions; and in a fourth, uncolored two-dimensional versions. This design allowed the researchers to probe not only which shapes children preferred, but whether those preferences depended on whether a shape was rendered in color or in three dimensions.

The trial structure generated 240 individual responses, nested within the 60 participants, with each child contributing four choices. This nesting posed a statistical challenge, because repeated responses from the same child are not independent of one another. A simple analysis that treated all 240 responses as separate data points would have inflated the apparent strength of the findings. To handle this properly, the team used a multinomial generalized estimating equation model, a regression approach specifically designed for correlated, non-normal outcome data. The model explicitly accounted for the correlation among each child’s four responses, allowing the researchers to test the effects of country, gender, age, and presentation condition while controlling for within-child clustering.

The results were clear on several fronts. After controlling for the clustering of responses within participants, country remained significantly associated with shape preference, with a Wald chi-square statistic of 22.702 on three degrees of freedom and a p-value below 0.001. Gender also showed a strong and significant association with shape choice, yielding a Wald chi-square of 37.962 on three degrees of freedom and a p-value below 0.001. Age group mattered as well, with a Wald chi-square of 13.273 on six degrees of freedom and a p-value of 0.039. In contrast, the way shapes were presented made no detectable difference: presentation condition was not significantly related to preference, with a Wald chi-square of 5.150 on nine degrees of freedom and a p-value of 0.821, and neither was the interaction between country and presentation condition, which produced a Wald chi-square of 7.168 on nine degrees of freedom and a p-value of 0.620. In other words, whether a shape appeared as a colored cube or a plain flat outline, children’s choices stayed essentially the same.

The descriptive patterns behind those statistics are intriguing. Omani children chose the circle in 59 of their 120 trials, a rate of 49.2 percent, suggesting a marked affinity for curvature among the Muscat sample. Iranian children, by contrast, distributed their choices more heavily toward angular forms, selecting the square in 44 trials, or 36.7 percent, and the rectangle in 46 trials, or 38.3 percent. Age produced its own developmental gradient: four-year-olds chose rectangles most commonly, five-year-olds gravitated toward circles, and six-year-olds most often picked squares. Gender differences emerged as well, with girls more likely to choose squares while boys showed a stronger preference for circles and rectangles. The triangle, notably, was the least favored shape across most subgroups, rarely topping any child’s list.

Taken together, the authors argue, these findings demonstrate that children’s geometric preferences are not random or purely idiosyncratic but are systematically related to gender, country, and developmental stage. To interpret the cross-national difference, the researchers propose a cultural-developmental framework. According to this framework, broad perceptual predispositions, such as the widely reported human affinity for curved contours, may be modulated by factors including visual familiarity, educational exposure, and the surrounding architectural context. A child growing up in Muscat, surrounded by the domes, arches, and curvilinear ornamentation characteristic of much Omani architecture, may encounter and internalize curved forms differently than a child in Tehran, where the built environment and educational materials may present a different balance of rectilinear geometry. Classroom materials, textbooks, and culturally specific design conventions could similarly shape which forms feel familiar and comfortable.

The framework is offered carefully, and the authors are explicit about its limits. Because mechanisms such as visual familiarity and architectural exposure were not directly measured in this study, the country difference should be read as an exploratory contextual association rather than proof that national culture causes a particular geometric preference. The same caution applies to the gender and age findings, which are based on a modest sample of sixty children contributing four responses each. Still, the fact that presentation condition had no significant effect is itself informative: it suggests that children’s shape preferences are robust enough to survive changes in color and dimensionality, hinting that the preferences reflect something deeper about the child’s perceptual and cultural makeup rather than fleeting responses to surface features of the stimuli.

For architects, interior designers, and early childhood educators, the practical implications are potentially significant. If children in different cultural and developmental groups respond differently to the geometric composition of their surroundings, then a one-size-fits-all approach to designing kindergartens and preschool classrooms may be leaving value on the table. Designers working in Gulf contexts might consider how curvilinear elements could support the sense of comfort and belonging that Omani children in this sample seemed to find in circular forms, while those working in Iranian settings might explore how rectilinear geometry resonates with local children’s established preferences. Equally, the developmental shift observed across the ages of four, five, and six suggests that classroom geometry could be tailored to children’s stage rather than applied uniformly across early childhood.

The study also makes a methodological contribution. By using a child-centered visual interview technique with forced-choice trials, the researchers demonstrated a feasible way of eliciting preference data from very young children, a population that is notoriously difficult to study with conventional survey instruments. The statistical treatment, using generalized estimating equations to respect the nested structure of the data, offers a template for future studies of repeated preference tasks in child populations. The ethics of the work were reviewed by the Ethics Committee of Islamic Azad University, which waived the need for formal approval on the grounds that the study involved only non-invasive visual preference assessments in educational settings, posed no more than minimal risk, and collected no sensitive personal information. Written informed consent was obtained from the parents or legal representatives of all participating children, and each child gave verbal assent before taking part. Participation was voluntary, and children were free to decline or stop at any time.

The research was conducted without external funding, and the authors declare no competing interests. The article, published open access under a Creative Commons license, was received by the journal on 8 May 2026, accepted on 22 August 2026, and published on 4 September 2026. The authors describe the released version as an early share of peer-reviewed, accepted research, citable and carrying a permanent DOI, with a final Version of Record to follow.

Whether the preference for curves seen among children in Muscat reflects deep perceptual biology, cultural shaping, or an interaction between the two remains an open question, and the authors themselves stop short of claiming causation. But the study adds a valuable data point to a growing body of work suggesting that the environments we build for children are not neutral backdrops. The shapes embedded in walls, floors, and furniture may be quietly participating in children’s emotional and cognitive development, and understanding which forms feel right to which children, in which places, and at which ages, may help designers create spaces that genuinely fit the people they are meant to serve. Future research with larger and more diverse samples, and with direct measurement of the environmental exposure variables this study could only hypothesize about, will be needed to turn these exploratory associations into design guidance.