“Revisiting P. Schilder’s Theory on Body- Image, by the use of Repertory Grid: Perspectives on Systemic and on Dance Movement Psychotherapeutic Approaches”


The thesis revisits Schilder’s theory on body-image, informed by systems theory and Dance/Movement Therapy (DMT) literature. A research conducted with college students (N=317), focuses on the body-images of Self (S), Mother (M), Father (F) and Ideal Self (IS) and questions their interconnections as a structure, in relation to the emotional distance between a person and his/hers parents, the degree of differentiation from mother and father, the image of the relational self, self – esteem and sex of the participant.
Spatial dimensions and dynamic variations of the (S-M-F-IS) structure are both examined by the use of our “Relational-Body Image Grid” (RBIG) created for the needs of this research. Results confirm Schilder’s theory: spatial distances between inner- body images are related to emotional distances between persons. Yet, this relation appears to be more complex than the simple analogy he speculated. The use of the RBIG reveals the role of the Ideal body image as a balancing factor in the dynamic process of the differentiation, in relation to the parents. This insight creates a common, mutually enriching ground that can be shared by Family Therapy and Dance Therapy with anorectics. Sex differences in Body Image’s inner space organization seems an important factor that must be taken in consideration while planning DMT interventions and approaching phenomena like body transference and counter transference.
Using Schilder’s postulations as starting point of the research has been a fruitful choice:
The exploration of structural aspects of the body-image system offered reflections and research questions that supply a constructivist approach on body-image and have implications for theoretical advances and therapeutic interventions in the DMT field.
The exploration of dynamic aspects in relation to the emotional environment led to conclusions that inform a constructionist psychological understanding of the body-image concept and have implications in the context of family therapy and anorexia nervosa.
The discussion highlights the potential of these findings as a building step towards conceptual and methodological tools needed, in order to achieve a meaningful integration of the diversity that characterizes the body-image field and to formulate relevant research questions.

Aims of research

This PhD thesis is an attempt to elaborate on the following ideas:
1) Applying a systemic way of thinking on body image is a necessary step to be taken towards an encompassing theory on the topic.
2) Dance Movement Therapy (DMT) would particularly benefit from a systemic approach on body-image.
3) Schilder’s interdisciplinary approach provides a rich ground for such an effort.
The aim of the research is
1) To examine Schilder’s theory on body-image under the light of Living Systems theory and to investigate his statements’ potential for providing meaningful hypotheses and data.
2) Explore the subjective-intersubjective space of personal body-image based on Schilder’s conceptions, systemic- constructivist methods and DMT ideas.

Principal research question(s)

According Schilder (1935) each personal body-image is “an inner unity” formed by “body images of other people, and parts thereof, together with one’s body-image”. Body-images interact as parts as well as wholes, through the synergic action of identification and distinction, or, the recognition of partial similarities and differences between the bodies of self and others. Permeability of boundaries, identification and similarity is expressed in small distances, while defined boundaries, distinction and difference is expressed in large distances between the body images contained in this “inner unity”. Distances between body-images are analogous to the emotional distances between the corresponding persons.
To examine this proposition the research focuses on 4 body-images: Self-(S), Mother- (M), Father-(F) and Ideal Self-(IS). We hypothesized that the (S-M-F-IS) sub-unity makes part of the “core” structure of a person’s own body-image, having the potential to influence his/hers behaviors on different levels. The research objective is to explore the existence (or not) of spatial and dynamic interconnections between (S)-(M)-(F)-(IS) body- images and then, to investigate if these are meaningfully related with emotional distances between a person and his/hers parents, with the degree of differentiation from mother and father, the image of the relational self, and self – esteem. Possible influence by the sex of the participant is also investigated.

Principal findings

Statistical processing of the RBIG data focused mainly on relations between Elements: spatial interconnections (PCA), repartition of similarities/differences (Euclidean distances) and strength of interrelations (Pearson, r)
-Differences between Men-Women:
PCA plots reveal distinct male and female patterns: Male pattern is characterized by larger distances, lower degree of density and adjacent disposition between the elements that consist the Self, Father, Ideal Self body-images, while Mother body-image elements are situated on the diametrical opposite quadrant in the plots. Female pattern is characterized by narrower distances, higher degree of density and adjacent arrangement of elements between all four body images. Euclidean distances in Men’s data show emphasis on similarity between (S)-(F) and on difference between (S)-(M). (I S) body image is also estimated as similar to (F) and different to (M). Women’s data show emphasis on similarity between (S)-(M) and (S)-(F) while (IS) body image appears different from both (M) and (F). Results show differences from mother’s body as the main differentiation factor in men while in women relevant necessary space is provided by the differences of Ideal Self body from both Mother and Father. The above results are discussed in relation to the Shape / Effort analysis of male – female movement patterns (Davies, E. 2006) and to family therapy literature (Skynner, R., 1987) Statistical processing of the construct ratings (percentage of extreme ratings (1, 5) per element and sex) shows different dispositions for Men and Women in the use of constructs provided, coherent to the previous outcomes.
-Common patterns in both sexes:
Control of correlations (Pearson, r) between (S)-(M), (S)-(F), (S)-(IS), (M)-(F), (M)-(IS) and (F)-(IS) distances shows large number of statistically significant correlations, revealing dynamic interconnections.
Euclidean distance rates between (S)-(M) and (S)-(F) body-images correlate respectively with the (S)-(M) and (S)-(F) emotional closeness/farness rates (Men: r=.31 / r=.36, p<.o1, Women: r=.20 / r= .23, p<.01) Control of correlations between all sets of data revealed the following pattern of interconnection: emotional proximity facilitates the discernment of differences between the real body-image of self and that of the parents. At the same time though, it enhances recognition of similarities and identification to the parents, on the ideal-self- body level. Emotional farness is related to identification between Self and Parent body- images, and differentiation between Ideal Self and Parent body-images. The “ideal” image appears to have an important role in the adjustment of the subject’s body-image to the needs of his /hers relationship to the parents. The triadic relation between self/parent/ideal body-images behaves as a negative feedback loop, regulating separateness / connectedness of body-image under either conditions of emotional closeness or farness. The outcome is discussed in relation to Object Relations & Rep Grid literature (Boker, H. 2004)
The Johnson subscales showed higher degree of differentiation- individuation in relation to parents among men than among women. The two groups were significantly different in relation to Mother (t= 4, 02, p= 0,000) and Father (t= 3, 86, p= 0,000). No significant correlations were found between the Johnson’s scale data and any other variable of this research. On the contrary, and supporting the research hypotheses, a large number of significant correlations were observed between the RBIG Distance rates and the variables revealed by Factor Analysis on the Independence-Interdependence Scale ratings. High rates on Self-Esteem Scale were shown to be significantly correlated to similarity between S-IS (Women: r=-.41, p<.001, Men: r=-.23, p<.001).

Research Supervisors
Papadioti Vassiliki, Karagiannopoulou Evangelia, Barbousi Vassiliki
Client group
Any psychotherapy setting
Nature of data collection
317 students of the University of Ioannina participated in this research (women=237, men=80, mean age 21yrs, SD= 2,3). Data on spatial dimensions and dynamic variations of the (S-M-F-IS) structure were retrieved by the use of the “Relational-Body – Image Grid” (RBIG) created for the needs of the specific research. Eight relational-body aspects (Gaze, Embrace, Face, Body, Voice, Physiognomy, Figure, Person as a whole), related in turn to (S), (M), (F), (IS) generate 32 Elements as “my own voice”, “my mother’s gaze”, etc. Participants were asked to rate each one, in terms of 14 Construct dimensions such as “tender … violent, tense … relaxed” (five-point scale). They were also asked, by independent question, to estimate the emotional closeness / farness of Self-Mother, Self-Father and Mother-Father on a seven-point scale. Data were also collected by the “Relation to Mother- Father Scales” (Johnson, A. 1998), the “Independence- Interdependence Scale” (Singelis, T. 1994), and the “Self- Esteem Scale” (Fleming, J.)
Ethical board approving research
Giotsa Artemis Zaragas Harilaos Savrami Katia Papastathopoulos Stathis
Research design
The research plan includes exploratory and descriptive research parts. The first (exploratory research) aims to investigate the body image unity (S-M-F-IS) per se as a structure, while the second one (descriptive research) aims to give information about elements possibly related to the characteristics of the body image unity under question. The Repertory Grid method (Kelly G., 1955) has been used in the exploratory part. Rep Grid is a constructivist method “compromising idiography and nomothetics: the idiographic information collected with it can be statistically analyzed” (Schoeneich F. et al., 2004). Quantitative methods have been used in the second part (descriptive research). Evaluation of the results takes in consideration objective criteria as distinctiveness and statistical significance. The process followed constructivist epistemology criteria: subjective coherence in between consecutive statistical processing of Grid results, inter-subjective consensus in between Grid outcomes and those from the rest of the tools used in the research and indirect comparison with “objective “environment defined as the theoretical context to which the research aims to deliver, as well as findings provided by other researchers (Rescher, N. 1973, Thagard, P. 1989)
Registered Academic Institution
University of Ioannina – School of \Philosophy / Social Sciences - Psychology Dpt.
Academic level of research
Related publications by members of research team
Kita E. (2012) “Inner and outer engendered space: A research by the use of the Relational- Body Grid” (poster) 17th Herbstakademie, Heidelberg
available at: http://baobablab.org/wp-content/uploads/2021/09/Poster-Heidelberg-FINAL.jpg
Kita E. (2012) “Grounding the body – image: A systems approach inspired by P. Schilder” 17th Herbstakademie, The implications of embodiment: Enactive- Clinical Social, Heidelberg
Available at: http://baobablab.org/wp-content/uploads/2013/12/grounding-the-body-image.pdf
Kita E. (2012) “From Perceptive Phenomena to Interpersonal Relations: Body-Image as a Nodal Representation in Systemic Dance Therapy” (in Greek) 3rd Conference of Cognitive Psychology, Ioannina
Available at: http://baobablab.org/wp-content/uploads/2021/09/---.pdf