Home > Funding > Grant opportunities > Abstracts > A study of the autism phenotype

A study of the autism phenotype

Abba Krieger, Ph.D.
University of Pennsylvania

Abba Krieger and Andreas Buja at the University of Pennsylvania (in collaboration with Eric Fombonne, Edwin Cook and Alex Lash) analyzed the relationship between body temperature and cognitive functioning as measured by intelligence quotients (IQ) in individuals with autism.

They found that if one compares two children with autism whose body temperatures differ by 2 degrees Fahrenheit, the child with the higher temperature will on average have an IQ that is higher by eight points. The investigators examined whether this temperature-IQ relationship is spurious and can be explained by uninteresting factors such as gender or time or day of the exam. They found that it could not, even when a wide range of up to 27 ‘uninteresting’ factors were included in the analysis. It is therefore likely that the temperature-IQ relationship has an interesting underlying biology.

Krieger and Buja also worked with the Simons Foundation to summarize the large autism database in the Simons Simplex Collection. For each child, the database contains many hundreds of answers to survey questions. These answers need to be summarized for each child in order to measure the severity of a child’s behavior and personality problems. This can be done in many ways and some summaries have been designed to measure, for example, the severity of problems in social interaction or the use of language, or problems caused by repetitive behaviors.

An issue that has plagued such summaries is that they tend to correlate with IQ and age: Children with low IQs have more problems in other areas as well, and older children with autism have progressively more problems than do their typical peers. Low IQ and progression with age, however, is not specific to autism; it is found in many other developmental disorders. Krieger and Buja’s contribution has been to ‘adjust’ autism summary measures for IQ and age, thereby making them more specific to autism and independent of age. The new measures allow them to properly compare specific autism problems across IQs and ages.


proposalCENTRAL #121221