Improved Scoring System for the Mutuality of Autonomy Scale

The proposed changes to the MOA scale include introducing positive and negative scores, making the scale symmetrical by allowing the same number of scale points for adaptive and maladaptive object relations and differentiating between relations based on mutual satisfaction, competition or aggression. Inter rater reliability was found to be comparable with values reported for the original scoring system, and the validity of the new scoring system was established.
Key Words: Rorschach, object relations, mutuality of autonomy.

Object relations - the internalized and generalized formative experiences with significant others during early childhood- form the basis for the appraisal of relationships later in life and for the strategies an individual will employ in initiating, maintaining and terminating them. .. The quality of one’s object relations is an important factor determining his or her psychological wellbeing and the degree of social adjustment. According to Fowler and Erdberg (2005) “The strength of object relations theory when applied to psychological assessment is that it provides an understanding of the complex interactions among self and object representations, defenses, pathological formations and ego strengths that make up the entire personality”.
Urist’s “Mutuality of Autonomy Scale” (1977) for the Rorschach, or MOA, measures a specific aspect of object relations, namely the ability (or lack thereof) to interact with others while maintaining a degree of autonomy. It is an indicator of the strength and integrity of the subjects’ internal boundaries. The MOA is a Likert type seven point scale, with the lower score indicating a more differentiated view of self seen in mutual interaction with other(s), and the higher score indicating decreasing differentiation, weaker boundaries, and increased dependency, loss of autonomy as well as malevolence and destructiveness. A short description of the scale points and scoring criteria is listed below:

Scale Point 1 .Mutual activity in which the individuality and autonomy of each participant is preserved. E.g. on Card III “Two people playing drums”
Scale Point 2. Parallel activity, without explicit acknowledgement of the figures’ individuality, but without loss of autonomy. E.g. on Card VIII “Two bears climbing a mountain.”
Scale Point 3. Diminished autonomy, as the figures are seen as needing support, leaning, holding on or hanging on each other. E.g. Card I “Two drunks hanging on to this woman in the middle.”
Scale Point 4. Increased loss of autonomy as one figure is seen as the reflection, shadow or imprint of the other. E.g. Card VII “A girl looking et her reflection in the mirror.”
Scale Point 5. Severe imbalance in the degree of autonomy. Relationships characterized by malevolent control. E. g. Card IX “A witch casting an evil spell to make this person do her bidding.”
Scale Point 6. The imbalance in autonomy is cast in destructive terms, for example in the depiction of a parasitic and/or violent relationship. E.g. Card I “Two vampires sucking the blood of this woman.”
Scale Point 7. Relationships characterized by overwhelming destructiveness and total loss of control. E.g. Card IV “An ogre devouring a child.”

The MOA has been used in numerous clinical studies , for instance to asses the object relations of male black children of absent fathers (Brown-Cheatham,1993), or to study its usefulness in making decisions regarding the continuation or termination of therapy (Ackerman ; Hilsenroth., Clemence; Weatherill and Fowler, 2000) Blatt, Tuber and Auerbach ( 1990) used the MOA to study the relationship between the representation of interpersonal relations and psychopathology, and Leifer, Shapiro, Martone and Kassem (1991) used the MOA, among other Rorschach measures to asses the functioning of sexually abused girls. . Lerner (1998) reported that using the MOA resulted in an improved insight in the nature of object relations.
The reported inter rater variability, according to Kelly (1999) ranges from .73 to .9, rising to .9 when scores that were within one point of each other where included.. The content validity of the MOA was examined by Bombel , Mihura and Meyer (2009) who found the MOA to be a “potent measure” of object relations and psychopathology...
As Fowler and Erdberg (2005) note, despite the fact that revisions of the MOA have been proposed (Urist & Shill, 1982; Spear & Sugarman, 1984; Holaday & Sparks, 2001) the majority of the studies continue to use Urist’s original scoring system.
Nonetheless, the present scoring system has certain limits and presents certain difficulties”. It is asymmetrical, with only two of the scale points reflecting well adjusted object relations and the remaining five maladaptive object relations. It is also counterintuitive, with the lower score representing the higher level of functioning and the higher score a lower level: three scores of 1, corresponding to the highest functional level add up to a score of 3, corresponding to a dependent relationship. Also, by rating only the actual relationship contained in the subjects’ answers, it overlooks the important dimensions of potential for interaction and past interactions as used in the AgPot and AgPast categories proposed by Gacono and Meloy (1994 ). Ag Pot (aggression potential), is used when the subject’s answer contains references to potential aggression, as in Card III : Two people getting ready to pull a crab apart”, and AgPast (aggression pas), is used when the subject’s answer contains references to past aggression as in Card VI: ” A cat beaten to death.”
We propose to widen the scorable interactions by including potential and past relationships following the criteria used by Gacono and Meloy (1994). The proposed changes would also result in a symmetrical scale, ranging from -5 to +5, making quantitative analysis easier. By making a distinction between relations based on mutual satisfaction, competition or aggressiveness regardless of the degree of autonomy involved, the new scoring system results in a more nuanced interpretation, while also maintaining the conceptual framework of Urist’s original work.

Here are the proposed changes in scale points and scoring criteria:
Scale Point 5: Mutual, positive and rewarding activity in which the autonomy of each participant is preserved E.g. Card III “Two people kissing.”
Scale Point 4: Mutual, neutral activity in which the autonomy of each participant is preserved. E. g. Card II “Two men shaking hands.”
Scale Point 3: Mutual competitive activity in which the autonomy of each participant is maintained. E.g. Card VIII: “Two lions pulling something in opposite directions”
Scale Point 2: Mutual, aggressive activity in which the autonomy and integrity of each participant is maintained. E.g. Card II “Two men fighting.”
Scale Point 1: Parallel neutral activity in which each participant remains autonomous. E.g. Card X “Two crabs. Each is carrying a leaf.”
Scale Point -1: Activity during which participants need support, indicating some loss of autonomy. E. g Card II “Two drunks leaning on each other”, Card III “Two people holding on to a railing”
Scale Point -2: Answers in which one participant is the mirror image, shadow or imprint of the other, implying a loss of mutual autonomy. E.g. Card VII “A girl looking in the mirror”, Card III “A bird and its shadow.”
Scale Point -3: Interactions characterized by malevolent control of one participant by the other E. g. Card IX ” An alien using his powers to control this other guy.”
Scale Point -4: Interactions characterized by a parasitic or destructive relationship, in which one participant lives off the other, and interactions which imply loss of body integrity. E.g. Card I “A vampire sucking the blood of a woman, Card II.”A guy cutting off the legs of this other guy.”
Scale Point -5: Interactions characterized by significant destructiveness E. Card IV “A giant stomping on these people”, Card VI “A tsunami washing away a village.”

The following indexes can be computed from these scores:
1. Highest positive score and Highest negative score
2. Average positive score and average negative score. Comparing these indexes allows the examiner to establish whether the subject has a pervasive manner of relating ( high positive or negative scores) or whether he/ she is ambivalent in regard to relationship ( both high positive and high negative scores)
3. The total positive score divided by the number of answers for which a MOA score was recorded, and the total negative score divided by the number of answers for which a MOA score was recorded. This would permit the comparison of protocols that differ in length.

In addition, the proposed changes permit a more comprehensive qualitative interpretation of the
subjects’ object relations. For example, the examiner can establish whether there is a pervasive pattern (relations that are mutually rewarding, competitive neutral, dependent, or destructive) or a mixed pattern and, in the latter case, make inferences based on other clinical indicators in the protocol such as blends, content, variations in form quality, special scores and defenses.

Inter rater agreement was found to be comparable with the values reported for Urist’ scoring system for the MOA. 27 graduate psychology students, blind to the hypothesis, were presented with 15 protocols containing 43 scorable interactions and were asked to score them using the above listed scoring system. 26 of the 27 subjects identified all 43 scorable interactions. The percentage of overall agreement, measured with Fleiss’ kappa was .80221 with a free marginal kappa of .75277, indicating a high degree of inter rater agreement.
The construct validity of the Urist’s original MOA scale was assessed positively by Bombel, Mihura and Meyer (2009), who concluded it was a strong and valid measure of object relations. In order to obtain a measure of concurrent validity, the same 27 students were asked to score the same 15 protocols using Urist’s original scoring system. The average score for each of the 43 scored interactions in each scoring system was computed and the Pearson correlation coefficient was calculated in order to obtain a measure of concurrent validity. The resulting values of r = -.7612, T = -7.2366, p/2 sided = 1.1794, p/1 sided = 5,897 confirm that the validity of the new scoring system .(The correlation is negative because in Urist’s scoring system the highest level of object relations receives the lowest score, while in the new system the opposite is true.)
I believe that the proposed changes make the scoring easier and more intuitive, and allow for more complete and complex interpretations of the subjects’ object relations, resulting in a more accurate understanding of their conflicts and defenses.


Ackerman, .S. J., Hilsenroth, M. J., Clemence A. J., Wetherill R.& Fowler, J. C.
The effects of social cognition and object representation on psychotherapy continuation. Bulletin of the Menninger Clinic 64 (3), 386-408, (2000)
Bombel,G. ,Mihura J.G.,& Meyer G.J. aAn Examination of the Construct Validity of the Rorschach Mutuality of Autonomy (MOA) Scale Journal of Personality Assessment, 91(3), 227 – 237. (1990)
Blatt, S.J., Tuber,S.B.,& Auerbach, J.SRepresentation of Interpersonal Interactions on the Rorschach and Level of Psychopathology Journal of Personality Assessment, 54( 3- ) , 711 – 728 .(1990 )
Brown-Cheatham, MThe Rorschach Mutuality of Autonomy Scale in the Assessment of Black Father-Absent Male Children Journal of Personality Assessment, 61(3), 524 – 530. .(1993)
Fowler, J.C.,& Erdberg P. The Mutuality of Autonomy Scale:An Implicit Measure of Object Relations for the Rorschach Inkblot Method. South African Rorschach Journal 2(2) 3-10. (2005/2006)
Gacono, C. B. & Meloy, J.R. The Rorschach assessment of aggressive and psychopatic personalities. Psychology Press. (1994 )
Holaday, M., & Sparks, C.L. Revised guidelines for Urist's mutuality of
autonomy scale (MOA). Assessment, 8, 145-154. (2001).
Leifer, M., Shapiro, J. P., Martone, M. W.. & Kassem. L Rorschach Assessment of Psychological Functioning in Sexually Abused Girls. Journal of Personality Assessment, 56 (1) 14 – 28. (1991).
Lerner, P.M. Psychoanalytical Perspectives on the Rorschach.Routledge. (1998)
Kelly, F.D. The Assessment of Object Relations Phenomena in Adolescents. TAT and Rorchach measures. Lawrence Erlbaum Associates. (1997)
Spear, W. E. & Sugarman, A. Dimensions of internalized object relations in
borderline and schizophrenic patients. Psychoanalytic Psychology, 1, 113-129. (1984).
Urist, J. The Rorschach test and the assessment of object relations. Journal of Personality Assessment, 41,3-9. (1977)
Urist, J.& Schill, M. Validity of the Rorschach mutuality of autonomy scale. Journal of Personality Assessment, 46, 541-454. (1982)