科技創新意識促進職業重建專業人員對沉浸式擴增實境職業技能培訓系統之接受度

Technological Innovativeness Promoted Acceptance of an Immersive Augmented Reality System for Vocational Skills Training Among Vocational Rehabilitation Professionals

本研究以科技接受模式為理論基礎，探討職業重建專業人員對沉浸式擴增實境職業技能訓練系統之接受情形，並納入科技創新意識作為外在變項，瞭解科技創新意識於科技接受模式中的角色，以補足過去相關研究之不足。
71位職業重建專業人員參與本研究，參與者實際體驗沉浸式擴增實境職業技能訓練系統，任務為依商品保存期限依序排列。任務結束後，填寫「專業人員科技創新意識問卷」，以瞭解科技創新意識、知覺可用性、知覺易用性、使用態度與行為意圖等構面，同時檢驗測量模式與結構模型之適配性與解釋力。
研究結果顯示，科技創新意識可顯著正向預測知覺有用性與知覺易用性，顯示專業人員對新科技的開放程度越高，越傾向評估該系統具有功能價值且易於操作。在模型路徑中，知覺易用性顯著影響知覺有用性；而知覺有用性與知覺易用性皆對使用態度產生強烈正向影響。最終，使用態度成為行為意圖最主要的預測因子。
本研究之結果支持將科技創新意識納入科技接受模式，並證實知覺可用性與知覺易用性對專業人員接受沉浸式擴增實境具中介效果。未來研究建議可進行縱貫性研究，以探討科技於真實情境之持續應用與訓練成效。

Purpose：A core objective of vocational rehabilitation is to support vocational skill development among individuals with disabilities. Trainees with intellectual and developmental disabilities require highly structured instruction，repeated practice，and low-error learning environments to master job-related competencies. Although traditional training approaches，such as direct demonstrations and authentic workplace exposure，remain valuable，they provide limited opportunities for safe repetition and place considerable demands on professionals to deliver individualized instruction. Augmented reality（AR）has emerged as a promising instructional solution to these challenges. By overlaying digital information onto physical environments，immersive AR（iAR）provides real-time visual cues，step-by-step guidance，and authentic simulations that mirror real-world workplace demands，thereby enhancing task comprehension and reducing training risks. However，the successful implementation of AR relies heavily on vocational rehabilitation and special education professionals，who serve as the primary decision-makers and“gatekeepers”in determining whether emerging tools are adopted and sustained within instructional routines. The technology acceptance model（TAM）has been extensively applied to explain how core cognitive evaluations of perceived usefulness（PU）and perceived ease of use（PEOU）influence technology adoption. However，these constructs are insufficient to explain technology adoption in specialized educational settings. Accordingly，scholars have called for further research into technological innovativeness—defined as an individual’s intrinsic tendency and openness toward exploring new technologies—to better explain the intention to adopt emerging technologies. Despite the theoretical implications of technological innovativeness，its role in the adoption of iAR within vocational rehabilitation has received limited attention. To bridge this gap，the present study extended the TAM framework to examine professionals’acceptance of an iAR system for vocational skills training（iARVST）. Technological innovativeness was conceptualized as a critical external antecedent，and its structural relationships with PU，PEOU，attitude toward technology use（ATT），and behavioral intention（BI）were investigated. Methods：A primarily quantitative research design was adopted. Partial least squares structural equation modeling（PLS-SEM）was supplemented with qualitative insights gathered from semistructured interviews to characterize technology adoption contexts. PLS-SEM was selected for its suitability for exploratory studies with limited sample sizes. In total，71 professionals actively engaged in vocational training for individuals with intellectual disabilities were recruited through convenience sampling. The cohort comprised 21 male and 50 female participants evenly distributed between vocational rehabilitation and school-based settings. Notably，60.6% of the participants reported no prior experience with AR technology. All participants engaged in a standardized experimental procedure. After providing informed consent and watching a brief video tutorial，they wore the Microsoft HoloLens 2 headset to perform a product expiration-date sequencing task within the bespoke iARVST system. The task simulated a common responsibility in convenience store jobs；that is，the participants were required to chronologically arrange physical cartons on the basis of virtual expiration-date labels presented in AR. The task lasted for a maximum of 8 minutes or until four difficulty levels were completed. Immediately after the task，the participants completed a Technological Innovativeness in Professions Questionnaire（TIPQ）. The TIPQ was rigorously developed and validated through expert reviews and a pilot study（N = 44），demonstrating high composite reliability（.92–.94）. The final 30-item instrument measured five constructs：technological innovativeness，PU，PEOU，ATT，and BI. Subsequently，a purposively selected subset of 12 participants completed a 15–20-minute semistructured interview exploring perceived usability，practical constraints，and pedagogical attitudes. SmartPLS 4.0 was used for data analysis using 500 bootstrap resamples. Sample size adequacy was confirmed through the 10 times rule，which indicated a theoretical minimum sample size of 30；thus，the sample of 71 provided robust power. During the qualitative phase，interview data were summarized and thematically aligned with the extended TAM framework to complement the statistical findings. Results：Mean scores indicated moderately high evaluations of PU（M = 3.97），ATT（M = 3.90），BI（M = 3.80），PEOU（M = 3.61），and technological innovativeness（M = 3.32）. The PLS-SEM path analysis strongly supported the extended TAM model. Technological innovativeness significantly and positively predicted both PU（β= .24，p < .05）and PEOU（β= .36，p < .001），indicating that intrinsically open professionals perceived the AR system as more useful and easier to operate. However，technological innovativeness did not exert a significant direct effect on ATT（β= .12，p = .09）. Regarding core TAM relationships，PEOU significantly predicted PU（β= .58，p < .001）. Both PU（β= .54，p < .001）and PEOU（β= .42，p < .001）strongly influenced ATT. ATT also emerged as the strongest predictor of BI（β= .83，p < .001）. The structural model demonstrated robust explanatory power，accounting for 50%，13%，77%，and 69% of the variance in PU，PEOU，ATT，and BI，respectively. Indirect effect analyses confirmed that technological innovativeness influenced BI primarily through sequential cognitive evaluations rather than direct affective responses. Qualitative analyses were then conducted to corroborate these results. Thematic summaries indicated that although technological innovativeness stimulated initial interest，professionals ultimately acted as gatekeepers constrained by workplace realities. High PU was attributed to the unique instructional affordances of the iARVST，such as visual sequence scaffolding，which alleviated one-on-one coaching burdens. Additionally，PEOU was critical to minimize extraneous cognitive load for both instructors and learners. The lack of a direct relationship between innovativeness and ATT was explained by practical concerns：the prohibitive cost and fragility of head-mounted AR hardware may inadvertently induce psychological distress among trainees，demonstrating that personal openness cannot override pragmatic usability. Conclusions：This study successfully extended the TAM to technological innovativeness，demonstrating that personal traits are major antecedents of technology adoption in special education. Professionals’acceptance of iAR is driven by technological affinity and mediated by their evaluations of pedagogical efficacy and operational intuitiveness. The iARVST was recognized as a valuable instructional tool that streamlines training workflows，offers automated performance logging，and provides a safe environment for individuals with intellectual disabilities to repeatedly practice work-readiness skills. These results have several practical implications for institutions and policymakers. Because technological innovativeness influences early acceptance，institutions must proactively cultivate an organizational culture that encourages innovation. Managers should implement pilot programs and hands-on workshops，allowing professionals to engage with emerging technologies in risk-free environments to reduce anxiety. Furthermore，to address hardware and financial barriers，future technological development should consider cross-platform compatibility. Core training functionalities can be migrated to more ubiquitous and financially sustainable nonimmersive platforms，such as tablets or smartphones. This study is limited by its reliance on convenience sampling and the short-term nature of the experimental task. Nevertheless，the findings provide a robust foundation for future research. Subsequent longitudinal studies should incorporate objective behavioral data and perform comparative experiments across AR platforms to explore sustained integration in authentic settings，contributing to inclusive employment preparation.