Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record, with a small difference in title or year; 74 of the 74 references it cites were found in a registry or an index; 9 of the 53 sources its claims cite were checked for support.

Sorting Insiders From Co-Workers: Remote Synchronous Computer-Mediated Triage for Investigating Insider Attacks

Open access
Dando, Coral J., Taylor, Paul J., Menacere, Tarek, Ormerod, Thomas C., Ball, Linden J., Sandham, Alexandra L. · 2024 · Human Factors: The Journal of the Human Factors and Ergonomics Society

Abstract

OBJECTIVE: for quickly highlighting persons of interest after an insider attack. BACKGROUND: Insiders maliciously exploit legitimate access to impair the confidentiality and integrity of organizations. The globalisation of organisations and advancement of information technology means employees are often dispersed across national and international sites, working around the clock, often remotely. Hence, investigating insider attacks is challenging. However, the cognitive demands associated with masking insider activity offer opportunities. Drawing on cognitive approaches to deception and understanding of deception-conveying features in textual responses, we developed InSort, a remote computer-mediated triage. METHOD: During a 6-hour immersive simulation, participants worked in teams, examining password protected, security sensitive databases and exchanging information during an organized crime investigation. Twenty-five percent were covertly incentivized to act as an 'insider' by providing information to a provocateur. RESULTS: Responses to InSort questioning revealed insiders took longer to answer investigation relevant questions, provided impoverished responses, and their answers were less consistent with known evidence about their behaviours than co-workers. CONCLUSION: Findings demonstrate InSort has potential to expedite information gathering and investigative processes following an insider attack. APPLICATION: InSort is appropriate for application by non-specialist investigators and can be quickly altered as a function of both environment and event. InSort offers a clearly defined, well specified, approach for use across insider incidents, and highlights the potential of technology for supporting complex time critical investigations.

Overview

CoChat Summary

## TL;DR

This paper develops and tests InSort, a remote synchronous text-based triage for identifying persons of interest after an insider attack. In a 6-hour simulation with 60 participants, insiders took longer, wrote less, and showed more answer-evidence inconsistency than co-workers.

## Key findings

  • Insiders took significantly longer to complete InSort than non-insiders: 696s vs 340s; F(1,54)=187.81, p < .001, ηp²=0.88.
  • For open target questions, insiders wrote fewer words than non-insiders: significant group effects for OQ1 and OQ2, with non-insiders writing 3 times more words on OQ1 and 2.5 times more words on OQ2.
  • Insiders provided less information in open responses: significant group effects for total information items in OQ1, F(1,36)=9.485, p=.003, ηp²=0.22 and OQ2, F(1,36)=34.75, p < .001, ηp²=0.49.
  • Insiders were more inconsistent with known evidence on closed target questions: U=43.00, z=-5.046, p < .001; insiders’ median answer-evidence inconsistency was 22.07 vs 15.85 for non-insiders.
  • In feedback, insiders rated InSort as more demanding and reported lower confidence in their answers: F(1,59)=39.11, p < .001, ηp²=0.67 for demand and F(1,59)=45.41, p < .001, ηp²=0.62 for confidence.

## Methods

  • The study used a 6-hour immersive simulation in the iCOS serious game, with participants working in teams investigating organized crime using password-protected databases; 25% were covertly incentivized to act as insiders.
  • 60 participants were paid £50; teams were randomly assigned to Fraud, Human Trafficking, or Narcotics, with four roles per team.
  • InSort included 56 questions (including repeated yes/no and open target questions), and analyses used two-way ANOVAs, Mann-Whitney tests, and coding of open responses into information items.

## Limitations

  • The authors state that the simulation differed from the real world, including the absence of a “world” outside the simulation and the fact that insiders were chosen at random without controlling/measuring personality, motivation, or personal circumstances.
  • They note that they compared known insiders to co-workers as a first step; they explicitly say more research is required to determine how to distinguish signal from noise when insider status is unknown.
  • Evident from the text: the study was conducted in a controlled serious-game simulation with a small sample from one experimental setting, so generalizability to real organizational incidents is limited.

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

Integrity signals

Registry

Crossref · 10.1177/00187208211068292

Cites

74 of 74 references verified0 citation errors · 0 not checkable

Statistics

9 tests recomputed · 0 with the reported and recomputed p on opposite sides of .05

Figures

Figure 1. Mean word count for each of open question (OQ1, OQ2 and OQ3) as a function of group (insider; non-insider).
Figure 1. Mean word count for each of open question (OQ1, OQ2 and OQ3) as a function of group (insider; non-insider).CC BY · PubMed Central

Verification

Claims the paper makes

7 Valid · 2 Not valid · 15 Not checked · 39 Own results and interpretations

26 more claims

Create a free account to see all 63 claims and to verify the 7 references still open.

See all claims — free account
  • Probably not a substantial claim. 25
  • It may state two separate claims as one. 14

Create a free account to see these claims.

Statistics that don’t add up

Each reported p-value recomputed from its test statistic and degrees of freedom.

1 of 1

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.

  • Doesn’t add upF(1, 36) = 9.485reported p = .003 · recomputed p = .004

    “F (1, 36) = 9.485, p = .003”

  • ConsistentF(1, 36) = 12.866reported p = .001 · recomputed p < .001

    “F (1, 36) = 12.866, p = .001”

  • ConsistentF(1, 36) = 23.95reported p < .001 · recomputed p < .001

    “F (1, 36) = 23.95, p < .001”

  • ConsistentF(1, 36) = 34.75reported p < .001 · recomputed p < .001

    “F (1, 36) = 34.75, p < .001”

  • ConsistentF(1, 36) = 29.268reported p < .001 · recomputed p < .001

    “F (1, 36) = 29.268, p < .001”

5 of 9 · the rest aren’t loaded

4 more recomputed tests

Create a free account to see every test the Stats check recomputed.

See all tests — free account

Self-check

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

1 of 1

Citations

Cites

Show all 74 references

Research works printed but not linked: 9

  1. Clearswift Insider Threat Index. (2017). https://www.clearswift.com/about-us/pr/press-releases/insider-threat-74-security-incidents-come-extended-enterprise-not-hacking-groups
  2. CPNI, 2020, Investigation and disciplinary
  3. Jenkins M. C., 2012, Proceedings of the 10th Biennial Conference of the International Conference of Forensic Linguistics
  4. Levine T. R., 2018, International Journal of Communication, 12, 2461
  5. Matsumoto D., 2011, FBI L. Enforcement Bull, 80, 1
  6. National Law review. (2020). Frequency and cost of insider threats continue to increase. https://www.natlawreview.com/article/frequency-and-cost-insider-threats-continue-to-increase
  7. Sandham A. L., 2020, Journal of Police and Criminal Psychology, 1
  8. Trzeciak R., 2019, SEI cyber minute: Insider threat mitigation, we can help!
  9. Walczyk J. J., 2013, Applied Psychology in Criminal Justice, 9, 1

Cited by

See all 10

Metadata

Authors
Dando, C. J., Taylor, P. J., Menacere, T., Ormerod, T. C., Ball, L. J., & Sandham, A. L.
Affiliations
  • University of Westminster (GB)
  • Lancaster University (GB)
  • University of Sussex (GB)
  • University of Lancashire (GB)
  • University of Gloucestershire (GB)
Corresponding author
Dando, Coral J.
Venue
Human Factors: The Journal of the Human Factors and Ergonomics Society
Published
2022-03-07
Volume
66
Issue
1
Pages
145-157
ISSN
  • 0018-7208 (ISSN-L)
  • 1547-8181
OpenAlex
W3216334198
Cited by
10
References
74
Citations by year
2026: 1 · 2025: 2 · 2024: 5 · 2023: 2
FWCI
2.13
Type
journal article
Language
en
Open access
Locations
  • Human Factors The Journal of the Human Factors and Ergonomics Society (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • Research Repository (University of Gloucestershire) (acceptedVersion)
  • CLOK (University of Central Lancashire) (acceptedVersion)
  • Research Repository (University of Gloucestershire) (submittedVersion)
  • Lancaster EPrints (Lancaster University) (submittedVersion)
  • WestminsterResearch (University of Westminster) (submittedVersion, cc-by)
  • Figshare (submittedVersion, cc-by)
Funders
UK Government (Project Ref. LUR 46.851)
Topics
  • Social Sciences › Psychology › Social Psychology › Deception detection and forensic psychology
  • Physical Sciences › Computer Science › Information Systems › Information and Cyber Security
  • Physical Sciences › Computer Science › Information Systems › Digital and Cyber Forensics
Keywords
insider attacks, deception detection, remote triage, investigative questioning, organizational security
MeSH terms
Algorithms, Computer Simulation, Computers, Humans, Triage, Computer Security

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.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

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).

Cited sources are checked with a language model’s help (qwen3.8-omni-flash and gpt-6-sol).

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

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Sorting Insiders From Co-Workers: Remote Synchronous Computer-Mediated Triage for Investigating Insider Attacks. https://app.cochat.ai/papers/10.1177/00187208211068292
The paper
Dando, C. J., Taylor, P. J., Menacere, T., Ormerod, T. C., Ball, L. J., & Sandham, A. L. (2024). Sorting Insiders From Co-Workers: Remote Synchronous Computer-Mediated Triage for Investigating Insider Attacks. Human Factors: The Journal of the Human Factors and Ergonomics Society. https://doi.org/10.1177/00187208211068292

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.