Abstract:
In recent years, engagement with the arts has emerged as a topic of increasing interest for politics, research, and society on both a national and international level (Scheunpflug & Prenzel, 2013; Scheunpflug et al., 2021). With regard to the visual arts, for instance, this special interest is reflected not only in a widely available infrastructure in the form of museums and exhibitions, which allow everyone access to the arts, but also in the fact that visual arts are expected to yield beneficial effects on personal development and thus are an integral part of school education systems throughout the world (Americans for the Arts, 2022; National Art Education Association, 2022). Additionally, it is claimed that visual arts engagement foster not only creative activities and participation of educated citizens in society but might also play an integral role in developing social competencies (cf. BMBF, 2015; Timm et al., 2020). Social competencies, in turn, might determine the success of social interactions (Dede, 2010) and might be related to well-being, and quality of life (Collie, 2019; Eisenberg et al., 2015; Goodman et al., 2015). Especially, the ongoing Covid-19 pandemic has highlighted the consequences of social isolation and lack of visual arts engagement, for instance, on individual well-being (Hetland & Kelley, 2022; UNESCO, 2020, 2021).
Nowadays, it is not surprising that many scientists were particularly interested in studying what, when, how, where, and under which circumstances positive “side effects” (in the following mentioned as transfer effects) beyond direct effects from artistic activities occur (e.g., Mozart effect, Bastian et al., 2000; Rauscher et al., 1993). However, it become apparent that finding evidence for such assumed transfer effects is rather challenging as many studies could not convincingly demonstrate their existence (e.g., Hetland & Kelley, 2022; Winner, Goldstein, et al., 2013). Winner, Goldstein, et al. (2013) could point out in their meta-analysis that most studies in this research area possess theoretical and methodological flaws (e.g., a lack of theoretically justified measures, a lack of experimental studies, and a lack of sufficiently large group sizes). As a result, this leads to the conclusion that robust and reliable studies on transfer effects regarding different art forms have only rarely been conducted so far.
For that reason, a central aspect of this dissertation is to investigate presumed transfer effects of visual arts engagement with an optimized experimental study design that avoids many of the pitfalls of previous research as reported by Winner, Goldstein, et al. (2013). The main research question, thereby, is whether visual arts engagement can lead to socio-emotional transfer effects. In addition, the question arises how visual arts course programs must be designed and which underlying mechanisms might play a role, such as drawing activities and personal prerequisites. To answer the research questions, this dissertation developed a course program “Emotions, Self-Concept, & Epochs – Exploring Portraits With a Digital Drawing Pencil” consisting of three visual arts courses that differed only in their content (for an overview, Kastner et al., 2020). Two of the visual arts courses addressed socio-emotional skills, empathy, or self-concept using an instructional design, while the control course focused on historical periods. The course program was conducted over a period of three weeks at the Herzog Anton Ulrich Museum in Braunschweig. The data set underlying this dissertation for the evaluation of the course program was acquired in two collection phases. Generally, the empirical studies in this dissertation are based on data from N = 294 adolescents between the ages of 12 and 19, although not all data were included in all studies. Study 1 conducted an overall analysis (“general look”), examining all three visual arts courses for their specific effectiveness on socio-emotional transfer effects. In contrast, Studies 2 and 3 focused on the specific target constructs, empathy, and self-concept, separately for which transfer effects were found. In Study 2, the focus was primarily on the first target construct (empathy/emotion recognition), drawing activities and mechanisms known from the learning-by-drawing literature. Whereas in Study 3, the focus was on the second target construct (self-concept/self-complexity) and the personal prerequisites, such as personality traits.
Study 1 (Designing Visual-Arts Education Programs for Transfer Effects: Development and Experimental Evaluation of (Digital) Drawing Courses in the Art Museum Designed to Promote Adolescents’ Socio-Emotional Skills) investigated whether transfer effects of visual arts engagement on socio-emotional skills, namely empathy and self-concept, exist. Therefore, two “psychological instructed” visual arts courses focusing on socio-emotional skills (i.e., emotion course, self-concept course) were compared to a traditional course focusing on historical periods (i.e., epoch course, not focused on socio-emotional skills). The results showed that transfer effects on socio-emotional skills could be achieved, but only when they were properly designed into the visual arts engagement. Concretely, specific effects on the emotion recognition abilities as subfacet of empathy could be found for the emotion course (i.e., measured with an animated morph task, Schönenberg et al., 2014), as well as on self-complexity, for the self-concept course (self-concept task, following McConnell, 2011), were found. However, the effect on self-complexity was moderated by adolescents’ empathy. In contrast, no specific transfer effects of the epoch course on socio-emotional skills were found. In sum, this study shows that transfer effects on socio-emotional skills can occur, but only if the visual arts engagement is designed properly. To understand the socio-emotional transfer effects in detail, in two subsequent in-depth studies the underlying mechanisms and influencing factors, such as drawing activities and personal prerequisites, were analyzed.
Study 2 (Focusing on Emotions in Digital Drawing Courses Fosters Emotion Recognition: The Influence of Drawing Tasks and Drawing Activities) focused on the first socio-emotional target construct of empathy/emotion recognition in detail and potential influencing factors. Since specific improvements in the emotion course on emotion recognition abilities could be demonstrated, the aim of this study is to analyze drawing parameters that are known from “classical” learning-by-drawing literature as effective, such as prior knowledge about emotions, drawing quality and drawing tasks itself (e.g., Ainsworth et al., 2011; Ainsworth & Scheiter, 2021; Cook, 2006; Cromley et al., 2020). The results of this study showed that prior knowledge had no effect on drawing quality and, in turn, no effect on the emotion recognition abilities. In contrast, declarative knowledge about emotions was mediated by drawing quality. Since drawing quality could not explain the specific findings on emotion recognition abilities, a special interest of this dissertation was on the question whether emotion recognition abilities could be predicted by drawing tasks differing in their elaboration depth or by drawing patterns. For drawing tasks with deeper elaboration (i.e., intensifying, transfer, selfie drawing tasks) more correlations with the emotion recognition abilities were found than in tasks with less deep elaboration (i.e., tracing drawing task). Since these findings were only correlative in nature, a more holistic exploratory analysis was conducted based on a machine learning approach to generalize for unseen data and to have a look at the bigger picture. The aim of this approach is to go beyond individual correlations between drawing patterns and emotion recognition abilities by training a supervised machine learning algorithm to identify complex patterns of drawing behavior that can robustly predict high (or low) improvements in emotion recognition abilities when all features are considered simultaneously. Thereby, it could be shown that time pencil on tablet or pencil pressure might be able to predict course effectiveness on the acquisition of emotion recognition abilities.
Study 3 (How and Under Which Circumstances Might Visual Arts Engagement Support Adolescents’ Self-concept Development? Psychological Boundary Conditions for Socio-Emotional Transfer Effects of Visual Arts Programs) focused on the second socio-emotional target construct, self-concept/self-complexity, and the comparison with the epoch course. The baseline model used to explain the effects on self-complexity could only explain ~25 % variance in the data. In the following, special interest was on other influencing factors that might help to explain the data variance comparing the self-concept course with the epoch course . Therefore, an exploratory machine learning approach was again used to identify the most relevant patterns from personality (as well as drawing data) to predict course outcomes on a complex self-concept. Thereby, strongest feature importance values could be found for the personality traits of agreeableness and the drawing feature altitude. Based on these exploratory findings, moderators influencing the effects on a complex self-concept could be identified. Adolescents with low levels of empathy and agreeableness benefited most in terms of building a complex self-concept from visual arts engagement, while adolescents with high levels of empathy and agreeableness could not benefit. In sum, this study provides significant insights into the relation between personality, drawing and learning in the visual arts.
Overall, this dissertation makes an important theoretical and methodological contribution to research on the transfer effects of visual arts engagement in the context of promoting socio-emotional skills. This dissertation shows using an interdisciplinary approach that the design of visual arts engagement is important to promote socio-emotional skills (e.g., empathy or self-concept). However, the results showed that socio-emotional transfer effects could only be found under certain circumstances. Therefore, an important step ist to consider the underlying mechanisms of drawing and personality traits. Lastly, this dissertation provides implications for the design of cultural education programs (also in other areas of cultural education) for museums and health psychology.