Dr. Leasha M Barry

Only 55.1% of young adults with Autism Spectrum Disorder (ASD) have held paid employment in the six years following high school graduation (Cameto et al., 2004). The interview process is recognized as a barrier for employment for people with ASD due to social skill deficits (Zaboski & Storch, 2018). Sereno and Hawkins (1967) suggested that people who use high rates of filler words (e.g., so, um, or uh) are considered less competent. Additionally, a lengthy response latency can indicate the respondent's uncertainty in their answer (Barker, 1946). The purpose of this study was to explore these variables during a virtual mock job interview. METHOD Participants were college students diagnosed with ASD between the ages of 18 and 21. Three males and one female volunteered to participate. Participants met virtually via Zoom for mock interviews where they were asked eight of the most common interview questions. These questions included: 1. Tell me about yourself. 2. Why are you looking for a new opportunity 3. Why do you want to work here? 4. Describe your current or most recent job 5. Where do you see yourself in five years? 6. What are your strengths? 7. What are your weaknesses? 8. Why should we hire you? During the first mock interview, participants were not provided with any feedback after answering the questions. Participants met with the investigator in-person to complete an online training module, which provided specific coaching for developing answers to the 8 interview questions. Participants then completed a second virtual mock job interview, answering the same 8 questions. The participants were provided with feedback after the second interview. The behaviors selected for analysis in both interviews were frequency of filler words (so, um, or uh) and question response latency. All interviews were recorded for data collection and Inter Observer Agreement. RESULTS This online curriculum intervention did not generate a clear pattern of change across participants. Participant 1 performed well in baseline, using no more than two filler words per question and had a maximum latency of 3.2 seconds, representing minimal change in latency and filler word frequency post-intervention. Participant 2 demonstrated marked improvement in response latency and use of filler words. In the baseline interview, Participant 2's longest response latency was 32.6 seconds. During the post-intervention interview, the longest latency for any question was 5.0 seconds. Additionally, filler words decreased in total by eight from baseline to post-intervention. Participant 3's data reflected increases in filler word usage in post-interview compared to baseline interview. However, the participant's response latency decreased for all but one question post-interview, indicating positive change for this participant. Participant 4's data varied significantly and showed no clear pattern in responding. The data reflect an increase in the frequency of filler words by seven from baseline to post-intervention and a decrease in response latency post-intervention. Figure 3 illustrates the cumulative decrease in response latency for all participants and across all questions with an almost 50% decrease in time.