I remember the day I discovered human factors. I took the course as an elective, because I’d already taken every other psychology class available to me. At the time, I knew I loved psychology but also knew I didn’t want to be a clinician or a counselor. The first day, we were assigned a typical “bad design” project – find a frustrating object in the world and write about what was wrong with it. The hardest part of the assignment was limiting myself to one bad design – I had a grievance list pages long. I knew then that I had found a field that matched my desire to ‘fix’ the world.
So many of us who work in human factors had a similar epiphany, and yet our field is virtually unknown to the public. I wanted to change that by writing the kind of non-fiction book I like to read: jargon-free content augmented by stories from history, the news, and the author’s own experiences.
This month my book was published by Cambridge University Press: All Too Human. The intended audience spans the general public, students, and practitioners in related fields like engineering and computer science.
The Process
I began this project over three years ago. For my cognitive class (undergraduate), I wrote chapters to show the application of basic processes: perception, attention, memory. For my human factors psychology course (graduate), I wanted to tell stories that paired with the primary source readings and spurred discussion.
Each semester I user-tested the chapters by having students write reaction papers, and then iterating on the text to make sure I met my three goals: 1) jargon-free and accessible writing, and 2) conveyed the message that to build our world, we must understand the capabilities and limitations of the human mind, and 3) there’s a profession that does just that: human factors psychology. I edited the chapters repeatedly to make them approachable with fun figures and end-of-book notes rather than in-text citations (to keep the text clean and easy to read.)
The Audiences
For educators: At the undergraduate or graduate level, I wrote this book to be a companion to text and lecture. I often use companion texts in my own courses, such as The Design of Everyday Things or Set Phasers on Stun, to grab students’ attention and generate discussion. Email [email protected] to ask for an examination copy.
For practitioners: I envisioned this book as one you could recommend to co-workers to give them a fun overview of “what we do” and what we can do. I want it to inspire multi-disciplinary conversations. I want others to understand where our recommendations come from.
For non-fiction lovers: This book has a message for you. It is that we can control the design of our human world. Products and systems should work for us, offer us pleasure, efficiency, and safety. But for that to happen, we must demand it. This book will give you the background to help you understand just how much control we deserve, and in how many ways we can influence companies, governments, and ourselves to make a better world.
Where to Buy
Available from all bookstores, including:
Amazon
Barnes & Noble
Cambridge University Press
Excerpt from the chapter Needles in Haystacks, on signal detection theory:
Sensitive and Specific
In August of 2020 Magawa received the PDSA Gold Medal for bravery and devotion to duty. A native of Tanzania, he was described as brave, friendly, and a determined worker. Over the course of four years, he found thirty-nine buried landmines in Cambodia. In a country with millions of mines there is still a ways to go, but experts like Magawa are on the front lines of mine detection. Also, Magawa is a rat.
A great attribute of mine-seeking rats is their weight. Although this particular species, African giant pouched rats, are very large for rats, they are still too light to trigger a landmine. They are also smart, have noses at least as discerning as the eyesight of pigeons, and are easy to train. Their training is a real rat race – they work about five hours each weekday on learning to discriminate the smell of a landmine from other scents, all for some mashed banana. They indicate finding a mine by stopping and digging with their little paws.
Magawa was trained by the APOPO, a Belgian non-profit with a mouthful of a name (the Anti-Persoonsmijnen Ontmijnende Product Ontwikkeling, which translates to “Anti-Personnel Landmines Removal Product Development”). The rats’ training is extensive and their success is measured in hits, misses, false alarms, and correct rejections, with more available data than the Project SEA HUNT pigeons. Why not just use a metal detector? The answer fits the theme of this chapter: the high cost of false alarms. The ground is full of nonexplosive metal bits and trash and one estimate is that there are 1,000 false alarms per mine found using just detectors. The rats, meanwhile, can only graduate by demonstrating a 100 percent hit rate on four buried mines and no more than one false alarm. The rats have to be both sensitive and specific. The sensitivity rate is calculated as the number of mines they identified in an area created by their handlers divided by the number of total mines in that area. Their specificity is the number of false alarms per 100 square meters – a slightly different calculation than for other signal detection where specificity is the probability of a false alarm. It may be too much to ask of any single rat to be such a perfect performer, so typically every area is searched by a few rats and their combined scores make the final sensitivity and specificity rates for the team.
In a study led by psychologist Alan Poling, a professor at Western Michigan University, the trained rats were taken to Mozambique where they explored 93,400 square kilometers of mine-infested area. The rats found 41 mines. Humans used other tools to check the area as well, like metal detectors, but found no additional mines. The rats succeeded with extreme sensitivity (100 percent hits and zero misses). But what about their specificity? In this real-world test, they had about 0.33 false alarms for every 100 square meters searched (617 false alarms), or about 39,383 fewer than would be expected from a metal-detecting human. My math is undoubtedly a brash generalization, as the mines aren’t distributed equally and differ by country, but in terms of both hits and false alarms the rats are clear winners.
The post “All Too Human” Now Available first appeared on the Human Factors Blog.