From the Student Success edition of September 17, 2026
Cogniti GenAI academic-agent pilot: perceived learning/success up; engagement and wellbeing flat
Flinders University / Technology, Knowledge and Learning (Springer) · Higher education teaching and learning / assessment support · Australia (Flinders University; undergraduate courses)
- Publisher
- Technology, Knowledge and Learning (Springer Open)
- Original publication
- Accepted 17 December 2025; © The Author(s) 2026 (online OA)
- Source retrieved
- 2026-09-18
What happened
Pilot study of course-embedded Cogniti GenAI academic agents for assessment preparation across four undergraduate courses (education, psychology, social work, sport). N=139 survey completers (ages 18–55) pre/post agent deployment. Mann–Whitney U tests: significant improvements in perceived student learning and perceived success (both p=0.04; w≈1642 and 1639.5), with no significant change in engagement (p=0.30) or wellbeing (p=0.59). Qualitative themes valued 24/7 context-specific support and confidence; students also stressed retaining human academic interaction. Authors conclude agents may enhance targeted aspects of experience when carefully integrated to complement—not replace—human-centred learning.
Why it matters
Newly relevant after 2026-09-16 archived hybrid/reflective feedback and RAG-tutor nulls: this unarchived 2026 OA pilot supplies HE evidence on personalized academic agents’ effects on student-success perceptions versus engagement/wellbeing. Why now: fills agent-product gap for advising/tutoring-adjacent on-demand support without overclaiming objective grades.
Evidence and measured results
Source findings: pre/post perception surveys on one assessment; Mann–Whitney and rank-biserial effect sizes; content analysis of open responses. Distinguishes perception gains (learning/success) from null engagement/wellbeing. Limitations: pilot/non-randomized; self-report only; single-assessment window; Australian one-university sample; response rates vary by course (e.g., psychology 9.6% of enrolled).
Limitations and uncertainty
Perception outcomes ≠ measured GPA/retention. Non-experimental design cannot support causal claims. Not a U.S. community-college advising RCT. Publication accepted late 2025 / ©2026; included as newly archived for this stream.
Put this evidence to work
Lighthouse Advisory interpretation, grounded in this source. Enriched 2026-09-18; this does not change the original publication date. Each role follows the same checklist so takeaways compare across sources.
Sales
Role takeaway
- Customer problem
- Campuses want always-on academic agents but lack evidence separating perception gains from engagement/wellbeing and objective learning.
- Stakeholders
- Teaching & learning center
- Faculty course leads
- Student success / retention
- Academic technology
- Discovery questions
- Will you measure objective marks as well as perceptions?
- Is human escalation required in the workflow?
- Which courses own agent knowledge updates?
- What integrity rules apply to agent use on assessments?
- Value hypothesis
- Course-embedded GenAI agents can improve perceived learning/success for assessment prep when faculty-scoped and paired with humans; engagement/wellbeing gains are not evidenced here.
- Potential engagement
- Bounded multi-course pilot with pre/post surveys plus grade checks and explicit no-replacement staffing commitment.
- Unsupported claims
- Guaranteed retention or wellness improvements
- That perception gains equal learning
- Staff-cost savings from replacing tutors/advisors
Pre-sales engineering
Role takeaway
- Fit
- Fits LMS-embedded, course-specific academic agents for assessment clarification—not enterprise early-alert alone.
- Architecture and integration
- Per-course agent configs + retrieval over approved materials + LMS entry points + logging; human ticket/escalation channel.
- Prerequisites
- Faculty-authored briefs/rubrics
- Campus-approved GenAI endpoint
- Survey/analytics pipeline
- Access control by course roster
- Constraints
- Non-random pilot bias
- Variable course response rates
- Data residency for student prompts
- Security
- Prompts may include personal academic struggles; minimize retention and block consumer tools for roster data.
- Proof of value
- Replicate learning/success perception contrasts
- Add objective assessment outcomes
- Track escalation rates to humans
Delivery
Role takeaway
- Work
- Faculty prompt design, student communications, integrity guidance, and helpdesk escalation playbooks.
- Dependencies
- Course coordinator time
- LMS admin
- Integrity office language
- Support staffing
- Ownership
- Center for teaching and learning with participating departments.
- Skills and adoption
- Faculty need agent-authoring skills; students need disclosure/how-to-use guidance.
- Governance checkpoints
- Pedagogy approval of agent scope
- Privacy review of prompts
- Integrity syllabus statements
- Acceptance criteria
- Agents live in ≥2 pilot courses
- Pre/post instruments completed
- Human escalation path documented
- Risks
- Overclaiming from self-report
- Uneven adoption across disciplines
- Students substituting agents for staff
Implementation considerations
Lighthouse Advisory interpretation across the operating dimensions a public-sector buyer must settle before this evidence becomes a design. Each note answers the question under its heading for this specific source.
Architecture and integration
What must connect, and where does the AI sit in the workflow?
Prefer course-scoped agent configs grounded in assessment briefs over generic chat; instrument both perception and objective performance; keep human escalation paths. Integration: LMS/course sites with agent endpoints and usage/outcome logging.
Governance
Who approves, reviews and stays accountable for outcomes?
Approve agents as supplemental academic support with disclosure, academic-integrity boundaries, and faculty ownership of prompts/knowledge—not unmanaged consumer bots.
Security and privacy
What data, permissions and controls need testing?
Source emphasizes institutional integration; does not detail FERPA-equivalent flows. Interpretation: treat prompts/uploads as education records; require vetted campus tools, retention limits, and no-train vendor terms.
Accessibility and workforce
Who is affected, and what skills or accommodations follow?
24/7 agents may help students balancing work/caregiving, but flat wellbeing and student insistence on human contact caution against using agents as disability or counseling substitutes; co-design with accessibility offices.
Procurement
What should contracts, pricing and exit terms secure?
Procure course-configurable agents with faculty authoring rights and evaluation clauses—not seat licenses sold on engagement/wellbeing claims unsupported here.
Operating model
Which teams own the service once it runs?
Operating model keeps faculty as knowledge owners and agents as supplemental capacity; staffing models should reserve human contact for complex cases.
Publication history
- 2026-09-17Student Success · Issue 125 resources
Stable resource ID: elliott-cogniti-genai-agents-pilot-2026