Drawings

Children's drawings drawings may be interpreted in a variety of ways. Tests such as the Bender-Gestalt and the House-Tree-Person provide information about perceptual-motor abilities and developmental level and can be useful for this purpose. But children's drawings are also used to assess possible sexual abuse (Conte et al., 1991; Kendall-Tackett, 1992). This latter use lacks empirical support.

The assumption underlying this use is that, since emotionally disturbed children are believed to reflect their problems in their drawings (e.g., Di Leo, 1973; 1996; Handler, 1996; Koppitz, 1968; Myers, 1978; Yates, Beutler, & Crago, 1985), the drawings of children who have been abused will differ from those of nonabused children. Free drawings, as well as the as the House-Tree-Person, Draw-A-Person, and Kinetic Family Drawings are used and qualitative features of the drawings, such as the colors used, the size and detail of body parts, and the shape of the figures may be interpreted in terms of the presence or absence of sexual abuse.

Burgess, McCausland and Wolbert (1981) claim that drawings in which a child exhibits a shift from age-appropriate figures to more disorganized objects or drawings with repeated stylized, sexualized figures indicate suspected sexual abuse. Sahd (1980) recommends using drawings as part of the evaluative interview of the sexual abuse victims and gives several examples of drawings that reflect abuse histories. Kelley (1984, 1985) believes that human figure drawings can be analyzed for "emotional indicators" (signs) in young children who are unable to verbalize their trauma.

Cantlay (1996) claims that distress and trauma, including sexual abuse, is reflected in drawings that include such signs as large heads, large, empty eyes, abundant hair, shaded clouds, knotholes in trees, large hands, large heads, large pointed teeth, abnormally tiny eyes, eyes without pupils, crossed eyes, excessive details, box-shaped bodies, poorly integrated body parts, lack of gender differentiation, hair that is long at the sides or thinning at the crown, wedge-shaped windows, extraneous circles, and large smoke trails coming out of the chimney. She cautions, however, that trauma can only be determined from a series of drawings which contain a number of these signs and that a single characteristic is not enough to indicate abuse.

The presence of genitalia is often considered a sign of sexual abuse because it is considered rare for normal, nonabused children to include genitals in their drawings (Di Leo, 1973, 1996). Cantlay (1996), Hibbard, Roghmann, and Hoekelman (1987), and Kelley (1984, 1985) claim that the presence of genitalia in drawings means possible sexual abuse, Yates et al. (1985) believe that incest victims either exaggerate or minimize sexual features in their drawings and Miller, Veltkamp, and Janson (1987) state that sexualized drawings indicate sexual knowledge beyond a child's years.

Empirical support for all these claims is extremely weak. In addition, a major difficulty with these assumptions is that sexually abused children are likely to have been interviewed about sexual abuse, have perhaps undergone a distressing genital examination, and/or have been placed into sexual abuse therapy. They are likely to have been shown undressed anatomical dolls where their attention is focused on the genitals. They may have been asked repeatedly to describe the details of the abuse. Therefore, genitalia may well become salient and it is not surprising that some of these children will include genitals in their drawings. In addition, other factors, such as family nudity, the birth of a sibling, or viewing an X-rated video may affect the tendency of a child to include sexual details in a drawing.

We have seen many examples of drawings by children that are erroneously interpreted as bolstering a conclusion of sexual abuse: For instance:

· A 4-year-old girl drew nothing but circles which she called "caves." One of the drawings looked to the psychologist like male genitalia-when asked what it was, the child said it was "a ball rolling into a lion's cave." The psychologist said this meant the child has been subjected to some type of traumatic experience. But when we saw evaluated the child, now age 5, we found that she was of borderline intelligence and could not do any kind of drawing task. All she could do was make scribbled circles.

· The child, when asked to draw a tree, also drew a cactus (Figure 2). This was interpreted in terms of "unconscious expression of danger and fearfulness." However, the child was not asked if she had a cactus in her yard (this was in Texas).

· The 7-year-old girl drew a picture of herself and her sister with their hands up in the air with the father standing next to them and smiling (Figure 3). She told the psychologist that she and her sister were "cheering at a show." But the psychologist claimed that this really signified a "helpless posture." She saw it as significant that there were no fingers drawn on the hands and that the hands were large on the father — she claimed that abused children put large hands on the drawings of their perpetrators. She also said that the thick lines in the crotch in the picture of the father meant an emphasis on genitals, was probably a penis, and showed anxiety about the father. She concluded that the girl, who continued to deny allegations of sexual abuse by the father, had, in fact, been sexually abused by the him.

· Hundreds of drawings over two years were interpreted by a psychologist who saw two girls in a day care case with allegations of ritualistic satanic abuse complete with costumes, masks, dead animals, sacrificed babies, blood, feces, skeletons, and monsters. These allegations only surfaced during therapy. The psychologist depended heavily upon the children's drawings in forming conclusions about satanic ritual abuse. But all of the drawings were typical of the types of scribbles and rudimentary figures drawn by 3- and 4-year-olds. In her deposition the psychologist said that what she believed was significant included practically every characteristic she believed she saw in the drawings:

Shapes that are untypical for 3- and 4-year-old children
Shapes that are phallic symbols.
Jiggly lines that indicate anxiety
Straight mouths that mean people can't say anything.
Jagged mouths that mean anxiety.
A mouth that is open and oval shaped
Darkened eyes
Eyeballs that are scribbled around
Eyes that are two different colors
Drawing something and then covering it up
Drawings something and not talking about it
Colors are very important and significant:
Black means the child is frightened or distressed; black is a morbid down color
Red means angry, unless the child is drawing a pretty red flower, when it is healthy
If everything is the picture is red or red and black, this is very suspicious.
Blue, brown, and orange mean fear, anger, and depression
Pink, red, and green are healthy colors

There are no empirical data to support these types of interpretations. Despite their frequent use in child abuse investigations, drawings are subject to the same criticisms as the anatomical dolls (Underwager & Wakefield, 1990; Wakefield & Underwager, 1988, 1989, 1994). Interpretations of drawings lack validity and reliability as projective assessment devices. In a review of the Draw-A-Person test in the Seventh Mental Measurements Yearbook, Harris (1972) states that there is little evidence for the use of "signs" as valid indicators of personality characteristics. In fact, there is so much variability from drawing to drawing that particular features of any one drawing are too unreliable to say anything about them. Reviews by Cundick (1989) and Weinberg (1989) in the Tenth Mental Measurements Yearbook note that there are no normative data establishing reliability and validity of the Kinetic Drawing System. Smith and Dumont (1995) state that four decades of research have failed to support the validity of human figure drawings in assessing personality, behavior, or intelligence. They observe that the confirmatory biases expressed in anchoring errors predispose clinicians to find support for their initial opinions in whatever material the client provides.

There are serious methodological problems with the few studies that claim to find differences between sexually abused and nonabused children. As with the anatomical dolls, studies with drawings with children suspected of being abused would have to take place before children underwent physical examinations, interviews, or therapy since these could affect their drawings. The persons who rated the drawings would have to be blind concerning the child's status regarding abuse. Ideally, a study would involve children who were all suspected of being abused who, based on subsequent evidence, could be divided into those who have and have not been abused. None of the studies meet these criteria.

Another common drawing technique is an outline of the back and the front of a naked male or a female. These cards are ostensibly used to enable the interviewer to learn what terms the child uses for body parts. However, this suggests that the purpose of the interview is to talk about sexual matters. The interviewer also may also tell the child to put an X where he or she was touched, which gives the message, "You were touched, now show me where." A recent New Zealand research report questions whether this constitutes a prompt or priming technique and states that drawings increased errors and confabulation in children's accounts (Rawls, 1996).

Nevertheless, drawings may be useful evaluation aids. They can provide information about possible perceptual-motor difficulties and developmental delays. They can help build rapport and encourage narrative accounts. Goodwin (1982) had children complete the Draw-A-Person task and the Kinetic Family Drawing and also asked them to draw whatever they wanted, to draw the whole family doing something together, and then to draw a picture of the alleged perpetrator. She reports that drawings were helpful in understanding the child's fears and anxieties, her view of the family, and her self-image, and in opening up a workable line of communication between the evaluator and the child. She cautions, however, that by themselves drawings are not sufficient to form diagnostic conclusions. Miller et al. (1987) report that, once a child has drawn a picture, he or she becomes highly verbal regarding the contents of the drawing. They believe that drawings are less threatening to a child who is hesitant to talk.