Quality of Communication in Young Adults with Autism during a Simulated Employment Interview

Friday, May 15, 2015: 11:30 AM-1:30 PM
Imperial Ballroom (Grand America Hotel)
W. Mitchell and J. Volden, University of Alberta, Edmonton, AB, Canada

Many adults with autism spectrum disorder have difficulty finding a job. Even when they are not intellectually disabled and are therefore “high functioning” (HFA), many are un- or under-employed. Communication skills play a key role in obtaining a job and one concern is that interviewers may judge the communication of adults with HFA negatively.    If so, adults with HFA may not advance beyond an interview despite being qualified. 


Our objectives were to determine:  (1) Whether listeners judge the language quality of adults with HFA differently than that of matched controls in a simulated employment interview? (2) What dimensions of communication (i.e., semantics, syntax, pragmatics or prosody) influence community listeners’ judgments of language quality in the simulated interview?  We expected that average ratings of quality would be poorer for the adults with HFA and that pragmatic items would most influence quality ratings in the HFA group.


Fifty-nine university students rated the “quality” of audio-recorded interviews collected from 20 young adults with HFA and 20 matched controls.  Quality was defined as: a) the amount of information conveyed b) the ease of understanding interviewee responses and c) how easy it seemed for the interviewee to participate in the interview.  Second-by-second reactions to communication quality were recorded electronically using an interval scale ranging from 0-100.  Following each interview, a 14-item questionnaire was administered to assess which dimensions of communication influenced listeners’ judgments of language quality. Listeners rated each item on a 7-point sliding scale (1 = no influence and 7= significantly influenced).  


Average communicative quality of the adults with HFA was judged as poorer (M=45.00) than communicative quality of the controls (M=73.63, t(27.23)= 7.76, p = .001). On the questionnaire, multivariate analyses for each communication domain revealed significant group differences on all dimensions; pragmatic, F(6,33)= 12.23, p<0.01; syntactic, F(2,37) = 6.70, p=0.003; vocabulary, F(2,37) = 17.70, p<0.01 and prosody F(4,35) = 7.55, p<0.01. Follow up ANOVAs revealed that specific pragmatic features, such as providing irrelevant or inappropriate details, using overly formal language, shifting topic unexpectedly, repeating topics, excessively talking about an event and providing limited responses were influential as were grammatical errors, vague vocabulary, use of unusual words, unusual voice quality, excessively fast/slow speech, and excessive use of pauses/interjections.

After listening to each interview a ‘yes/no’ question was posed to the listeners about whether or not they would proceed with a second interview or offer a job to the interviewee.  A Fisher’s exact test was conducted comparing the number of participants in each group where more than 70% of the listeners for each interview would recommend a second interview or offer the interviewee a job.  Only 30% in the HFASD group compared to 75% in the control group would be offered a second interview or job (p < .01).



Listeners readily recognize differences in communicative quality and they can have a negative impact on judgments made by conversational partners. These results may help explain why adults with HFA sometimes fail to advance beyond a job interview, despite being well-qualified.