CoChat
Sign Up
Cited by 298Intelligent Tutoring Systems and Adaptive LearningInnovative Teaching and Learning MethodsVisual and Cognitive Learning Processes

Effective Tutoring Techniques: A Comparison of Human Tutors and Intelligent Tutoring Systems

Merrill, Douglas C., Reiser, Brian J., Ranney, Michael, Trafton, J. Gregory·1992·Journal of the Learning Sciences
DOI 10.1207/s15327809jls0203_2
Open this paper in CoChat — free account

Abstract

There has been much debate about instructional strategies for computerized learning environments. Many of the arguments designed to choose between the various philosophies have appealed, at least implicitly, to the behavior of effective human teachers. In this article, we compare the guidance and support offered by human tutors with that offered by intelligent tutoring systems. First, we review research on human tutoring strategies in various domains. Then we investigate the capabilities of a widely used technique for providing feedback, model tracing. Finally, we contrast the types of guidance and support provided by human tutors with those in intelligent tutoring systems, by examining the process of recovering from impasses encountered during problem solving. In general, the support offered by human tutors is more flexible and more subtle than that offered by model tracing tutors, but the two are more similar than sometimes argued.

Integrity signals

Registry

Crossref · 10.1207/s15327809jls0203_2

Cites

9 of 9 references verified0 citation errors · 0 not checkable

Statistics

Recomputing p-values needs the paper’s full text.

Claims the paper makes

This paper’s full text isn’t openly available, so its claims and self-check weren’t run. The registry and citation checks still apply.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Expert tutors base their interventions on observations about the student's emotional or knowledge state and the state of the conversation and their previous interventions.

    “Expert tutors base their interventions on observations about (1) the student's emotional or knowledge state (Lepper et al., 1993), and (2) the state of the conversation and their previous interventions (Merrill et al., 1992).”

    From Towards the Pedagogical Steering of Large Language Models for Tutoring: A Case Study with Modeling Productive Failure (2025) · cites this paper as Merrill et al., 1992

  • It is more efficient from a learning standpoint to promote students' active engagement and let them cognitively engage with the problem, for example by providing an indirect hint or asking a guiding question.

    “Yet, it is more efficient from a learning standpoint to promote students' active engagement and let them cognitively engage with the problem, for example, by providing an indirect hint or asking a guiding question (Freeman et al., 2014; Merrill et al., 1992).”

    From Towards the Pedagogical Steering of Large Language Models for Tutoring: A Case Study with Modeling Productive Failure (2025) · cites this paper as Merrill et al., 1992

  • Existing LLM tutors are optimized and mostly evaluated for single-turn requests.

    “Moreover, existing LLM tutors are optimized and mostly evaluated for single-turn requests, while a tutoring dialog should be envisioned as a multiturn teacher-student interaction where the teacher uses various pedagogical cues (Merrill et al., 1992).”

    From Towards the Pedagogical Steering of Large Language Models for Tutoring: A Case Study with Modeling Productive Failure (2025) · cites this paper as Merrill et al., 1992

  • A tutoring dialog should be envisioned as a multi-turn teacher-student interaction where the teacher uses various pedagogical cues.

    “Moreover, existing LLM tutors are optimized and mostly evaluated for single-turn requests, while a tutoring dialog should be envisioned as a multiturn teacher-student interaction where the teacher uses various pedagogical cues (Merrill et al., 1992).”

    From Towards the Pedagogical Steering of Large Language Models for Tutoring: A Case Study with Modeling Productive Failure (2025) · cites this paper as Merrill et al., 1992

Cites

Flagged first, then by citations

Cited by

Sorted by citations

Authors
Merrill, D. C., Reiser, B. J., Ranney, M., & Trafton, J. G.
Affiliations
Princeton University (US)
ORCID
Venue
Journal of the Learning Sciences
Publisher
Informa UK Limited
Published
1992-07
Volume
2
Issue
3
Pages
277-305
ISSN
  • 1050-8406 (ISSN-L)
  • 1532-7809
DOI
10.1207/s15327809jls0203_2
OpenAlex
W1983352506
Cited by
298
References
9
Citations by year
2026: 3 · 2025: 2 · 2024: 6 · 2023: 1 · 2022: 14 · 2021: 10 · 2020: 12 · 2019: 16 · 2018: 10 · 2017: 7 · 2016: 17 · 2015: 11 · 2014: 9 · 2013: 12 · 2012: 8
FWCI
3.57
Type
journal article
Language
en
Open access
closed
Locations
Journal of the Learning Sciences (publishedVersion)
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › Intelligent Tutoring Systems and Adaptive Learning
  • Social Sciences › Psychology › Developmental and Educational Psychology › Innovative Teaching and Learning Methods
  • Social Sciences › Psychology › Experimental and Cognitive Psychology › Visual and Cognitive Learning Processes
Keywords
intelligent tutoring systems, human tutoring, model tracing, feedback, problem solving, impasse, guidance and support, computerized learning environments, tutoring strategies

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review