Running Head: THERAPEUTIC STORYTELLING GROUPS

Therapeutic Storytelling

Using Therapeutic Storytelling Groups to Increase Social/Emotional Awareness and Mind-Reading Skills in Preschoolers with Autism: A Case Study

Joan M. Reid, MSEd. and James Bellini, PhD.

SyracuseUniversity, Syracuse, New York

Correspondence concerning this article should be addressed to Joan Reid, 6633 Bennetts Corners Road, Memphis, New York, 13112. Electronic mail may be sent via Internet to .

Abstract

The purpose of this case study was to determine if a preschooler with autism would demonstrate increased social skills following participation in storytelling groups. Empirically-derived behavioral variables reflecting social initiation and social reciprocity were measured. Pre/post ratings of the communication and socialization subdomains of the Vineland Adaptive Behavior Scales were completed by the classroom teacher and parents. Results indicated a steady increase in the number of social initiations, but no clear pattern emerged for social reciprocity. Post-intervention observer ratings increased by one standard deviation for communication and by one-half standard deviation for socialization. The findings are discussed in the context of the theoretical linkages among narrative or storytelling, social-emotional skill development, and the development of mind-reading skills in young children.

Using Therapeutic Storytelling Groups to Increase Social/Emotional Awareness and Mind-Reading Skills in Preschoolers with Autism: A Case Study

Introduction

Autism is a disability characterized by a limited ability to communicate and interact with other people. One hypothesis for this limitation is that these children are impaired in the development of a theory of mind (Baron-Cohen, Tager-Flushberg, & Cohen, 1993). Most researchers now agree that a rather sophisticated level of reasoning about mental states of self and others is evident in three- and four-year olds (Wellman, 1993). The typical child creates mental representations of mental states through a set of inferences covered under a theory of mind, while the child with autism demonstrates a deficit in metacognition (Baron-Cohen, Leslie & Frith, 1985). Baron-Cohen (1995) describes this lack of awareness for people with autism to their own and others’ beliefs, desires, or intentions as mind-blindness.

Children with autism, as compared to typical children, exhibit facial expressions that are less expressive or inappropriate for the situation, and use fewer gestures that express an emotional state (Yirmiya, Sigman, Kasari, & Mundy, 1992). Typically, children with autism who are high functioning appear to have limited understanding of socially derived emotions. Although facial expressions are learned for each emotion, they are not used to communicate socially. A person with autism would have difficulty modifying the beliefs of another person by displaying a deceptive emotion (Dennis, Lockyer & Lazenby, 2000). For example, he/she can physically hide an object for a person but may not block the other person’s access to information about it (Baron-Cohen, 1992). Rieffe, Meerum-Terwogt and Stockmann (2000) predict that children from the autism spectrum have theory of mind capacities that remain latent but could be developed or revived when conditions are optimal.

In a study done with children diagnosed along the autism spectrum, Rieffe et al. (2000) reported findings that suggest that these children dohave the capacity for mind reading (i.e., they can acknowledge a mental state as an explanation of another’s emotion or behavior), but simply do not apply it as spontaneously as typically developing children do. Jarrold, Boucher and Smith’s (1996) findings strengthen the idea that the performance of children with autism will improve when they are appropriately prompted. Their research has shown that children with autism may engage in pretend play if it is highly structured or prompts are used to encourage pretence. Based on their findings, the authors suggest that these children have difficulty with generating pretend play, rather than with the mechanics of pretence itself.

In addition to increasing social engagement for children with autism, there is a need to increase the capacity for long-term inter-personal relationships. We have better methods to assess social relationships than we have to assess qualities of relationships (Rogers, 2000). Baron-Cohen’s (1995) theory of mind-blindness is promising as an explanation for the difficulty people with autism have in identifying their own and other’s emotions correctly, but more research is needed. He maintains that typical children can recognize not only simple emotional states such as happy and sad, but also belief-based emotions such as surprise. The child with autism has difficulty with this. For example, in children with autism, pictures of surprised expressions were mistaken for non-cognitive states such as yawning or being hungry, with the focus on the open mouth. When using story characters, Baron-Cohen (1995) found that children with autism could understand situations as causes of emotion when based on the characters’ desires. However, when using story characters’ beliefs, the children with autism had a great deal of difficulty predicting the characters’ emotions (1995). For example, in the story Snow White, when the wicked witch, disguised as a kind lady, offers Snow White the poisoned apple, the child with autism would predict Snow White’s emotion as scared rather than happy to get an apple. The typical child would know that Snow White believedthe witch was a kind lady and react accordingly. The child with mind-blindness is unaware that another person’s mental state or emotions may be different than his/her own.

Peter Caruthers (1996) built on Baron-Cohen’s theory of mind for children with autism by hypothesizing that these children might have severe difficulties accessing their emotions. While these persons with autism surely must have prepositional thoughts and emotional feelings, they do seem to have difficulty in knowing introspectively what their thoughts and emotions are. He further added that the lack of social imaginative play is likely due to their difficulty in manipulating their own mental states, a feature that is rewarding to typical children during pretend play. Wolfberg (1999) notes that play serves an important role in children’s emotional development. It can have a cathartic effect by allowing children to play out and master confusing or traumatic experiences. For children with autism, this is a special concern as they have difficulty with pretence even in the best of circumstances.

To be considered socially competent during interactions, children with autism need to engage in spontaneous social interactions. Research designed to develop interventions that teach spontaneity and initiations is needed (Koegel, 2000). Moreover, without adequate early social input, the neurological and behavioral development of children with autism may be further deflected from a typical developmental path (Mundy & Crowson, 1997). Strain and Hoysen (2000) stress the need for longitudinal, intensive social skill intervention for children with autism. “The new approaches should hold greater promise for children and families but also greater challenges to our ways of designing, executing and describing credible studies and longitudinal data sets. These are worthy challenges of all eager pioneers with a visionary spirit” (p.121).

The emotional development of children with autism is just beginning to be explored. The incidence of children diagnosed with autism appears to be on the rise. In the last ten years, the number of children diagnosed with autism served under IDEA has risen by more than 500 percent (Shaul, 2005). Children with autism are mainstreamed in our schools, and adults with autism are in the workforce. “Although a growing number of studies have assessed and developed interventions for improving social interactions between typically developing and disabled children, social-pragmatic areas of communication remain an area in need of further research” (Koegel, 2000, p. 383).

The therapeutic story telling group is an intervention based on the belief that the tasks of identifying and expressing feelings in self and others can be learned, and are best taught in a natural social setting. Preschoolers diagnosed along the autism spectrum, through participating in a story telling group that focuses on feelings and social skills, can gradually learn and practice their newly acquired skills. The purpose of this case study was to determine if a preschooler diagnosed with autism would demonstrate increased social/emotional awareness through participating in a therapeutic storytelling group.

Method

Participants

The subject of the case study, Billy (pseudonym), was four years old when he was diagnosed with autism by a team of developmental specialists, which included a licensed psychologist. At the beginning of the study, Billy was four years, six months old. The other story group members were two typical peers from Billy’s class. One member was a four year old girl and the other member was a three year old girl. Both girls were chosen as peer role models because they demonstrated average to above average social skills. The children chosen were members in the same class in an inclusive early education setting so that their opportunities to use their newly learned skills in a generalized classroom setting would be increased. The two cofacilitators had been conducting storytelling groups for three years. One cofacilitator had a Master’s degree in Marriage and Family Therapy and the other cofacilitator (first author) was a special education teacher with a Master’s degree in Special Education.

Operationalization of Variables

The story group intervention, implemented over the course of thirteen sessions during the school year (September - April), represents the independent variable in this study. The procedures used to operationalize the story group intervention will be described fully in the section on Procedures.

To operationalize the dependent variables in this study, a focus group comprised of a special education teacher, play therapist, psychologist and parent of a child diagnosed along the autism spectrum met to discuss what observable behaviors would demonstrate social skills. The group met for one hour on the campus of the early education center. After viewing segments of the taped sessions, the first author served as a facilitator as the group discussed what behaviors would best serve to indicate the use of social skills. Based upon what was observable in the video tapes, the group decided that two operational skills could serve as dependent variables.

The first variable, social initiation, was defined as instances when Billy used a person’s name to gain attention, verbalizations such as “Look at this,” looking in the direction of the person he was speaking to, and asking questions. The second variable, social reciprocity, included behaviors of responding to others’ attention by looking in their direction, verbalizations such as “What?”, following another’s agenda (i.e., not fixated on his own), listening to what others said by following directions, and completing a conversation circle.

In addition to the operational variables that were the main focus of this study, the researcher also used pretest and posttest measures of relevant adaptive behaviors to evaluate the effectiveness of the story telling group. The Vineland Adaptive Behavior Scales (VABS) were rated by both the child’s parents and special education teacher during the first month of the intervention and again at the end of the intervention.

Instrumentation

The subject attended thirteen groups during the ten month period, but two were eliminated for evaluation purposes due to shortened sessions. The eleven tapes were rated by two members of the focus group. The raters counted each behavior observed for the two dependent variables. The first five taped sessions were viewed simultaneously and the level of agreement between two raters was computed for target behaviors. When the two raters were within 95% agreement they independently rated alternating sessions for the remaining six sessions.

Pre and post intervention assessments using the Vineland Adaptive Behavior Scales (VABS) were completed by both Billy’s teacher and parents. The VABS was chosen as an additional indicator of treatment outcomes for several reasons. First, the observer rates how the child functions in everyday settings, contributing to the ecological validity of the assessment (Vig & Jedrysek, 1995). The VABS focuses on what the child can do, instead of what he/she cannot do (Sparrow, Balla & Cicchetti, 1984). Also, the scale is broken into domains which better describe the child’s functioning as related to the construct of social/emotional awareness, particularly the Communication and Socialization domains. The VABS provides data about actual behavior instead of potential behavior that intelligence tests predict (Meyers, Nihiri & Zetlin, 1979). The adaptive behaviors scored are typical behaviors needed for the performance of daily activities required for personal and social sufficiency (Meyers et al., 1979). The VABS standard scores are reported with a mean of 100 and standard deviation of 15. Lastly, the test authors report solid reliability data. Reliability of the survey form is as follows: Internal consistency – Split half means is .91 - .95. Classroom edition coefficient alpha means is .80-.95. Test/retest survey form means is .81 - .86 and inter-rater reliability is .62 - .78.

Only the subdomains of Communication and Socialization were used as pre- and posttest measures of adaptive behavior as these scales more closely reflect the constructs of social/emotional functioning that were the focus of the intervention. Some of the communication skills or related tasks include: listens to a story for at least five minutes, spontaneously relates experiences in simple terms, and expresses ideas in more than one way without assistance. Some items in the socialization subdomain include: responds to the voice of caregiver or another person, shows interest in novel objects or new people, expresses two or more recognizable emotions such as sadness, pleasure, fear or distress, shows interest in children or peers other than siblings, plays very simple interaction games with other children, laughs or smiles appropriately in response to positive statements and engages in elaborate make-believe activities alone or with others.

For the pretest assessment of adaptive behavior, parents and teacher were within 99% agreement on the Communication subdomain, and 93% agreement on the Socialization subdomain. For the posttest, the teacher and parents were within 96% agreement the Communication subdomain, and 92% agreement for the Socialization subdomain. The results of the data collected from rating the video tapes are shown in Figures 1 and 2. Table 1 lists the pre- and posttest scores obtained from the VABS.

Procedures

For the first story group, the cofacilitators taught the group members how to tell a story. They explained that there were three parts to a story. The beginning of the story always began with “Once upon a time,” so that the other members were alerted that it was time to listen to a member’s story. Using cues to alert the listener was modeled by the facilitators as well. The middle of the story was the part where the storyteller told what happened in the story. They were instructed that there was no right or wrong way to tell a story because it was their own story. However, the children were reminded that they could not use a story that they had seen on television or in a movie or from a book, because that story was someone else’s. The third, or last, part of the story was described as the part of the story where they could choose how it ended and that the other group members would know by the storyteller using the words “The end.” Before the children began their turns, one of the facilitators modeled how to tell a story. We learned from experience that although there is a risk of influencing the members’ stories, the concept of story telling was too abstract at first without modeling. Although the children may have started their stories similar to a modeled one, they soon digressed and were immersed in their own story.

Although various materials and activities were used to facilitate storytelling in this group, for the first two months, photos and pictures were primarily used. The children took turns telling their stories using a picture of their choice which included characters with different emotional expressions. While they were waiting for their turn, the other children were given materials such as playdoh, crayons and markers to meet their sensory needs. This provided opportunities to negotiate for space and materials. Once the students demonstrated that they (1) understood storytelling techniques and (2) could model/identify basic “feelings” pictures, we used a more spontaneous approach for the group. The children learned to use what was in their environment to tell stories, giving them more avenues of emotional expression. Lists of ideas for story telling materials and busy hand activities are included in the appendix.

While the child told his/her story, one of the facilitators wrote it down to read back to the group. This was an empowering experience for the child. While one facilitator wrote the story, the other facilitator asked questions or interrupted to ask what the person’s face in the story may look like. Although the content and story themes were important, the description/modeling of the actual feeling while storytelling was what we tried to elicit.