Did the Course Spark Interest? The Four-Phase Model of Interest Development as an Evaluation Lens

In brief

A course-evaluation item like "I enjoyed this course" captures a diffuse, momentary feeling. Educational-psychology research on interest offers a sharper target. Hidi and Renninger''s (2006) four-phase model of interest development distinguishes triggered situational interest (a momentary spark from something novel or surprising), maintained situational interest (that spark held across the term by meaningful tasks and support), emerging individual interest (the student begins to seek the topic out on their own), and well-developed individual interest (a durable, self-sustaining disposition to re-engage). This matters for evaluation because these phases have different educational value and different causes. A course that reliably triggers interest but never maintains it is a different — and more fixable — problem than one that never sparks anything. Measuring "did the course move students along the interest pathway?" is more diagnostic, and more predictive of later outcomes, than measuring satisfaction.

What the research says

Suzanne Hidi and K. Ann Renninger''s "The Four-Phase Model of Interest Development" (Educational Psychologist, 2006) synthesised decades of interest research into a developmental sequence. Their central claims are that interest is both affective and cognitive — it involves positive feeling and stored value and knowledge — and that it develops through phases that can deepen or fall away depending on environmental support. The four phases are:

1. Triggered situational interest — a short-term psychological state sparked by conditions in the environment: novelty, surprise, hands-on tasks, a compelling problem, personal relevance.
2. Maintained situational interest — the state persists and is sustained by the meaningfulness of tasks and by involvement over time, not just a one-off hook.
3. Emerging individual interest — a beginning predisposition to re-engage with the content voluntarily; the learner starts generating their own questions.
4. Well-developed individual interest — a relatively enduring predisposition to seek out and re-engage with the topic over time, with accumulated knowledge and value.

The pedagogically crucial point is that situational interest can be deliberately triggered and then either supported into individual interest or allowed to lapse — the transition is not automatic and depends on the learning environment. This is what makes interest an evaluable property of a course rather than a fixed trait of the student.

Corroborating and extending evidence comes from Rotgans and Schmidt (2011), who measured situational interest repeatedly across a single active-learning session and found it rose sharply when a problem was introduced, decayed, and rose again — and, importantly, that situational interest predicted achievement-related classroom behaviours, which in turn predicted academic achievement. Their "micro-analytical" approach — short interest measures administered repeatedly at critical moments — also showed that aggregating interest into a single end-of-session number produced worse prediction than capturing it in the moment. That is a direct methodological warning for end-of-term evaluation: a single retrospective "was it interesting?" rating discards the dynamics that actually matter.

The deeper literature ties interest to consequential outcomes. Interest is associated with attention, persistence, deeper processing, and — through those channels — learning and later course-taking and career choices. Schiefele''s work on interest and text learning, and Harackiewicz and colleagues'' longitudinal studies linking situational interest to sustained individual interest and subsequent choices, establish that interest is not a "nice to have" affective by-product but a predictor of the outcomes universities claim to care about.

Why it matters for course evaluation in practice

Reframing evaluation around interest development changes both what you ask and what a result means.

• It separates "enjoyed" from "want to continue." A stand-up-comedian lecturer can generate high enjoyment and high satisfaction scores while producing zero individual interest — no student leaves wanting to read more. Conversely, a demanding course can be reported as only moderately "enjoyable" yet move many students into emerging individual interest. Satisfaction and interest development can diverge, and the second is closer to the university''s mission. This connects to the "desirable difficulties" paradox, where the teaching that most improves learning can lower momentary satisfaction.
• It localises the failure. If a course triggers interest (students report the opening problem was gripping) but does not maintain it (the middle weeks flatten out), that is specific, actionable feedback about course design — very different from a low global rating that tells a designer nothing about where the course lost people.
• It predicts downstream behaviour — elective uptake, major choice, continued study — that programme directors and accreditation frameworks increasingly want evidence for. "Did this course move students toward the discipline?" is a programme-level QA question that a satisfaction mean cannot answer.
• It rewards the right teaching. Anchoring evaluation on interest development credits instructors who take pedagogical risks to spark and sustain curiosity, rather than those who simply maximise comfort — partially counteracting the incentives that a satisfaction metric creates (see our discussion of Campbell''s Law).

Limitations and honest caveats

Several caveats keep this honest.

First, self-reported interest is still self-report, subject to social desirability, mood, and the peak–end and recency effects that distort any end-of-term rating. A student who found the last two weeks dull may under-report interest that was genuinely high mid-term — precisely the aggregation problem Rotgans and Schmidt identified.

Second, measuring individual (durable) interest at the end of a course is premature. By definition, well-developed individual interest reveals itself later, through voluntary re-engagement. An end-of-term instrument can credibly capture triggered and maintained situational interest and emerging individual interest, but claims about durable interest need longitudinal follow-up (did students take the next elective, join the reading group, choose the major?). Overclaiming durable interest from a single survey wave would be a validity error.

Third, interest is not the only goal. Some essential courses are unlikely to be intrinsically fascinating for most students; a required statistics or research-ethics module may legitimately aim for competence and compliance more than passion. Interest should be one lens among several, not a universal yardstick, or it risks penalising necessary-but-unglamorous teaching.

Fourth, attribution is hard. Interest is co-determined by the student''s prior individual interest, the topic, and the teaching. A cohort of already-committed majors will report high interest regardless of teaching quality — the "prior subject interest" confound documented in the SET literature. Interest measures therefore need the same contextual and baseline care as any other evaluation signal.

Fifth, the four-phase model is a framework, not a validated scale. Operationalising the phases requires careful item design, and conflating the phases (treating a momentary spark as evidence of durable interest) would misrepresent the construct.

How Koji incorporates this

Koji is well suited to measuring interest as a developmental property rather than a single satisfaction number, because its instrument is conversational and can be deployed across the term rather than only at the end.

• Phase-aware question design. Koji supports structured question types (open_ended, scale, single_choice, ranking) that can be written to distinguish the phases — separating "what in this course grabbed your attention?" (triggered), "what kept you engaged across the term?" (maintained), and "has this course made you want to learn more on your own?" (emerging individual interest) — rather than collapsing them into one "interesting?" item.
• In-semester and micro-analytical collection. Because Rotgans and Schmidt showed that aggregating interest destroys predictive signal, Koji''s support for mid-cycle and experience-sampling collection lets institutions capture interest while it is happening — after a key project, mid-module — instead of relying on a single retrospective rating vulnerable to peak–end distortion.
• Conversational probing of the trigger and the lapse. Koji''s AI-moderated conversational interviews probe beyond the Likert number: when a student says interest faded, the moderator can ask when and why, surfacing whether the course triggered but failed to maintain — the specific, fixable diagnosis the four-phase model points to. A number cannot locate where a course lost its audience; a probed conversation can.
• Thematic analysis mapped to the pathway. Koji''s automatic thematic analysis of open text can organise students'' own accounts of what sparked and sustained their curiosity, giving course designers evidence about which elements did the developmental work.
• Triangulation and longitudinal tracking. Because durable individual interest reveals itself later, Koji''s ability to run linked studies across cohorts and over time supports the honest, longitudinal evidence the construct demands — connecting mid-term interest to later elective uptake — rather than overclaiming durable interest from one wave.

The framing is deliberately modest: Koji is designed to measure movement along the interest pathway and to locate where a course sparks or loses students; it does not claim that a survey alone proves durable interest, which only later behaviour can confirm. Koji''s core research platform at koji.so applies the same conversational engine to product research, where the parallel distinction — a feature that sparks momentary curiosity versus one that earns durable, self-initiated use — is just as decision-relevant.

Related resources

• Self-determination theory as an evaluation lens
• Achievement emotions and control-value theory
• Experience sampling for in-semester feedback
• Prior subject interest as a bias in evaluations
• Sense of belonging as an evaluation construct
• Self-regulated learning and metacognition as an evaluation lens

References

• Hidi, S., & Renninger, K. A. (2006). The four-phase model of interest development. Educational Psychologist, 41(2), 111–127. https://doi.org/10.1207/s15326985ep4102_4
• Rotgans, J. I., & Schmidt, H. G. (2011). Situational interest and academic achievement in the active-learning classroom. Learning and Instruction, 21(1), 58–67. https://doi.org/10.1016/j.learninstruc.2010.07.001
• Renninger, K. A., & Hidi, S. (2016). The Power of Interest for Motivation and Engagement. Routledge. https://doi.org/10.4324/9781315771045
• Schiefele, U. (1991). Interest, learning, and motivation. Educational Psychologist, 26(3–4), 299–323. https://doi.org/10.1080/00461520.1991.9653136