Aspiration-Driven Design in EdTech

• Aspiration-driven design is an approach that aligns learners' self-articulated goals with core SDT needs to enhance motivation and agency.
• It employs scaffolded curricula and participatory co-design processes to transition learners from passive users to active creators of educational tools.
• Analytics frameworks like LAI and TES offer quantifiable insights into tool efficacy and learner agency, guiding iterative improvement.

Aspiration-driven design of educational technologies is an approach that structures learning experiences and tool development according to learners’ self-articulated goals and values, formally aligning these aspirations with core psychological needs—such as autonomy, competence, and relatedness—to enhance learner agency, motivation, and sustainable growth. Recent research defines and operationalizes this approach through theoretical frameworks (including Self-Determination Theory and the Aspire to Potentials for Learners model), scaffolded curriculum designs, analytics-driven evaluation, and iterative participatory co-design processes supported by generative AI and learning analytics (Huang et al., 22 Oct 2025, Mao, 29 Apr 2025).

1. Theoretical Underpinnings: Self-Determination Theory and Aspirations

Self-Determination Theory (SDT) posits that intrinsic motivation thrives when three innate needs are satisfied: autonomy (volition and agency), competence (effectiveness and mastery), and relatedness (connection and value within a community). Aspiration-driven educational technologies treat the learner’s aspiration profile $$\textbf{Asp} = (\mathrm{asp}_1, ..., \mathrm{asp}_n)$$ as a central input for design, with aspirations such as developing AI competencies, influencing peers, or achieving creative autonomy explicitly mapped to SDT needs:

• Autonomy: $$f_A(\mathrm{Asp}_{\text{choice}}, \mathrm{Asp}_{\text{ownership}})$$
• Competence: $$f_C(\mathrm{Asp}_{\text{skill}}, \mathrm{Asp}_{\text{mastery}})$$
• Relatedness: $$f_R(\mathrm{Asp}_{\text{collab}}, \mathrm{Asp}_{\text{impact}})$$

A linear aggregation for overall motivation $M$ is formalized as: $$A = \alpha_1 \mathrm{asp}_{\text{choice}} + \alpha_2 \mathrm{asp}_{\text{ownership}};\quad C = \beta_1 \mathrm{asp}_{\text{skill}} + \beta_2 \mathrm{asp}_{\text{mastery}};\quad R = \gamma_1 \mathrm{asp}_{\text{collab}} + \gamma_2 \mathrm{asp}_{\text{impact}}$$

$$M = A + C + R \qquad \alpha_i, \beta_i, \gamma_i > 0$$

This explicit formalization enables direct mapping of aspiration profiles to design levers that satisfy psychological needs, thereby structuring the motivational affordances of educational technologies (Huang et al., 22 Oct 2025).

2. Curriculum and Participatory Co-Design Processes

A two-stage curriculum operationalizes aspiration-driven design by bridging passive consumption of AI tools and participatory creation:

Term 1: Tool Use and Familiarization

• Introduction to research methods and initial engagement with instructor-built GPT-based knowledge tools (Observation Station GPT, Interview Simulation GPT, Design-thinking GPT).
• Scaffolded walkthroughs (prompt formats, live demos).
• Low-stakes use in routine academic tasks.

Term 2: Tool Redesign and Creation

• Reflection on initial experiences to identify strengths, gaps, and pain points.
• Formation of small teams (3-4 students) to select and redesign tools based on collective aspiration statements.
• Workshops in prompt engineering and no-code GPT builder tools.
• Iterative sprints involving prototype development, peer feedback, design refinement, and culminating presentations documenting the workflow and rationales.

This process is designed to increase autonomy (choice in function/design), build competence (prompt engineering, technical workflows), and foster relatedness (peer collaboration, shared outcomes), facilitating a transition in student identity from tool user to tool creator (Huang et al., 22 Oct 2025).

3. Case Implementations: Custom GPT Knowledge Tools

Two student-built tools exemplify how aspiration-driven design materializes:

Tool	Intended Outcomes	Technical Features
Interview Companion GPT	Real-time interview skills, critical analysis	Multi-modal feedback (text analysis, live audio via WebRTC), persona engine, demo-based refinement
Observation Station GPT	Systematic observation, pattern recognition	Framework selector, photo input + object detection (OpenAI API + Google Vision API), 3x3 matrix generator

In both cases, prompt architectures embedded scaffolded examples and role-instructions; technical workflows combined no-code builders with targeted API microservices. Student reflections highlighted increased realism, analytical depth, and a desire for further customization (e.g., expanded language outputs, richer audio personas) (Huang et al., 22 Oct 2025).

4. Analytics, Measurement, and Evaluation Frameworks

Aspiration-driven design leverages both qualitative thematic analysis and quantified evaluation rubrics:

Agency Rubric (1–5 scale for each SDT dimension):

• Autonomy Score ($A_s$): extent of user-defined choices
• Competence Score ($C_s$): self-evaluated expertise in prompt engineering and debugging
• Relatedness Score ($R_s$): quality and frequency of peer engagement

Composite Learner Agency Index (LAI): $$\mathrm{LAI} = \frac{A_s + C_s + R_s}{3}, \qquad 1 \leq \mathrm{LAI} \leq 5$$

Tool Efficacy Rubric (1–5 scale):

• Functional Completeness (learning goals, stated features)
• Usability (prompt and interface clarity)
• Feedback Quality (depth, specificity, authenticity)

Composite Tool Efficacy Score (TES): $$\mathrm{TES} = \frac{\mathrm{FC} + \mathrm{U} + \mathrm{FQ}}{3}, \qquad 1 \leq \mathrm{TES} \leq 5$$

A plausible implication is that such rubrics enable systematic tracking of both process- and product-level affordances, offering quantifiable evidence of the impact of aligning design with aspirations (Huang et al., 22 Oct 2025).

5. Formalization within Learning Analytics: The A2PL Framework

The Aspire to Potentials for Learners (A2PL) model positions learner aspirations as dynamic, vectorized entities embedded within cyclical growth analytics (Mao, 29 Apr 2025). Core constructs include:

• Self-assessment vector: $x^{(t)} \in \mathbb{R}^n$ (skills, strategies, resources)
• Component weights: $w \in \mathbb{R}^n$
• Relational logistics: $R \in \mathbb{R}^{n \times n}$
• Vectorized progression: $v^{(t)} = x^{(t)} - x^{(t-1)}$
• Aspiration alignment score: $$S^{(t)} = w^{T} x^{(t)} + x^{(t)T} R x^{(t)} + \lambda \|v^{(t)}\|$$

$$A^{(t)} = \frac{x^{(t)} \cdot p}{\|x^{(t)}\| \|p\|}$$

Here, $p$ is a prototypical aspiration vector.

This formal structure enables invisible (diagnostic) scoring and interactive (guided) analytics, where generative AI does not prescribe “correct” answers but prompts iterative learner reflection, thereby supporting sustained, aspiration-aligned self-directed growth (Mao, 29 Apr 2025).

6. General Principles and Methodological Guidelines

Research (Huang et al., 22 Oct 2025, Mao, 29 Apr 2025) converges on key design and methodological principles:

1. Surface and Align Aspirations: Begin with explicit goal-mapping exercises and connect stated aspirations to SDT needs.
2. Scaffold Entry and Gradual Release: Use instructor-created prototypes followed by dynamic transfer of design control to learners.
3. Embed Iterative, Reflective Cycles: Require repeated prototype testing, peer review, and reflexive documentation.
4. Leverage No-Code/Low-Code Systems: Minimize technical barriers, maintain focus on aspiration-directed design.
5. Build for Community Impact: Catalog and share learner-built tools for future cohorts to foster sustained relatedness and legacy.
6. Restrict Prescriptiveness: Maintain open-ended reflection and refrain from revealing system rubrics or enforcing a single “correct” solution.
7. Integrate Complex Thinking and Summative Self-Assessment: Use real-world, ill-structured tasks coupled with self-authored rubric frameworks to scaffold multi-level analytic growth (Mao, 29 Apr 2025).

7. Significance and Implications

Empirical and conceptual studies converge on several outcomes:

• Transformation of learner identity from consumer to co-creator and collaborator.
• Heightened critical engagement with generative AI, including awareness of system limitations and the epistemic risks of over-reliance.
• Quantifiable improvements in intrinsic motivation, skill development, and community formation, as indexed by LAI and TES scores.
• Methodologies that foreground aspirations and agency are positioned as central to achieving equitable, adaptive, and sustainable learning systems in AI-rich educational environments (Huang et al., 22 Oct 2025, Mao, 29 Apr 2025).

A plausible implication is that the aspiration-driven approach constitutes a blueprint for educational technologies that not only deliver content, but also activate and sustain deep learner agency, reflective growth, and collaborative innovation.