People with autism spectrum disorder (ASD) often struggle with imagining or understanding another person’s perspective or state of mind, so-called “theory of mind abilities.” Such individuals also have difficulties with social and conversational language (termed “pragmatic” skills). Research on ASD has been guided by the assumption that pragmatic difficulties are a simple reflection of problems with theory of mind. Thus, we might imagine that someone with ASD may not tailor his language based on what another person already knows (such as when conversational partners share background knowledge).
Our recent study published in the Journal of Child Language unveils a more complicated and perhaps surprising picture of conversational interactions and pragmatic language in ASD. We studied storytelling in order to answer the central question, “how do adolescents with ASD respond to shared knowledge with a conversational partner?”
Our results demonstrated that adolescents with typical development subtly altered their language in response to shared knowledge; specifically, their stories were shorter in the context of shared knowledge. In contrast, adolescents with ASD did not make these subtle adjustments – their stories were no shorter – demonstrating that, in this sense, they did not communicate differently based on the shared social context. On the other hand, additional study measures revealed that teens with ASD were sensitive to the social context and attempted to modify their stories accordingly. Specifically, we asked college students to rate participant stories for overall communicative quality. We found that college students were sensitive to differences in story quality based on the participants’ social context (that is: shared knowledge or no shared knowledge). These ratings revealed that adolescents with ASD did change how they communicated based on what their conversational partner knew; however, their strategy for incorporating shared knowledge was unsuccessful, resulting in less effective communication.
Next we probed why teens with ASD were essentially telling worse stories when they shared background knowledge with their conversation partner. We discovered that, in the context of shared knowledge, those with ASD were less likely to clarify or correct themselves when they stumbled during speech. One way to interpret this is that, since they were aware that their partner already knew what they were talking about, they exerted less effort in explaining some parts of their stories, resulting in stories that were harder for others to follow.
This method of combining the big picture (for example, ratings of communicative quality) with a detailed analysis of discourse revealed that adolescents with ASD were indeed aware of the social context and its relevance, highlighting a critical, yet under-recognized strength. Unfortunately, their strategy for incorporating the shared experience was unsuccessful, perhaps because storytelling itself is highly effortful. Pinpointing exactly where and how communication breakdowns occur will help inform targets for pragmatic language interventions.