A universal feature of autism spectrum disorders (ASDs) is disruption in social cognition—the ability to interpret thoughts and feelings of oneself and others. This disruption is related to struggles with behavioral, communicative, and social functioning. A young boy with an ASD might, for example, physically push a peer instead of asking her to step aside. Telling the boy not to push may seem like the default course correction. But a more promising approach may be to target the core social-cognitive deficit through a story that bolsters the boy's social understanding of the pushing event.

Known in the literature as a "social story," the approach has become one of the most popular ways to address the social-cognitive difficulties characteristic of ASDs (Hess, Morrier, Heflin, & Ivey, 2008; Reynhout & Carter, 2009). What's less clear, though, is what exactly social stories are and how best to use them.

These questions are complicated because guidelines for using social stories have been offered not only by their originator—educational consultant and former teacher Carol Gray (Gray & Garand, 1993)—but by other professionals working with children with ASDs over the last 25 years (Hutchins, in press). For Gray, a Social Story™ "describes a situation, skill, or concept" narrated according to 10 criteria. According to The New Social Stories™ Book, published in 2010, some of these key criteria include the use of:

• Background research and the collaboration of a team to answer who, what, when, where, why, and how.
• A patient, reassuring, and descriptive, rather than directive, approach.
• Language that is literally accurate and meaningful to the child.
• Careful consideration of formatting choices (e.g., length of story, font style/size, organization of text and illustrations).
• Specific sentence "types" that describe social information more than they direct audience behavior.
• A plan for editing, comprehension checks, and story review schedules.

When it comes to research, however, most studies have examined social stories (lowercase) that don't necessarily follow Gray's 10 criteria or formal definition. Indeed, researchers and practitioners have not adhered to a uniform method of development and implementation, and a number of social-story "myths" have emerged (see sidebar). Despite this variability, the research in this area is substantive enough to provide some empirically supported recommendations for best practice.

A Social Story in Action

"Jacob needs visual supports to understand the who, what, where, when, how, and why of our culture and how he fits and belongs."

—Mother of a 17-year-old boy diagnosed with pervasive developmental disorder—not otherwise specified, an ASD

So what, according to the existing research, goes into developing accurate, meaningful social stories? The following example provides an illustration. Jacob was enrolled in a general education classroom at his local high school and was receiving services from a speech-language pathologist, special educator, case manager, and paraeducator. He loved wind chimes and music and could play a variety of instruments in a truly gifted manner. Previous assessments indicated that Jacob had deficits in auditory comprehension, his verbal expression was limited to two- to five-word sentences, and his areas of strength were reading and writing.

One important communication-enhancement goal involved his voice, which was often soft and whispered and trailed into silence. Following an examination that revealed normal structure and function of the larynx, Jacob began voice treatment with a university-based SLP. Jacob worked hard over the course of several weeks to lengthen his sentences and find what he called his "excellent voice," which was louder and easier to understand. Unfortunately, he failed to generalize gains made in the clinic to other settings.

Because we knew that Jacob could produce an "excellent voice," we decided to develop a social story with some specific objectives. One was to enhance Jacob's understanding of the social importance of using his "excellent voice." Another was to recruit the participation of those with whom he had frequent and direct contact. We collaborated with Jacob, Jacob's mother, and his SLP to learn about the "who, what, when, where, and why" that surrounded the production of his different voices. We used the Motivation Assessment Scale (MAS; Durand & Crimmins, 1992)—a tool designed to help service providers understand the purpose behind behaviors—and our previous observations of Jacob.

We drafted a social story titled "Using my Excellent Voice [PDF]" and revised it with input from Jacob's mother to ensure accuracy of content. We also incorporated vocabulary important for Jacob that reflected his great interest in music (e.g., a "basso profundo voice"). When asked about pictures, Jacob noted that he would like "BoardMaker pictures" in his story so we added Picture Communication Symbols. We incorporated particular font and formatting choices to highlight key messages (for instance, different fonts represent soft and loud voices). We also made specific choices about the length and complexity of sentences and illustrations on each page.

We shared hard and electronic copies of the social story with Jacob, Jacob's mother, and his special education team members. They agreed to read the social story with him across contexts, such as home and various school settings, to facilitate acquisition and generalization of skills. The electronic version of the social story was particularly important for Jacob, who often used an iPod touch; we expected use of this medium to enhance Jacob's interest in and access to the story.

Our team identified a topic—Jacob's voice—that seemed critical to his success as a communicative partner. Through testing, discussion, and observation, we determined that the target represented an area in which Jacob had demonstrated some competency, which increased the likelihood that it was a developmentally appropriate target. We learned that Jacob was more likely to use his "excellent voice" when others spoke to him using short sentences and gave him more time to respond.

Research Basis

This example of the development and application of a social story is supported by research. But first, a caveat from descriptive reviews (Ali & Frederickson, 2006; Nichols, Hupp, Jewell, & Ziegler, 2006; Rust & Smith, 2006, 2011; Sansoti, Powell-Smith, & Kincaid, 2004) and quantitative literature reviews (Kokina & Kern, 2010; Reynhout & Carter, 2006, 2011; Test, Richter, Knight, & Spooner, 2010): Most authors agree that social stories are a promising intervention, but they universally acknowledge such methodological limitations as confounding treatment variables, lack of experimental controls, use of weak designs, and lack of research on efficacy, which demands rigorous experimental controls.

The conclusions of others have been more positive. For instance, The National Standards Project (National Autism Center, 2010) identifies social story intervention as an established treatment for ASDs for children ages 6–14, based on sufficient quality, quantity, and consistency in the evidence base.

Much of that evidence base points to effectiveness of the following practices, used in our work with Jacob:

• A team approach to gathering information. Although the research contains a few examples of appropriately rich data collection practices (Crozier & Tincani, 2005, 2007; Delano & Snell, 2006; Hutchins & Prelock, 2008; Lorimer, Simpson, Myles, & Ganz, 2002), experience suggests that these aspects are often underdeveloped or entirely overlooked in practice. To understand more fully the cause and purpose of problematic behaviors, social story authors are encouraged to—as done with Jacob—conduct observations and interview relevant others (e.g., parents, teachers, and other involved professionals).
• Performance of a functional behavioral analysis (FBA). This problem-solving process is designed to identify the purpose of a maladaptive behavior and strategies to address it. In our case, we used observation and the Motivation Assessment Scale to explore the purposes behind Jacob's behavior. In support of this approach, a recent meta-analysis concluded that studies that used some form of FBA yielded higher effect sizes than those that did not (Kokina & Kern, 2010).
• Use of comprehension checks. Before formally launching our intervention with Jacob, we introduced the story to him and checked his comprehension of the text and images. Ensuring that key messages in a social story are understood as intended has received empirical support (Kokina & Kern, 2010; Reynhout & Carter, 2006).
• Use of repetition in teaching of the story. In Jacob's case, he and all participants agreed to read the social story once or twice a week for several weeks to ensure constant exposure and reinforcement (Hutchins & Prelock, 2008; Norris & Datillo, 1999).

The research base is murkier on other aspects of social stories. They've been used successfully to address an extraordinarily broad set of intervention targets (Hutchins, in press), including behavioral targets such as tantrums and aggression, communicative targets such as greeting and eye contact, and social interaction targets such as peer engagement and joint attention.

In a recent meta-analysis, Kokina and Kern (2010) suggested that social stories targeting the reduction of problem behaviors may be more effective than those that teach appropriate social skills; however, researchers have rarely considered whether and which prerequisite skills or social understandings are needed to achieve social interaction goals (Kokina & Kern, 2010). In fact, from a theoretical perspective, any behavior that would benefit from enhanced social cognition may be an appropriate target of social stories when intervention is responsive to the individual's developmental level.

Social stories have been used successfully with children and adolescents with ASDs who vary widely in their cognitive and linguistic profiles. Gray (Gray & Garand, 1993) suggested
early on that social stories were likely to be most effective for individuals with good language skills. Subsequently, Gray (1998) suggested that, with modification, social stories can be used successfully for individuals with ASDs with more severe challenges. However, investigations of the use of social stories in this population are comparatively rare and the evidence so far is mixed (Barry & Burlew, 2004; Quirmbach et al., 2009; Swaggart et al., 1995).

A number of questions also persist about the best strategies for developing social stories. Gray (e.g., 1998, 2010) has asserted that sentences in a Social Story™ must be of a particular type and those types must adhere to a specific ratio. Readers familiar with the history of the social story literature will recognize various sentence types (e.g., perspective sentences, control sentences, directive sentences) that have undergone revision over the years. When researchers have systematically examined this issue, however, the usefulness of Gray's sentence types and ratio has been seriously challenged (see Okada, Ohtake, & Yanagihara, 2008; Quirmbach et al., 2009; Reynhout & Carter, 2006, 2011).

One point, however, is clear: What's critical in the development of a social story is the gathering of accurate and individualized information. Ideally, this process involves collaboration among family members, caregivers, and professionals. The collaborative process we used in Jacob's case helped us refine our thinking, coordinate workable plans, and develop shared responsibility for supporting his success. Indeed, the increasing popularity of social stories
may be explained not only by the positive effects they may have on the audience, but also by their potential benefits to authors (Ali & Fredrickson, 2006).