Mental workload is defined as the cognitive demand of a task. It has been established that human performance drops where the mental workload of a task exceeds a persons capacity. In this article I want to give a personal opinion on mental workload in the lab.
Mental Workload - a definition (skip)
Mental workload (MWL) is an often missused term. It is simply defined as the cognitive demand of a task, but in theory it is a complex concept. Longo et al. (2022) found that 22 different theories have been proposed to (in part) explain the underlying processes. MWL is much easier to define by looking at the way it is measured. The most popular measure of MWL is the NASA Task Load Index (NASA-TLX). Although originally created as a measure of overall workload, it is widely accepted as a meausre of MWL (Babaei et al 2025).
The NASA-TLX is a quesitonaire that asks workers to rank and score six seperate scales of workload: mental demand, physical demand, temporal demand, performace, effort and frustration (link). One often missunderstood aspect in analysis of the quesitonaire is that all the factors are correlated and should be assumed to measure the same dimension (unless evidence gives proof to the opposite).
High Mental Workload Tasks - A Lab Example (skip)
Although Mental Workload (MWL) is difficult to define without looking the standard measurement (NASA-TLX), it is useful to have an idea of the underlying processes. I will explain some of these processes with the help of a lab example. The example I will use is the sample allocation for large scale duckweed assays (SA), a task with problematically high MWL.
For this example lets assume that the researcher has to allocate 5 different plant samples and 27 different growth media into 1296 wells in one sitting. In practice the researcher sits on a laminar flow hood with a stack of well plates, multiple fasks containing different growth media, muliple containers with different genotypes and a sheet of paper with a table detailing how to label a plate and what to allocate into each well. As the researcher works down the list she/he has to track:
- The position of the target well
- The type of sample (medium/genotype)
- The ID of the sample (often five or more arbitrary nubmers and letters)
- The position of the current well and sample on the sheet
For this task the key dimension determining MWL is mental demand. The high amount of information that has to be kept in short term memory. Short term memory acts very much like working memory in computers, it is a store that keeps information to be processesed. Unlike computers, human "working memory" is very limited, with the average human being unable to store more than three pieces of information. For our example the information that has to be memorized exceed the short term capacity of most humans, especially considering that the sample ID can be long and arbitrary. But the key challenge of the example task is the volume of work and hence the time-frame of constant mental demand.
Mental Workload in Context of Day-to-day Work (skip)
The majority of tasks completed by researchers do not score high in MWL in isolation, but only when seen in context. The main external facotrs that can push mid MWL task to a problematic load level are:
1) Context switching
Context switching means switching between tasks that require different information to be kept in short term memory. Frequently switching context increases the Menal Workload (MWL) of tasks, as often observed for multitasking. Researchers often lead or contribute to multiple projects at the same time and lab work often involves doing tasks concurrently (e.g. setting up the next experiment in the idle time of the current one). Research is a field where frequent context switches are required to be effective.
2) Time restriction
By time restricion I do not mean the inherent time sensitivity of certain tasks but the pressure put on reseachers to complete a certain volume of work in a given time (often self-imposed). The pressure to fit more tasks into a days work acts synergistcally with the other points on this list.
3) Reliance on Records
There are two reasons reseachers are highly reliant on records and labels. Firstly, the identitiy or state of many resources is invisible (e.g. genotype of a plant, concentration of a solution). Secondly, the sheer volume of different resources being generated. Record keeping by it's nature takes time and interrupts the flow of tasks (causing mild context switches).