Images of Mary:
Effects of Style and Content on Reactions to Marian Art

Donald J. Polzella, Johann G. Roten, and Christopher W. Parker
University of Dayton

Abstract 105 college students rated digitized facsimiles of works of art depicting the Virgin Mary, which varied in style (Pre-Renaissance versus Renaissance and Post-Renaissance) and content (depictions of events versus devotional images). Viewer reactions were affected strongly by style, content, and their interaction. For example, the newer style was perceived to have greater power and clarity than the older style, and this effect was greater for event depictions than for devotional images. Other findings included differences in reactions between men and women.

This study fits within the long tradition of experimental aesthetics, through which scientists have attempted to understand and predict people's reactions to works of art. (See Berlyne, 1971.) Marian scholarship has a distinguished history dating back over 500 years, but there have been few attempts to examine scientifically viewers' reactions to artistic depictions of the Virgin Mary one of the most frequent subjects in Western art. A recent example is a survey reported by Roten (1992) in which college students were asked to indicate their preferences for various Marian images. A subsequent analysis of the results of that survey (Polzella, 1996) found that viewers preferred "devotional" rather than "affective" images, and preferred Renaissance and 19th century styles over the earlier Romanesque and Gothic styles. The original survey (Roten, 1992) was limited, however, in two ways. First, the images were presented as small black and white copies of the original works; much of the detail and subtlety of the originals was thus lost. Second, each depiction, theme, and artistic style was represented by a single image. There could be no way to determine whether the results might generalize beyond those particular examples, or whether artistic style and content might interact to affect viewer reactions. For the present study, digitized color reproductions of paintings and sculpture were used, and each depiction and artistic style was represented by several examples.


Participants The participants were 105 college students 38 males and 67 females who were enrolled in the Introductory Psychology course at the University of Dayton. Participation was voluntary, although it partially fulfilled a course requirement.

Stimuli and Apparatus The stimuli were 24 digitized facsimiles of paintings and sculpture depicting the Virgin Mary from the collection of the Louvre museum (Hatala, 1990). Half the works were from Renaissance and post-Renaissance periods and half were from pre-Renaissance periods. Half the works of each style depicted events (e.g., Annunciation); the other half were "devotional" images (e.g., Madonna and Child). A list of the art is shown in Table 1 and examples are shown in Figures 1- 4.

Full automation of the study was achieved through the use of a Macintosh Quadra 700 computer, which controlled the display of each work and prompted and stored the participants' reactions (Brake, Polzella, & Kozar, 1997).

Procedure and Design Participants indicated their reactions to each work of art on 10 seven-point semantic differential rating scales that measured such dimensions as complexity, clarity, power, beauty, and evoked pleasure (cf. Berlyne, 1973). A list of the scales is shown in Table 2. Participants were tested individually. The works were displayed one at a time, in random order, as were the rating scales, and each work remained visible while it was being rated. The design was a 2 X 2 factorial (style by content) with repeated measures on both factors.


A multivariate analysis of variance revealed that viewer reactions were affected strongly by style, content, and their interaction ( p < .001). Compared to the older Romanesque and Gothic works, the Renaissance and Post-Renaissance works were rated more pleasing, more beautiful, more powerful, and possessing greater clarity. Compared to the event depictions, the devotional images were rated more pleasing, less complex, less powerful, but possessing greater clarity. The two variables interacted in interesting ways as shown in Figures 5-10

For example, the effect of style on rated power and rated clarity was greater for event depictions than for devotional images. Many other effects were also observed, including differences between men and women participants. For example, women found Marian devotional art to have greater clarity than did student men. They also found the newer style of Marian art to have greater power, greater clarity, and greater intimacy than did the men.


The results of this study reinforce previous studies showing that viewers are sensitive to differences in artistic style (Bachmann & Vipper, 1983; Hasenfus, et al., 1983). More interesting and unique is the finding that the effects of style on viewers' perceptions were mediated strongly by a work's content. For example, the newer, more "realistic," style seemed particularly suited for depicting events, whereas both styles were effective for devotional works. The results also reinforce previous studies showing differences in men's and women's reactions to works of art (Bernard, 1972; Cupchik & Gebotys, 1990), although methodological differences preclude any direct comparisons. Finally, the results suggest that the methods of experimental aesthetics can provide insight into profound religious issues.

Two important questions remain:

(1) Will the findings apply to another group of art works? and
(2) How can the findings be explained?

This paper was presented at the 10th Annual Convention of the American Psychological Society, May 21 24, 1998, Washington, DC. The research was supported by the University of Dayton Research Council and Catholic Intellectual Tradition Forum. Send correspondence to D. J. Polzella, Department of Psychology, University of Dayton, 300 College Park, Dayton, OH 45469 1430 or INTERNET (


Table 1. Marian Art Works Used in the Study


Event Depiction

The Annunciation by Bernardo Daddi (ca. 1336)
Nativity by Guido da Siena (ca. 1270)
The Presentation in the Temple by Bartolo di Fredi (ca. 1400)
Crucifixion with a Carthusian Monk by Jean de Beaumetz (ca. 1390)
Death of the Virgin (French, early 13th c.)
Coronation of the Virgin (English, ca. 1350)


The Virgin and Child Enthroned by Cimabue (ca. 1270)
Virgin and Child with Saints by Simone dei Crocifissi (14th c.)
The Virgin, the Infant Jesus, and an Angel by Peter Paul Reubens (ca. 1617)
Virgin and Child of Sainte-Chapelle (French, ca. 1255)
Virgin and Child (French, late 12th c.)
Virgin and Child (Italian, 13th c.)


Event Depiction

The Annunciation by Charles de La Fosse (ca. 1705)
The Nativity by Giovanni di Petro Spagna (1506)
The Presentation in the Temple by Simon Vouet (ca. 1641)
Christ on the Cross, the Virgin, and St. John by Anthony Van Dyck (ca. 1622)
The Death of the Virgin by Michelangelo Caravaggio (1606)
The Coronation of the Virgin by Jacob Jordaens (17th c.)


The Virgin and Child Enthroned by Sassetta (1444)
The Virgin with Grapes by Pierre Mignard (17th c.)
The Virgin of Seville: The Holy Family by Bartholomé Murillo (ca. 1670)
The Virgin of Olivet by Ateliers of Touraine (ca. 1515)
Virgin and Child by Jean Dubois (17th c.)
Virgin and Child by Jacques Sarrazin (17th c.)

Table 2. The Scales

Figures 1-4

Figure 1. The Annunciation
Figure 2. The Presentation at the Temple
Figure 3. Virgin and Child Enthroned
Figure 4. Devotional Sculpture

Figures 5-10

Figure 5. Powerful
Figure 6. Clear
Figure 7. Makes Me Feel
Figure 8. Feminine
Figure 9. Intimate
Figure 10. Superior to Us

Return to the top of the article

Return to the Research and Publications Index


This page, maintained by The Marian Library/International Marian Research Institute, Dayton, Ohio 45469-1390, and created by Jim Danis , was last modified Thursday, 01/02/2014 15:19:54 EST by Kiran Shakapuram . Please send any comments to

URL for this page is