Measuring response time in complex decision tasks

Background

For background, please refer to:
https://knowlabunimelb.github.io/SCHEDULE_RT/index.html

Research Questions / Hypotheses

This is an ongoing study that aims to investigate how the optimal task scheduling strategy changes based on the presence or absence of reward. The optimal policy to maximise the number of tasks completed before the deadline is to complete the shortest subtask first. This policy does not necessarily hold when importance varies between tasks of increasing difficulty (eg. differing reward values for task completion). An optimal strategy would be to complete the task producing the highest value reward per unit of time. We will determine whether participants follow this optimal schedule. We will compare scheduling with two typing tasks under deadline: a baseline without rewards, and a follow-up experiment with rewards. We will expect participants to follow the precedented optimal strategy under a deadline and without reward, and the alternative optimal strategy under a deadline and with reward.

Participants

42 participants were recruited with 38 participants completing the study.

Methods

Participants completed multiple trials of selection and completion of typing tasks online. On each trial, participants were shown a set of four tasks labelled Short, Medium, Long, and Very Long, describing the length of the typing task. Participants selected and completed one task at a time, in the order of their choosing, attempting to complete as many as possible before the deadline. After each complete trial, participants were presented with a word game similar to Wordle with clues corresponding to the difficulty of the typing tasks they were able to complete before the deadline. Participants guessed a word based on the clues obtained.

Results

This is an ongoing study. Please refer to:
https://knowlabunimelb.github.io/SCHEDULE_RT/index.html

Implications

We expect that participants will follow the precedented optimal strategy of shortest processing time when completing tasks of varying difficulty under a deadline without reward. We expect that participants will use an alternative optimal strategy (weighted shortest processing time) to schedule tasks under a deadline when differing in difficulty level and assigned reward value. This has implications for using reward as a goal motivator, when scheduling tasks under time pressure, to optimise workloads. Results will be presented in an Honours oral presentation and written thesis.