{
  "metadata": {
    "title": "Use of artificial-intelligence tools and learning motivation among undergraduate students: a quantitative survey study",
    "work_type": "final_project",
    "research_type": "empirical_study",
    "language": "english",
    "direction": "ltr",
    "duration_minutes": 10,
    "target_slide_count": 12,
    "paper_summary": "This quantitative cross-sectional survey study examined whether frequency of AI-tool use predicts intrinsic motivation among 200 undergraduates, and whether year of study moderates that relationship. Results confirmed a significant positive association (r = .34) and showed the effect was significantly stronger for first-year students than for advanced-year students, supporting a motivational scaffolding account grounded in Self-Determination Theory."
  },
  "config": {
    "duration_minutes": 10,
    "target_slide_count": 12,
    "palette_id": "academic_blue",
    "palette": {
      "primary": "#1E40AF",
      "secondary": "#DBEAFE",
      "accent": "#60A5FA",
      "background": "#FFFFFF",
      "text": "#0F172A",
      "title_background": "#1E3A8A"
    },
    "font": "Arial",
    "use_icons": false,
    "direction": "ltr",
    "language": "english",
    "logo_storage_path": null
  },
  "slides": [
    {
      "index": 1,
      "layout": "title",
      "section": "title",
      "title": "AI Tool Use and Learning Motivation in Undergraduates",
      "subtitle": "A Quantitative Survey Study",
      "bullets": [],
      "visual": {
        "type": "none",
        "items": [],
        "caption": null,
        "description": null
      },
      "speaker_notes": "Welcome the audience and introduce the study. Frame the talk as a 10-minute overview of a quantitative investigation into whether how often students use AI tools for learning predicts their intrinsic motivation, and whether that relationship differs by year of study.",
      "estimated_duration_seconds": 30
    },
    {
      "index": 2,
      "layout": "icon_rows",
      "section": "introduction",
      "title": "AI in Higher Education: The Motivation Gap",
      "subtitle": null,
      "bullets": [
        "AI tools are now structurally embedded in undergraduate learning.",
        "Research tracks what students do with AI, not what AI does to…",
        "Intrinsic motivation drives depth, persistence, and genuine engagement.",
        "No study has tested AI-use frequency as a predictor of intrinsic motivation.",
        "Year of study as a moderator is entirely unexplored."
      ],
      "visual": {
        "type": "diagram_idea",
        "items": [
          {
            "label": "AI Tool Adoption"
          },
          {
            "label": "Motivational Outcomes"
          },
          {
            "label": "Research Gap"
          }
        ],
        "caption": "The field has moved faster on adoption than on motivation",
        "description": null
      },
      "speaker_notes": "Establish the problem: the literature on AI in education has focused on perceptions, task performance, and adoption — not on whether regular AI use shapes the motivational orientations that sustain learning. This gap is the paper's entry point.",
      "estimated_duration_seconds": 55
    },
    {
      "index": 3,
      "layout": "icon_rows",
      "section": "research_question",
      "title": "Research Questions and Hypotheses",
      "subtitle": null,
      "bullets": [
        "RQ1: Does AI-tool use frequency predict intrinsic motivation?",
        "RQ2: Does year of study moderate that relationship?",
        "H1: More frequent AI use → higher intrinsic motivation.",
        "H2: Association is stronger for first-year than advanced students.",
        "Grounded in Self-Determination Theory (Deci & Ryan, 1985)."
      ],
      "visual": {
        "type": "diagram_idea",
        "items": [
          {
            "label": "AI-Tool Use Frequency"
          },
          {
            "label": "Intrinsic Motivation"
          },
          {
            "label": "Moderated by Year of Study"
          }
        ],
        "caption": "Conceptual model: direct effect moderated by academic stage",
        "description": null
      },
      "speaker_notes": "State the two hypotheses clearly. H1 is a bivariate directional prediction; H2 adds a moderation layer. The SDT grounding is important: autonomy, competence, and relatedness are the needs whose satisfaction sustains intrinsic motivation, and AI scaffolding may support all three — especially for novice students.",
      "estimated_duration_seconds": 55
    },
    {
      "index": 4,
      "layout": "icon_rows",
      "section": "methodology",
      "title": "Design, Sample, and Measures",
      "subtitle": null,
      "bullets": [
        "Cross-sectional online survey; N = 200 undergraduates.",
        "Power analysis: α = .05, power = .80, medium interaction effect.",
        "AI-tool use frequency: researcher-developed 5-point Likert scale (α = .76).",
        "Intrinsic motivation: Academic Motivation Scale composite (α = .81).",
        "Controls: gender, GPA; moderator: year of study (first-year vs. advanced)."
      ],
      "visual": {
        "type": "icon_grid",
        "items": [
          {
            "icon": "users",
            "label": "200 Undergraduates",
            "description": "Convenience sample; 44.5% first-year"
          },
          {
            "icon": "clipboard-list",
            "label": "AMS + AI-Use Scale",
            "description": "Validated + researcher-developed items"
          },
          {
            "icon": "bar-chart-2",
            "label": "PROCESS Moderation",
            "description": "Hayes Model 1; Pearson r for H1"
          }
        ],
        "caption": null,
        "description": null
      },
      "speaker_notes": "Summarise the design in one breath: cross-sectional survey, 200 students, two validated composite scales, and a Hayes PROCESS moderation model. Mention that the sample was sized by power analysis to detect a medium interaction effect — a methodological strength worth flagging.",
      "estimated_duration_seconds": 65
    },
    {
      "index": 5,
      "layout": "two_column",
      "section": "methodology",
      "title": "Sample Characteristics",
      "subtitle": null,
      "bullets": [
        "59% women, 37% men, 4% non-binary / undisclosed.",
        "44.5% first-year; 55.5% advanced (2nd year and above).",
        "Mean GPA = 82.4 (SD = 8.3) on institutional scale.",
        "Mean AI-use frequency = 3.42 / 5; mean intrinsic motivation = 3.71…"
      ],
      "visual": {
        "type": "pie_chart",
        "items": [
          {
            "category": "Social sciences & humanities",
            "value": 68
          },
          {
            "category": "Natural sciences & engineering",
            "value": 54
          },
          {
            "category": "Education & teaching",
            "value": 46
          },
          {
            "category": "Other fields",
            "value": 32
          }
        ],
        "caption": "Sample by disciplinary cluster (N = 200)",
        "description": null
      },
      "speaker_notes": "Briefly orient the audience to who the participants were. The disciplinary spread across four clusters reduces the risk that findings reflect a single-field effect. The moderately high mean GPA is worth noting as a potential ceiling consideration for motivation scores.",
      "estimated_duration_seconds": 55
    },
    {
      "index": 6,
      "layout": "stat_callout",
      "section": "results",
      "title": "H1 Supported: Positive Association Confirmed",
      "subtitle": null,
      "bullets": [],
      "visual": {
        "type": "stat",
        "items": [
          {
            "value": "r = .34, p < .001",
            "caption": "Medium positive correlation between AI-tool use frequency and intrinsic motivation (N = 200; 95% CI [.21, .46])"
          }
        ],
        "caption": null,
        "description": null
      },
      "speaker_notes": "This is the headline result for H1. A Pearson r of .34 falls squarely in the medium range by Cohen's benchmarks. The confidence interval does not include zero, and the effect holds across the full sample after normality checks confirmed parametric assumptions were met.",
      "estimated_duration_seconds": 50
    },
    {
      "index": 7,
      "layout": "two_column",
      "section": "results",
      "title": "Intrinsic Motivation Rises with AI-Use Frequency",
      "subtitle": null,
      "bullets": [
        "Higher AI-use bands correspond to higher mean intrinsic motivation.",
        "Trend is monotonic across all five frequency bands.",
        "No evidence of a non-linear ceiling or floor effect."
      ],
      "visual": {
        "type": "bar_chart",
        "items": [
          {
            "category": "1.0–1.8",
            "value": 2.81
          },
          {
            "category": "1.8–2.6",
            "value": 3.12
          },
          {
            "category": "2.6–3.4",
            "value": 3.48
          },
          {
            "category": "3.4–4.2",
            "value": 3.79
          },
          {
            "category": "4.2–5.0",
            "value": 4.05
          }
        ],
        "caption": "Mean intrinsic motivation by AI-tool use frequency band",
        "description": null
      },
      "speaker_notes": "Walk the audience through the bar chart. Each bar represents mean intrinsic motivation for students within that AI-use frequency band, as reported in the paper's Figure 1 data. The monotonic rise across bands visually reinforces the Pearson r finding and shows the relationship is not driven by outliers at either extreme.",
      "estimated_duration_seconds": 60
    },
    {
      "index": 8,
      "layout": "stat_callout",
      "section": "results",
      "title": "H2 Supported: Year of Study Moderates the Effect",
      "subtitle": null,
      "bullets": [],
      "visual": {
        "type": "stat",
        "items": [
          {
            "value": "ΔR² = .043, p = .015",
            "caption": "Interaction term (AI-use × year of study) in PROCESS Model 1; overall model R² = .189, F(4, 195) = 11.37, p < .001"
          }
        ],
        "caption": null,
        "description": null
      },
      "speaker_notes": "The interaction term is statistically significant and accounts for an additional 4.3% of variance in intrinsic motivation beyond the main effects. The overall model explains nearly 19% of variance — a meaningful result for a motivational outcome measured by self-report in a cross-sectional design.",
      "estimated_duration_seconds": 50
    },
    {
      "index": 9,
      "layout": "icon_rows",
      "section": "results",
      "title": "First-Year Students: Steeper Motivational Slope",
      "subtitle": null,
      "bullets": [
        "First-year slope: B = 0.41, p < .001 — strong and significant.",
        "Advanced-year slope: B = 0.14, p = .082 — smaller, marginal.",
        "Gap widens as AI-use frequency increases.",
        "Scaffolding benefit concentrated at academic entry point."
      ],
      "visual": {
        "type": "none",
        "items": [],
        "caption": null,
        "description": null
      },
      "speaker_notes": "The simple slopes tell the core story of H2. For first-year students, each unit increase in AI-use frequency is associated with a 0.41-point rise in intrinsic motivation — a practically meaningful gain. For advanced students the slope is less than a third of that size and does not reach conventional significance.",
      "estimated_duration_seconds": 70
    },
    {
      "index": 10,
      "layout": "icon_rows",
      "section": "discussion",
      "title": "Interpretation: Scaffolding Where It Matters Most",
      "subtitle": null,
      "bullets": [
        "AI tools reduce affective friction at the hardest transition point.",
        "Competence support (SDT) is most valuable before habits are formed.",
        "Advanced students have internalised strategies — scaffolding adds less.",
        "Intrinsic, not extrinsic, subscales drove the effect — theoretically key.",
        "Causal direction cannot be established from cross-sectional data."
      ],
      "visual": {
        "type": "diagram_idea",
        "items": [
          {
            "label": "AI Scaffolding"
          },
          {
            "label": "Perceived Competence"
          },
          {
            "label": "Intrinsic Motivation"
          },
          {
            "label": "Strongest in Year 1"
          }
        ],
        "caption": "Theorised mechanism (mediation not directly tested)",
        "description": null
      },
      "speaker_notes": "Interpret the pattern through the SDT scaffolding lens: AI tools lower the threshold for experiencing competence, which is the psychological need most acutely unmet in first-year students. Stress the cross-sectional caveat — motivated students may also be more likely to adopt AI tools, so the arrow could run both ways.",
      "estimated_duration_seconds": 70
    },
    {
      "index": 11,
      "layout": "icon_rows",
      "section": "conclusion",
      "title": "Limitations and Future Directions",
      "subtitle": null,
      "bullets": [
        "Cross-sectional design: association, not causation.",
        "Convenience sample may over-represent tech-engaged students.",
        "Self-reported AI use — no objective usage logs available.",
        "Future work: longitudinal design, mediation via self-efficacy.",
        "Replicate across disciplines and institutional contexts."
      ],
      "visual": {
        "type": "diagram_idea",
        "items": [
          {
            "label": "Cross-sectional → Longitudinal"
          },
          {
            "label": "Association → Mediation"
          },
          {
            "label": "Single context → Multi-institutional"
          }
        ],
        "caption": "Three priority extensions for future research",
        "description": null
      },
      "speaker_notes": "Be direct about the limits: the design cannot rule out reverse causation, the sample is self-selected, and self-report measures carry social-desirability risk. Frame the limitations constructively by mapping each onto a concrete next-step design — longitudinal tracking, mediation models, and cross-institutional replication.",
      "estimated_duration_seconds": 60
    },
    {
      "index": 12,
      "layout": "icon_rows",
      "section": "conclusion",
      "title": "Key Takeaways",
      "subtitle": null,
      "bullets": [
        "AI-tool use frequency positively predicts intrinsic motivation (r = .34).",
        "Year of study moderates the effect — first-year students benefit most.",
        "Motivational gains are intrinsic, not merely extrinsic or grade-driven.",
        "Integrate AI tools intentionally in first-year curricula for greatest impact.",
        "Longitudinal and mediation designs are the critical next steps."
      ],
      "visual": {
        "type": "icon",
        "items": [
          {
            "icon": "trending-up"
          }
        ],
        "caption": null,
        "description": null
      },
      "speaker_notes": "Close by restating the three substantive conclusions: a meaningful positive association exists, it is concentrated in first-year students, and it operates through intrinsic rather than extrinsic motivational channels. Leave the audience with the practical implication: targeted AI integration at the point of entry into higher education is where the motivational payoff is greatest.",
      "estimated_duration_seconds": 55
    }
  ]
}