Cognitive basis for the development of aesthetic preference: Findings from symmetry preference (2024)

Journal List
PLoS One
PMC7549785

As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsem*nt of, or agreement with, the contents by NLM or the National Institutes of Health.
Learn more: PMC Disclaimer | PMC Copyright Notice

PLoS One. 2020; 15(10): e0239973.

Published online 2020 Oct 12. doi:10.1371/journal.pone.0239973

PMCID: PMC7549785

PMID: 33045015

Yi Huang, Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Validation, Visualization, Writing – original draft, Writing – review & editing,^#^1,^2,^* Jinyun Lyu, Data curation, Investigation, Methodology, Resources,^#^2,³ Xiaodi Xue, Data curation,² and Kaiping Peng, Funding acquisition, Supervision²

Emily S. Cross, Editor

Author information Article notes Copyright and License information PMC Disclaimer

Associated Data

Data Availability Statement

Abstract

Where is the visual aesthetic preference rooted from and what’s its relationship with the perceptual preference that is emerging early? Do young children naturally prefer some visual stimuli or do they learn to appreciate visual stimuli for aesthetic pleasure? Here, for young preschool children who are on the age that the preferences are developing, we provide findings from a study to show that the interplay between early emerging perceptual sensitivity and perceptual exposure promotes the emergence of preschool children’s aesthetic preferences for simple visual patterns. Specifically in the experiments, 4-year-old children were exposed to either symmetric or asymmetric non-figurative forms in a perceptually demanding game; the group of children who received exposure to symmetric patterns showed aesthetic preference to the exposed patterns, while no preference was found in the group that received exposure to asymmetric patterns. The following recognition test then showed that the symmetric objects were differentiated better and remembered more clearly by the children, indicating that the symmetry was perceptually encoded better. These findings suggest that the early emerging perceptual sensitivity to ‘good features’ such as symmetry provides the prior cognitive prerequisites, allowing visual perceptual exposure to nourish the eventual formation of aesthetic preference. Thus, the preferences for aesthetic appreciation are likely the outcome of the interplay between biological and ecological adaptation.

Introduction

Visual aesthetic preference refers to people selectively preferring certain visual stimuli that cause them to feel aesthetic pleasure [1]. For example, people tend to find symmetric human faces more attractive than less symmetric ones [2,3]. Such pleasure could occur to people without being explicitly aware of their sources or without clearly associating them to a specific positive value [4,5]; in other cases, people find themselves becoming particularly attracted to an object at first glance, such as an abstract painting in a gallery. Although the visual aesthetic preferences exist across human’s history and different cultures [6,7], and affect people’s daily appreciation of pleasure, their choices of purchase [8] and even the development of their inter-person relationship [9,10], and furthermore, influence many professionals, such as graphic designer [11], artist [12,13] and even marketing strategy maker [14–16], the origin and mechanism of forming these aesthetic preferences are not well understood.

The objectivist view from aesthetic theory, dating back at least to Plato, “saw beauty as a property of an object that produces a pleasurable experience in any suitable perceiver” [17], emphasizing the role of graphic structure or of a feature itself in inducing aesthetic experience. Empirical studies with this view indeed had found human have aesthetic preferences towards certain visual “good features”, such as symmetry [18,19], curved shapes [7,20], moderate complexity [21] and high figure-ground contrast [22]. These features are either explained as being perceived less harmful and thus elicit positive, pleasurable feelings [20,23,24]; or having an organization of perceived “goodness”, meaning they possess a lesser amount of information [25,26] and are easier to perceive, therefore, the higher cognitive processing fluency causes positive aesthetic evaluations [27,28].

Among such features, symmetry is the one been studied extensively that imbedded in different objects from human faces and bodies [2,3,29], to concrete artworks and abstract visual patterns [30,31]. Beyond its link with the attractiveness of human faces, symmetry has been manifested as the best predictor of adults’ aesthetic preference for abstract patterns [32–35]. On the other hand, the perceptual preferences to “good feature” of symmetry have been manifested both for adults and children. For example, adults detect symmetric visual displays faster and more accurately than asymmetric ones [36,37], and remember symmetric patterns better [38,39]. Preschool children pay more spontaneous attention to a symmetric visual pattern over an asymmetric one [40]. Moreover, the perceptual sensitivity to symmetry has been found even emerging from as early as four months old [41]. According to perceptual processing theory, as long as the aesthetic evaluation is not associated with other sources, the experience of perceptual fluency will directly promote the a positive affective feeling [27,42]. Thus, the general aesthetic preference to symmetry seems naturally rooted from the perceptual preference to symmetry, just as it was mentioned by Weyl in his book, “beauty is naturally bonded with symmetry” [43].

However, a recent developmental study on 4-year-old children, on the contrary, had found the aesthetic preference for symmetry dissociates from early emerging perceptual sensitivity to symmetry [40]. With eye-tracking approach in this study, it was found that although preschool children paid more spontaneous attention to a symmetric visual pattern over an asymmetric one, they, unlike adults, did not report any preference when asked to indicate which pattern was more beautiful or preferred. Thus, the perceptual preference and aesthetic preference to “good feature” seem not naturally synchronized. Other recent studies even reported that the aesthetic preference to symmetry over asymmetry can be reversed due to long-term art-training expertise [44]; and that compared to explicit pleasure, the symmetry-induced implicit positive affect was found to be inconsistent across different task settings [45]. Therefore, these findings indicated that the formation of an aesthetic preference to symmetry may also be associated with perceiver’s interactive experience with the objects, as well as the evaluation contexts.

Indeed, beyond the objectivist view in aesthetic theory, researchers with a constructivist view emphasize the interactive experience between subjects and objects in the process of aesthetic preference formation [46,47]. Such research supports the claim that people prefer visual objects that have been exposed to them beforehand, with exposure potentially increasing the familiarity, which suggests that the objects are less harmful and are likely to elevate the feeling of pleasure [27,48]. The phenomenon of repeated exposure increasing people’s aesthetic evaluation is called the mere exposure effect [48–50]. According to this phenomenon, in the development of young children’s aesthetic preference, the large amount of visual experience they acquire every day should affect the formation of their aesthetic appreciation. However, it seems difficult to explain why most people eventually prefer certain kinds of stimuli, such as symmetry, and not their counterparts, such as asymmetry, because, theoretically people should have an equal chance of being exposed to all qualities of stimuli in their daily life. Would the processing fluency also benefit the exposure effect with respect to the formation of aesthetic preference?

Here in this study, we propose a possible mechanism to explain how the perceptual preference supports the development of aesthetic preference. We proposed the early emerging perceptual selectivity towards ‘good features’, e.g., symmetry, provides a perceptual prerequisite for developing children to be more sensitive to visual exposure of the features, causing later encounters with external exposures to selectively promote and shape the aesthetic preference towards those perceptually advantageous features. We focused on 4-year-old children, because in this age, children are already able to express preferences, but their preferences are still in developing, therefore, providing a natural time window to study the process of preference formation. To test the proposed hypothesis, across three experiments, we designed a picture-matching game to have children receive pure perceptual exposure before testing their preferences towards either symmetric or asymmetric pictures. We expected that through the proposed mechanism, visual exposure should affect the preference for symmetry but not asymmetry. After the preference test, we also implemented a recognition task by asking the children to recognize and differentiate a series of pictures they had or had not seen in the game. If symmetry indeed has a perceptual advantage, as reported by previous studies [38,39], symmetric pictures should be remembered and recognized better.

Experiment 1

Participants

Sixty-eight 4-year-old children (M = 4.45, SD = 0.63; 50% female) were recruited from kindergartens. They were randomly assigned to either an exposure to symmetry or an exposure to asymmetry group. Four children were excluded from the final data analysis because of missing data. Each condition included 32 participants (exposure to symmetry: M = 4.57, SD = 0.60; 50% female; exposure to asymmetry: M = 4.34, SD = 0.66; 50% female). The sample size was pre-determined by setting the statistic power as 0.95 and the effect size as 0.6 for one-sample t-test, and was calculated in the software of G*Power 3.1.9.2. All written consents were obtained from the parents before the experiments.

Stimuli and procedures

Exposure and preference task

Pairs of pictures of non-figurative forms were created as experimental stimuli. Each pair contained two pictures that showed objects with similar elements (e.g., straight lines, curved lines, rectangles, ovals) that were taken from Photoshop software but with one having a symmetric overall structure and the other asymmetric. Each asymmetric picture was formed by slightly moving the shapes from each symmetric picture either horizontally or vertically (Fig 1), so that with the exception of structural difference (symmetry or asymmetry), the other figural properties (e.g., the degree of curvature) within each picture pair were balanced as much as possible. These pairs were used both in the preference task and separated into individual pictures for the exposure game. To ensure that the children received adequate perceptual exposure and delivered equal amount of attention to symmetric and asymmetric pictures in the exposure game, a sticker matching game was designed. In this game, each participant received a pile of randomly mixed stickers, each having a single form that matched one of the 12 different sample forms on a sticker map. Children were instructed to match each sticker with its corresponding place on the map. The map was 21 cm×29.7 cm and had 12 grids printed on it. The grid size was 6 cm×6 cm, and in the center of each grid, one randomly selected sample form was shown. Each child would play only on the map containing either all symmetric samples or asymmetric samples to obtain exposure to symmetry or asymmetry (Fig 1). Children were told that each sticker would be scored as correct if put onto the right grid. To successfully complete the game, children had to observe and differentiate each picture carefully, which guaranteed perceptual exposure in both conditions.

Open in a separate window

Fig 1

The sticker matching game used to give children perceptual exposure and the stimulus samples of the paired non-figurative forms.

After completing the sticker game, participants took a short rest by moving around freely in the experiment room for a few minutes before being instructed to do the following preference task. For each trial, children would see one pair of forms presented on an iPad Mini screen. Each pair contained one form they had seen in the sticker matching game, symmetric or asymmetric, and a counterpart, asymmetric or symmetric, they had not seen. Children were instructed to play a role of a judge to point out which form in each pair was more beautiful and they liked better. All 12 pairs were shown in random order, and the location of the symmetric or asymmetric form was counterbalanced on the left or right side of the screen.

Recognition test

After completing the preference task, the children took a short break by moving around freely in the room before beginning the recognition test. In this test, the non-figurative forms that the children had played with in the sticker game were presented on an iPad Mini screen one by one in a random order. In addition to the 12 pictures children had seen before, 4 new pictures that were created under the same principles were mixed with the ones that appeared in the exposure paradigm. For each presented picture, children were asked if they had seen or played with this picture before, and their “yes” or “no” answers were recorded by the experimenter.

All the experimental protocols and participant consent were approved and performed under the regulation of Institutional Review Board (IRB) for human research in Department of Psychology in Tsinghua University.

Data analysis

For the preference task, the percentage of choice of symmetric forms for the symmetry group and the percentage of choice of asymmetric forms for the asymmetry group were calculated as the after exposure preference, and then each was compared with a 50% probability with a one-sample t-test to determine whether the exposure effect exists.

The independent-sample t-test was used when comparing the group differences.

The effect size for all tests across the tasks was Cohen’s d, which was calculated as d = $\frac{{\bar{X}}_{1} - {\bar{X}}_{2}}{S D}$ . And all the comparisons were two-tailed in order to test all the possible differences with a stricter statistical than one-tailed tests can offer.

Results

All children in experiment 1 were able to complete the sticker matching game within 5 minutes (M = 198.21 s, SD = 62.20 s) and have all the forms matched with the corresponding samples on the map. No group difference was found for the time duration of game completion (t (61) = -0.322, p = 0.748, d = 0.081, 95%CI [-36.90,26.66]; the completion time of one child in symmetry group was not recorded, so 63 children were included in this test). For the preference task, after the perceptual exposure introduced by the game, the group exposed to symmetry showed significant preference towards symmetric pictures (M = 76.56%, t (31) = 6.077, p < 0.001, d = 1.074, 95%CI [17.65,35.48]), but the group exposed to asymmetry did not show any preference (M = 47.14%, t (31) = -0.648, p = 0.522, d = 0.115, 95%CI [-11.88,6.15]) (Fig 2), thereby supporting our hypothesis that visual exposure would have differing effects regarding perceptually distinct features in influencing children’s aesthetic preference. Next, we further asked whether the symmetric forms we used in the experiment indeed had a perceptual advantage that was in line with previous studies using different formats of symmetry stimuli (e.g., dot patterns) [40].

Open in a separate window

Fig 2

Children’s aesthetic preference across all experiments.

Half violin plots with data points on the other half show a kernel density estimate of the full distributions of the percentage of choices for the preferred forms. Both in experiment 1 and 3, children who were exposed to symmetric forms showed a significant preference for symmetry compared with chance, while children who were exposed to asymmetric forms showed no preference. Children with no exposure to any type of form in experiment 2, showed no significant preference.

We tried to answer this question by determining whether the symmetric pictures were encoded more precisely by children in the following recognition test. The two groups of children, both the one given exposure to symmetry and the one given exposure to asymmetry, significantly recognized the pictures they had played with compared to a 50% probability of chance(symmetry: M = 79.17%, t (31) = 9.251, p < 0.001, d = 1.635, 95%CI [22.74,35.60]; asymmetry: M = 88.02%, t (31) = 12.708, p < 0.001, d = 2.246, 95%CI [31.92,44.12]) (Fig 3), which reflected a successful exposure induced by the game. The group difference showed marginal significance (t (62) = -2.037, p = 0.046, d = 0.509, 95%CI [-17.54,-16.55]). However, only the group in the exposure to symmetry condition made significantly correct rejections of the additional unseen pictures (symmetry: M = 75.78%, t (31) = 4.030, p < 0.001, d = 0.712, 95%CI [12.73,38.83]; asymmetry: M = 55.47%, t (31) = 0.827, p = 0.415, d = 0.146, 95%CI [-8.02,18.96]), and the correct rejection rate of the exposure to symmetry group was significantly higher than that of exposure to asymmetry group (t (62) = 2.207, p = 0.031, d = 0.552, 95%CI [1.92,38.71]) (Fig 3), suggesting a more precise encoding regarding symmetry during testing with the unseen pictures as distractors.

Open in a separate window

Fig 3

Correct recognition and rejection rates in the recognition test for experiment 1.

Half violin plots with data points on the other half show a kernel density estimate of the full distributions of the percentage of choices. Children who played with symmetric pictures had a significant correct rejection rate for the pictures that did not appear in the game, while children who played with asymmetric pictures could not correctly reject the pictures that they had not seen in the game compared with chance. Both groups of children correctly recognized the pictures they had played in the sticker game.

Experiment 2

In experiment 1, we found that no preference emerged from the exposure to the asymmetry group for neither asymmetric nor symmetric forms. This is in line with the previous finding that children at this age have not shown aesthetic preference for symmetry-related features [40]. However, we were still unsure if, with no exposure, children’s aesthetic preference for symmetry already existed for the current line-drawing forms and if the current results were caused by it. To rule out this possibility, we recruited another participant group in experiment 2 and tried to replicate the finding with the presently implemented forms to demonstrate that 4-year-olds do not yet show spontaneous aesthetic preference for symmetry.