Work Chat: An Interactive Virtual Workday

SIMmersion, LLC, Michigan Career and Technical Institute (MCTI), National Institute of Mental Health (NIMH), Michigan State University, University of Pittsburgh

Nearly 50,000 youth with autism spectrum disorder (ASD) transition from high school to adult life each year with only 25% of these transition-age youth with ASD (TAY-ASD) getting jobs within 2 years of graduation. TAY-ASD's ability to sustain employment is even more challenging due in part to their social cognitive deficits (e.g., poorly reading social cues) that disrupt communicating with customers, coworkers, and supervisors. Research shows nearly 90% of job dismissals among TAY-ASD are attributed to poor work-based social functioning (e.g., poorly communicating with upset customers). The subsequent unemployment has damaging effects on their mental, physical, and economic health. A critical gap in federally-mandated services to support youth with ASD as they transition from school-to-adult life is the lack of evidence-based practice to enhance work-based social functioning. Given that TAY-ASD report computerized training tools (developed by the investigators and others) are highly acceptable and improve their real-world outcomes, the investigators propose to address this critical barrier to sustained employment by developing and evaluating a novel and scalable computerized training tool to enhance participant conversations with customers, coworkers and supervisors at work. The investigators propose to develop and test the effectiveness of Social Cognitive and Affective Learning for Work (Work Chat: An Interactive Virtual Workday), a computerized training tool. Work Chat will have three tiers of instruction designed to help TAY-ASD prepare for effective workplace communication. Tier 1 will adapt existing evidence-based practices to design an e-learning curriculum that trains social cognitive strategies to help guide work-related conversations (e.g., reading social cues or regulating emotions during supervisor feedback). In Tier 2, SIMmersion's PeopleSim® technology will enable TAY-ASD to apply social cognitive strategies learned in Tier 1 to repetitively practice simulated conversations with a fictional customer, coworker, or supervisor. In Tier 3, SIMmersion will innovate PeopleSim to exist in an interactive 3-D environment to create a virtual workday with interconnected activities were the actions made early in the day influence conversations later in the day (e.g., a poor customer interaction may result in constructive feedback from a supervisor). Phase I (Feasibility) was completed with application HUM00177878. Phase II (Efficacy) Aims: Aim 1) Complete Work Chat development using an iterative process that includes initial and ongoing individual-level feedback from Phase I participants and the community and scientific advisory boards to complete the product that will be evaluated in Aims 2-3. Aim 2) Conduct a Randomized Controlled Effectiveness-Implementation hybrid trial in a school setting to evaluate Work Chat. The investigators will recruit and randomly assign n=338 TAY-ASD (90% of sample frame) enrolled in school-based standard transition services (STS) to the Work Chat group (STS+SW) or a STS group (STS). The hypothesis is that STS+SW, compared to STS, will show greater gains in: (H1) social cognition and (H2) work-based social functioning; as well as (H3) reduced anxiety about work-based social encounters, and (H4) greater sustained employment by 9-month follow-up. The investigators will test mechanistic hypotheses (H5a-b) that social cognitive ability (H5a) and work-based social functioning (H5b) mediate the effect of treatment (STS+SW vs. STS) on sustained employment. For the implementation evaluation, the investigators will conduct a multilevel, mixed-method process evaluation of Work Chat's acceptability, appropriateness, and feasibility. Aim 3) Use community participatory research methods to prepare for commercialization. The investigators will conduct a parallel multilevel, mixed method implementation evaluation that focuses on the Work Chat delivery system regarding its feasibility, sustainability, scalability, and generalizability by conduct focus groups with delivery staff and administrators. These groups will discuss potential facilitators and barriers to Work Chat implementation, adoption, and sustainability.

This arm will receive services as usual plus the Work Chat intervention over a series of a few weeks. We estimate approximately 15-18 hours of training will occur.

Study participants will receive Standard Transition Services plus Work Chat will be a simulation training using virtual characters to role-play conversations. It will contain conversations with a coworker, a customer, and a supervisor within a workplace setting.

Measured by the hinting task (Corcoran, Mercer, & Frith, 1995), which comprises 10 short stories describing an interaction between two characters, all of which end with one of the characters making a hint. At the end of each passage, the participant is asked what the hint meant. If they give a correct response (score 2), they move onto the next item. If they fail, they are provided with extra information to help interpret the hint. A correct recognition of the hint at this stage scores 1. Total score can be 0-20, with a higher score being a better outcome.

Measured by the MiniPONS, which is a short, multichannel version of the established Profile of Nonverbal Sensitivity (PONS) test (Rosenthal et al., 1979). The short multichannel version (64 items) available here correlates very highly with the full version and shows reasonable construct validity through significant correlations with other tests of emotion recognition ability. A total score is calculated at the end, total score can be 0-64, with a higher score being a better outcome.

Change in Emotion Regulation - Emotion Dysregulation Inventory Short Form (Reactivity Index Short Form)

Measured by the Emotion Dysregulation Inventory (EDI) Short Form (13 items), which is a caregiver-report questionnaire designed to capture emotional distress and problems with emotion regulation without requiring verbal information. The EDI is comprised of two scales: Reactivity (7 items) and Dysphoria (6 items). Items on the scale are 0 (Not at all) to 4 (Very severe). A total score for Reactivity can be 0-28 and a total score for Dysphoria can be 0-24. Higher scores indicate higher dysregulation, which is a worse outcome.

Self-report employment surveys completed at baseline and follow ups. Employment will be tracked as 0 (no job) and 1 (job). Higher scores indicate a better outcome.

The Mayer-Salovey-Caruso Emotional Intelligence Test (MSCEIT) is an ability-based test designed to measure the four branches of the EI model of Mayer and Salovey. This study will only use the Managing Emotions Subscale, which measures the ability to be be open to emotions and feelings and integrate emotions with thinking. There are 29 scenarios presented, with higher score ranges indicating a positive outcome and lower score ranges indicating a worse outcome. The score ranges are as follows: Improve (worse), Consider Developing, Competent, Skilled, and Expert (positive).

Measured by the Social Interaction Anxiety Scale (SIAS) self-report survey. Items are on a 5 point scale with 0 being "not at all" and 4 being "extremely". Total score can be 0-80, with higher scores indicating higher anxiety, a worse outcome.

Measured by the Workplace Behavior Inventory (WBI), a 36 item survey using a scale of 1 (inferior performance) to 5 (superior performance), and 2, 3, 4 representing intermediate points on the performance continuum. Total scores range from 36-180 with higher scores indicating a better outcome.

Measured by Social Skills Performance Assessment (SSPA). Roleplay performance: The SSPA includes two brief (3-minute) roleplays in which participants engage in a conversation with an unknown confederate who plays the role of a new neighbor(NN) or a landlord (LL). The NN and LL role plays were video recorded at the pre- and posttest visits and were blindly rated by use of an anchoring system. The NN roleplay was scored on eight items via a 5-point scale (e.g., 5,very interested, to 1, very disinterested), and the LL roleplay was scored on nine items via a 5-point scale (e.g., 5,very focused, to 1, very unfocused). The two role-play scores were combined to create one total pretest score and one total posttest score. Sum scores could have a minimum of 17 and maximum of 85. Higher scores indicated better skills.

Inclusion Criteria for Students: 18-26 years old Screened for ASD via parent report using the Social Responsivity Scale 2nd Edition (SRS-2) (a score of 65T or higher on the SRS-2 will be used to reduce false positives) or have a record with MCTI of ASD Have at least a 4th grade reading level measured by the Wide Range Achievement Test 4th edition (this will be provided by MCTI) Exclusion Criteria: Medical illness that may significantly compromise cognition (e.g., moderate or greater traumatic brain injury) uncorrected vision, or hearing problems that prevent using the software Inclusion Criteria for Teachers/Administrators at MCTI: 18 or older (no maximum age) Currently working at MCTI