“If you’ve met one person with autism, then you’ve met one person with autism.” This adage has become often-repeated in the autism community. It speaks to the uniqueness of each individual with autism and to the fact that no individual can be captured or fully described by his or her diagnosis. At the same time, it reflects what has become one of the greatest challenges facing autism science: variability in symptoms across people diagnosed with autism, otherwise known as heterogeneity. No one feature is reliably seen in every individual with autism, and any research sample can reflect only a subset of the diverse individuals who fall along the autism spectrum. In the latest diagnostic manual of the APA, a new, broader diagnostic category – Autism Spectrum Disorder (ASD) – replaced the categories we had all become familiar with: Autistic Disorder, Asperger’s, and PDD-NOS. For my colleagues at the Yale Child Study Center and I, the new DSM-5 umbrella “ASD” category opened the door for developing novel ways of conceptualizing and clustering heterogeneity among ASD features.
In a commentary published in the Journal of Autism and Developmental Disorders1 this month, we suggest a new way to consider autism symptoms. Borrowing concepts from the schizophrenia literature, we propose a new framework within which ASD-related features can be categorized. We introduce the idea of positive, negative, and cognitive feature clusters as a novel way to conceptualize ASD symptoms. Positive features include behaviors not present in typical development, but present in ASD, such as circumscribed interests or stereotyped motor movements. The negative feature dimension captures behaviors that are present in typical development, but delayed, deficient, or absent in some individuals with ASD, such as eye contact, social engagement, and spoken language. Finally, the cognitive dimension reflects patterns of thinking, behavior, and relating that are cognitively-driven and common among individuals with ASD, such as rigidity of thinking and difficulty with switching between tasks. These categories cut across social-communication and repetitive behavior domains that are currently the primary means of clustering symptoms in the DSM-5, which translates to how ASD is thought of and its features are grouped in both clinic and research settings. It may be easier to conceptualize this by looking at the table below.
|DSM-5 Social Communication
|DSM-5 Restricted/Repetitive Behaviors|
|Positive Features||Intrusive social initiatives; Exaggerated prosody or intonation of speech; Pronoun reversal||Echolalia and stereotyped speech; Repetitive use of objects;
Repetitive hand mannerisms
|Negative Features||Difficulty with conversation; Lack of pointing; Reduced eye contact and range of facial expressions||Non-functional play with toys; Narrowed range of interests;
Lack of imagination
|Cognitive Features||Difficulties with theory of mind and taking another’s perspective; Difficulty with non-literal language||Insistence on sameness;
Rigid adherence to routines;
In this way, we offer new ways to conceptualize and organize hallmark symptoms of ASD. In addition, this way of thinking offers an opportunity to describe specific characteristics with new precision. For example, instead of indicating a child has “atypical facial expressions,” the new dimensions would allow separating of children who have exaggerated affect from others who show limited range of facial expressions.
In schizophrenia, the notion of positive, negative, and cognitive feature clusters has been quite useful to both clinical and research communities. In the clinic, assessing and labeling symptoms along these dimensions has been useful for deciding which medications to prescribe, predicting which patients will continue to struggle versus which will have quick remittance of symptoms, and identifying individuals in a “prodromal” phase before the onset of more acute symptoms. In the lab, this framework has been useful for clustering symptom dimensions in ways that correspond with experimental task performance and underlying brain differences. In other words, positive, negative, and cognitive symptom dimensions help researchers understand which underlying brain differences or cognitive and social processes are most affected in patients showing more positive versus more negative symptoms. This, in turn, provides clues to genetic underpinnings of different symptoms as well as new leads from which to develop treatments.
In the future, with further research, we hope to use what has been learned in schizophrenia and apply it to autism. It is our hope that the dimensions we propose offer new ways to capture and organize the heterogeneity that makes each individual with ASD unique in a way that makes us better able to talk about what we see, provide good clinical care, and solve the remaining puzzles of autism.
1 Foss-Feig, J. H., McPartland, J. C., Anticevic, A., & Wolf, J. (2015). Re-conceptualizing ASD Within a Dimensional Framework: Positive, Negative, and Cognitive Feature Clusters. Journal of Autism and Developmental Disorders, 1-10. http://link.springer.com/article/10.1007/s10803-015-2539-x