Interactive Instruction in Bayesian Inference

Azam Khan; Simon Breslav; Kasper Hornbæk

doi:10.1080/07370024.2016.1203264

Interactive Instruction in Bayesian Inference

Azam Khan^*, Simon Breslav, Kasper Hornbæk

^*Corresponding author for this work

4 Citations (Scopus)

Abstract

An instructional approach is presented to improve human performance in solving Bayesian inference problems. Starting from the original text of the classic Mammography Problem, the textual expression is modified and visualizations are added according to Mayer’s principles of instruction. These principles concern coherence, personalization, signaling, segmenting, multimedia, spatial contiguity, and pretraining. Principles of self-explanation and interactivity are also applied. Four experiments on the Mammography Problem showed that these principles help participants answer the questions at significantly improved rates. Nonetheless, in novel interactivity conditions, performance was lowered suggesting that more interaction can add more difficulty for participants. Overall, a leap forward in accuracy was found, with more than twice the participant accuracy of previous work. This indicates that an instructional approach to improving human performance in Bayesian inference is a promising direction.

Original language	English
Journal	Human-Computer Interaction
Volume	33
Pages (from-to)	207–233
ISSN	0737-0024
DOIs	https://doi.org/10.1080/07370024.2016.1203264
Publication status	Published - 4 May 2018

Access to Document

10.1080/07370024.2016.1203264

Cite this

@article{2ce8f115920c47428e310122ad36ef6b,

title = "Interactive Instruction in Bayesian Inference",

abstract = "An instructional approach is presented to improve human performance in solving Bayesian inference problems. Starting from the original text of the classic Mammography Problem, the textual expression is modified and visualizations are added according to Mayer{\textquoteright}s principles of instruction. These principles concern coherence, personalization, signaling, segmenting, multimedia, spatial contiguity, and pretraining. Principles of self-explanation and interactivity are also applied. Four experiments on the Mammography Problem showed that these principles help participants answer the questions at significantly improved rates. Nonetheless, in novel interactivity conditions, performance was lowered suggesting that more interaction can add more difficulty for participants. Overall, a leap forward in accuracy was found, with more than twice the participant accuracy of previous work. This indicates that an instructional approach to improving human performance in Bayesian inference is a promising direction.",

author = "Azam Khan and Simon Breslav and Kasper Hornb{\ae}k",

year = "2018",

month = may,

day = "4",

doi = "10.1080/07370024.2016.1203264",

language = "English",

volume = "33",

pages = "207–233",

journal = "Human-Computer Interaction",

issn = "0737-0024",

publisher = "Taylor & Francis Online",

}

TY - JOUR

T1 - Interactive Instruction in Bayesian Inference

AU - Khan, Azam

AU - Breslav, Simon

AU - Hornbæk, Kasper

PY - 2018/5/4

Y1 - 2018/5/4

N2 - An instructional approach is presented to improve human performance in solving Bayesian inference problems. Starting from the original text of the classic Mammography Problem, the textual expression is modified and visualizations are added according to Mayer’s principles of instruction. These principles concern coherence, personalization, signaling, segmenting, multimedia, spatial contiguity, and pretraining. Principles of self-explanation and interactivity are also applied. Four experiments on the Mammography Problem showed that these principles help participants answer the questions at significantly improved rates. Nonetheless, in novel interactivity conditions, performance was lowered suggesting that more interaction can add more difficulty for participants. Overall, a leap forward in accuracy was found, with more than twice the participant accuracy of previous work. This indicates that an instructional approach to improving human performance in Bayesian inference is a promising direction.

AB - An instructional approach is presented to improve human performance in solving Bayesian inference problems. Starting from the original text of the classic Mammography Problem, the textual expression is modified and visualizations are added according to Mayer’s principles of instruction. These principles concern coherence, personalization, signaling, segmenting, multimedia, spatial contiguity, and pretraining. Principles of self-explanation and interactivity are also applied. Four experiments on the Mammography Problem showed that these principles help participants answer the questions at significantly improved rates. Nonetheless, in novel interactivity conditions, performance was lowered suggesting that more interaction can add more difficulty for participants. Overall, a leap forward in accuracy was found, with more than twice the participant accuracy of previous work. This indicates that an instructional approach to improving human performance in Bayesian inference is a promising direction.

U2 - 10.1080/07370024.2016.1203264

DO - 10.1080/07370024.2016.1203264

M3 - Journal article

AN - SCOPUS:85003890143

SN - 0737-0024

VL - 33

SP - 207

EP - 233

JO - Human-Computer Interaction

JF - Human-Computer Interaction

ER -

Interactive Instruction in Bayesian Inference

Abstract

Access to Document

Fingerprint

Cite this