CoChat

Investigating the effectiveness of mental workload as a predictor of opportune moments for interruption

Iqbal, Shamsi T., Bailey, Brian P. · 2005 · CHI '05 Extended Abstracts on Human Factors in Computing Systems

The DOI matches its Crossref record; 19 of the 19 references it cites were found in a registry or an index; none of the 17 sources its claims cite is checked for support yet.

Abstract

This work investigates the use of workload-aligned task models for predicting opportune moments for interruption. From models for several tasks, we selected boundaries with the lowest (Best) and highest (Worst) mental workload. We compared effects of interrupting primary tasks at these and Random moments on resumption lag, annoyance, and social attribution. Results show that interrupting at the Best moments consistently caused less resumption lag and annoyance, and fostered more social attribution. Results demonstrate that use of workload-aligned models offers a systematic method for predicting opportune moments.

Overview

CoChat Summary

## TL;DR

The paper tests whether workload-aligned task models can predict better moments to interrupt users, using low-workload (“Best”) and high-workload (“Worst”) task boundaries versus Random and no-interruption conditions. Interrupting at Best moments consistently reduced resumption lag and annoyance and increased respect/social attribution, with only a few seconds’ timing difference from other moments.

## Key findings

  • In a repeated-measures study with 12 users, interrupting at Best moments produced less disruption than Worst or Random moments across tasks.
  • Resumption lag was about 3× faster at Best moments (µ=2.06s) than at Worst (µ=6.69s) or Random (µ=5.66s) moments.
  • Annoyance was lower at Best moments (µ=1.69) than at Worst (µ=2.07) and Random (µ=2.35) moments; no interruption had the lowest annoyance (µ=1.26).
  • Respect/social attribution was higher at Best moments (µ=3.02) than at Worst (µ=1.85) and Random (µ=2.17); no interruption was rated most respectful (µ=4.02).

## Methods

  • Design: Repeated-measures within-subjects experiment with factors Timing (Best, Worst, Random, None) and Task (Route Planning, Document Editing, Object Manipulation).
  • Sample/data: Twelve users (6 female), ages 21 to 42, completed four trials per task category; they also completed an interrupting task (reading a news article and choosing the best title from three options).
  • Analysis: Mental workload was inferred from pupil size in prior workload-aligned task models; effects were analyzed with two-way ANOVAs (Task × Timing) on subjective workload, resumption lag, annoyance, and respect.

## Limitations

  • The paper explicitly notes a practical limitation: it requires workload-aligned task models, which must be developed or already available, though the authors say they only need to be developed once.
  • The authors state future work is needed to replicate the findings with a more diverse set of interactive tasks and to compare against alternative interruption-prediction methods.
  • Evident from the text: the study used only 12 participants and only three primary tasks, so generalizability is limited.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1145/1056808.1056948

Cites

19 of 19 references verified0 citation errors · 0 not checkable

Statistics

Verification

Claims the paper makes

0 Valid · 0 Not valid · 5 Not checked · 15 Own results and interpretations

2 more claims

Create a free account to see all 20 claims and to verify the 1 reference still open.

See all claims — free account
  • It may state two separate claims as one. 7
  • Probably not a substantial claim. 6

Create a free account to see these claims.

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?

Statistics recomputed

Every reported test result, recomputed from its test statistic and degrees of freedom. A result shows as not adding up only once an independent check has confirmed it.

  • Not confirmedF(2, 22) = 6.265reported p < .007 · recomputed p = .007

    “F(2,22)=6.265, p < .007”

  • Not confirmedF(2, 22) = 7.057reported p < .004 · recomputed p = .004

    “F(2,22)=7.057, p < .004”

  • ConsistentF(2, 22) = 22.01reported p < .001 · recomputed p < .001

    “F(2,22)=22.01, p < .001”

  • ConsistentF(2, 22) = 6.866reported p < .005 · recomputed p = .005

    “F(2,22)=6.866, p < .005”

  • ConsistentF(2, 20) = 0.701reported p < .508 · recomputed p = .508

    “F(2,20)=0.701, p < .508”

5 of 5

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Self-check results appear here once an independent check has confirmed them.

Citations

Cites

Cited by

See all 203

Metadata

Authors
Iqbal, S. T., & Bailey, B. P.
Affiliations
University of Illinois Urbana-Champaign (US)
Venue
CHI '05 Extended Abstracts on Human Factors in Computing Systems
Publisher
ACM
Published
2005-04-02
Pages
1489-1492
OpenAlex
W1971660807
Cited by
203
References
18
Citations by year
2026: 14 · 2025: 11 · 2024: 8 · 2023: 8 · 2022: 11 · 2021: 14 · 2020: 10 · 2019: 15 · 2018: 11 · 2017: 13 · 2016: 12 · 2015: 20 · 2014: 10 · 2013: 3 · 2012: 4
FWCI
9.06
Type
conference paper
Language
en
Open access
closed
Locations
CHI Conference Extended Abstracts on Human Factors in Computing Systems (CHI EA) (publishedVersion)
Topics
  • Social Sciences › Decision Sciences › Information Systems and Management › Personal Information Management and User Behavior
  • Social Sciences › Human Factors and Ergonomics › Human-Automation Interaction and Safety
  • Physical Sciences › Computer Science › Human-Computer Interaction › Usability and User Interface Design
Keywords
mental workload, task boundaries, social attribution

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.

A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

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

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.