The Basic Idea
Imagine someone asked you to describe how to walk. This would likely be a perplexing question, and might even be hard to answer, despite the fact that it is something you do on a daily basis. That’s because it’s something that you do without thinking. Habits are automatic, hard-to-break, and they form a pattern of behavior that responds to certain stimuli.1 Habits usually require a cue, repetition, and either a reward or punishment in order to be formed.
An activity like walking is an example of a habit. Walking might be more aligned with muscle memory, a physical kind of habit, but every day, we engage in dozens of habits both physically and mentally. Not only are our physical movements conditioned by habits, but the mental processes that we engage in eventually lead to habit-formation as well. This means that habits also influence our decision-making processes. When we are faced with a decision, we select a possible option based on the values that we associate with each. One way that we may decide an option is of value is if it is habitual. Since habits make up so much of our daily behavior, it is important that we understand them.
Theory, meet practice
TDL is an applied research consultancy. In our work, we leverage the insights of diverse fields—from psychology and economics to machine learning and behavioral data science—to sculpt targeted solutions to nuanced problems.
In part, habits are formed because we begin to associate certain triggers or situations with performing a particular behavior. Habit loops are neurological loops consisting of a cue, a behavior, and a reward.3 The reward means that your brain will start to associate a good feeling with the behavior that you just performed, making you likely to perform that same behavior the next time you encounter the cue. For example, imagine that you see a cookie. You decide to eat the cookie and feel happy because it tastes good. The next time you see a cookie, you will think about how eating it caused you to feel happy and will perform the same behavior.
While the habit loop describes the process through which habits are formed, a routine describes the process once the habit is formed. Routines are a sequence of actions or behaviors that you engage in on a regular basis, sometimes daily. Once a habit loop creates a habit, a routine is engaged in simply because it is a procedure, rather than because you are focused on the reward.4
At the core of behaviorism, also known as behavioral psychology, is the idea that all behaviors are acquired through conditions.5 That suggests that any individual can be trained to perform an action (within their physical limitations), because behaviorism suggests that actions are a product of responses to stimuli, rather than being products of subjective mental states like emotions, moods, or a priori cognitions.5
Reinforcement learning describes the ability to learn associations between stimuli, actions, and the occurrence of pleasant or unpleasant events6 – otherwise referred to as habit loops. The term suggests that associations can be strengthened through reinforcement. Reinforcement can come either in the form of a reward (a positive reinforcer) or a punishment (a negative reinforcer).6
Classical conditioning describes the process through which learning occurs by associating a behavior that naturally occurs – a reflex – to a new, conditioned stimulus. For example, imagine that you eat some shrimp and get food poisoning. The food poisoning is a natural reflex but you may come to associate it with shrimp, which might cause you to feel nauseous the next time you see shrimp. This is an example of accidental classical conditioning, but classical conditioning can be purposeful too.
Whereas classical conditioning forms habits by creating an association between a reflex and a created stimulus, operant conditioning is a method of learning that occurs through punishments and rewards, thereby being a form of reinforcement learning.5
Habits are often associated with the status quo bias. The status quo bias stipulates that when we are given a choice between leaving things as they are, or making a change, we are more likely to choose the option that leaves things as is. What is considered ‘status quo’ is what is habitual, which means that we may choose to perform the same behavior because it is familiar to us.
It is hard to pinpoint exactly when habits began to be studied because they are involved in so many different behaviors, such that they are now studied in a variety of fields. Psychology and behavioral science are fields deeply concerned with identifying and explaining the way that humans behave, which means that habits are of utmost interest to both.
All the way back in 1887, William James, a philosopher, and a psychologist, wrote a treatise titled Habits, in which he suggested that behavioral patterns shape who we are, in terms of our character and personality. The treatise demonstrated that habits have a very significant impact on our lives and our decision-making. He writes in Habits, “any sequence of mental action which has been frequently repeated tends to perpetuate itself.”11
One of the earliest famous studies on habit-formation was conducted by psychologist Ian Pavlov. Ian Pavlov was actually studying digestion, which he later won a Nobel prize for in 1904. During some of his studies, he discovered an interesting psychological phenomenon, referred to today as Pavlovian theory.7
In his experiments, Pavlov was trying to measure how much saliva dogs produced when they digested.7 His lab assistant would therefore present his dogs with a bowl of food that would then lead to the dogs salivating, an unconditioned response. After a while, Pavlov found that the dogs would salivate when they saw his lab assistant. They had come to associate the assistant with food, thereby causing the response. Based on this observation, Pavlov wanted to see if he could create a conditioned response. Pavlov decided to test whether a stimulus associated with food would be enough to create the salivation response.8
To begin, Pavlov would ring a bell when he gave the dogs food. After this was repeated a few times, Pavlov tried ringing the bell without giving the dogs food. Pavlov found that the dogs still salivated, suggesting they had come to associate the stimulus of the sound of a bell ringing with receiving food; a habit had been formed.8 This habit-forming process later became known as classical conditioning. Classical conditioning prompted the field of behaviorism because it suggests that individuals can be trained to perform an action through habit.
American psychologist B.F Skinner was the first to study operant conditioning. In 1948, Skinner examined how rats’ behavior changed based on different reinforcements. For positive reinforcement, he put hungry rats into a box that had a lever that would dispel food when pushed.9 The rats figured out that they would be rewarded for pressing a lever, and therefore after a while, the rats would go press the level immediately when they were put in the box. To show how negative reinforcement could also form habitual behavior, Skinner would place rats in a box with a level that would give them a small electric shock until they pressed a level that would turn off the electric current. After a while, the rats would go press the level immediately when they were put in the box.9 Skinner’s experiments demonstrated that habits could be formed through operant conditioning, which is a form of reinforcement learning.
Habits are also often studied in relation to health and fitness. Many people want to lose weight or get fitter, but in order to do so, we often have to break old habits and create new, healthier ones. Bad habits, like eating cookies because they make you feel good, are hard to break. Various studies have demonstrated that eating behavior is shaped by habits. There is a study by David Neal and Wendy Wool from 2011 that showed that people had formed a habit of eating popcorn while at the movie theatre and were thus likely to continue even if the popcorn they were served was stale.10
Another recent influential figure on habitual behavior is American author Gretchen Rubin. She has dedicated herself to researching habits and happiness. She has written various books and delivered numerous talks that discuss how habits are formed and strategies for creating better ones. To hear more about her work with habits, check out this podcast.
Habits impact virtually all aspects of our lives. They influence our decision-making, our physical actions, and our mental processes. And, unfortunately, many of our habits are incredibly unhealthy for us.
For example, addictions are related to habit. Cigarette smokers often will smoke at the same times each day because it has become part of their routine. They might associate having their morning coffee with having a cigarette or having an alcoholic drink with a cigarette, eventually forming a habit loop.
Habits are also manipulated through classical conditioning. In fact, marketing strategies often try to create habits in their consumers in order for their product to be successful. For example, Coca-Cola’s advertisements associated certain activities with their product. People in their ads are often playing sport or doing some kind of physical activity and are hot and thirsty. Next time people are thirsty, they may be prompted to think about Coca-Cola. The company has conditioned people to think of their product when they have the stimulus of thirst, sport, or heat, making it a habit to drink a Coca-Cola when you engage in such activities.8
Habits are often discussed in contrast to goal-oriented decision-making. Whereas habitual behavior often occurs on an unconscious level, as an automatic response to a particular trigger, goal-oriented decision-making requires putting time and energy into an action in order to get a reward.12 It is clear that as humans, we engage in both kinds of behaviors, but only recently has research shifted to examine goal-oriented choice mechanisms, whereas habits have long been researched. Goal-oriented decision-making does bring into debate the question of whether reinforcement learning is about creating habits or about influencing people to change their decision-making processes to adhere to creating goals.
Additionally, behaviorism and classical conditioning, which are concepts associated with habits, suggest that behavior can always be manipulated through changing conditions. However, this diminishes the influence of other factors, such as emotions, which, as various cognitive biases demonstrate, can be incredibly powerful influences on our decision-making. Some people believe that we overestimate the influence of habits at the cost of paying less attention to biological influences, our feelings, or other kinds of learning.5
COVID-19 and Habits
Think back to the beginning of the COVID-19 pandemic. All the changes felt incredibly jarring, because all of a sudden, you were no longer capable of engaging in your habitual behavior. Many people operate on a daily routine, but due to lockdowns and restrictions, people had to change their habits. However, as time has gone on, we have formed new habits that fall into the ‘new normal’. As habits create loops that are hard to break, it is likely that even as restrictions are lifted, some habits will remain. Behavioral scientists are interested in trying to predict what kind of behavior will stick, and what will be left behind.13 For example, will we continue to wear masks on public transport? Will meetings continue to be held via Zoom? Will digital menus stick over physical copies? Moreover, as some of our newly acquired habits have positive impacts for ourselves and the environment, behavioral science has a role to play in exploring the possibility of reinforcing the habits that we’ve formed so that they remain even when COVID-19 has diminished.13
Technology and Reinforcement Learning
Habits are usually discussed in relation to human behavior, however, reinforcement learning has recently been examined in relation to technology. Algorithms are not based solely on input functions; they become successful when they are able to understand the patterns of behavior of users.14 As The Social Dilemma, the popular Netflix documentary, discusses, algorithms are both conditioned by and help to create habits. Many algorithms function through model-based reinforcement learning, which uses user experience to form habits.14 For example, an algorithm is triggered on your phone when you open a social media app. It then shows you particular content, and it will either be rewarded (by you spending time on the app, by engaging through a like, etc), or punished (by you closing the app). It will come to understand what kind of behavior (content) leads to a reward and habitually begin to show you similar content. As you see more content that you like, you are also being influenced by reinforcement learning, and checking the social media app will become a habit.
Related TDL Content
In this article, our writer Zoe Adams examines the way in which individuals use technology like MyFitnessPal to develop and maintain healthy habits. She examines the concept of the quantified self, which refers to the behavior of using technology to track one’s habits and behaviors, and discusses the importance of the feedback loop that they create, acting like reinforcement learning.
Reflecting on how Panera Bread used a marketing strategy to create a habitual visit to their stores, our writer Yasmine Kalkstein discusses habits: how they are formed, why they work, and how to break them.
- Behavioral Economics. (2019, April 1). Habit. https://www.behavioraleconomics.com/resources/mini-encyclopedia-of-be/habit/
- Check Your Fact. (2019, June 26). FACT CHECK: Did Aristotle say, ‘We are what we repeatedly do’? https://checkyourfact.com/2019/06/26/fact-check-aristotle-excellence-habit-repeatedly-do/
- Duhigg, C. (2012, March 5). Habits: How they form and how to break them. National Public Radio. https://www.npr.org/2012/03/05/147192599/habits-how-they-form-and-how-to-break-them
- Dictionary.com. (n.d.). Routine. www.dictionary.com. Retrieved October 31, 2020, from https://www.dictionary.com/browse/routine?s=t
- Cherry, K. (09, September 24). History and Key Concepts of Behavioral Psychology. Verywell Mind. https://www.verywellmind.com/behavioral-psychology-4157183
- Palminteri, S., & Pessiglione, M. (2013). Nondopaminergic Neurotransmission in the pathophysiology of Tourette syndrome. International Review of Neurobiology, 112, 131-153. https://doi.org/10.1016/b978-0-12-411546-0.00004-4
- Weinschenk, S. (2019, April 19). The Science of Habits. Psychology Today. https://www.psychologytoday.com/us/blog/brain-wise/201904/the-science-habits
- Husson University. (2018, June 1). Consumer behavior theories: Pavlovian theory. https://online.husson.edu/consumer-behavior-pavlovian-theory/
- McLeod, S. (2007, February 5). Skinner – Operant Conditioning. Simply Psychology. https://www.simplypsychology.org/operant-conditioning.html
- Neal, D., & Wool, W., Wu, M., & Kurlander, D. (2011). The pull of the past: When do habits persist despite conflict with motives? Personality and Social Psychology Bulletin, 37 (11), 1428-1437.
- Popova, M. (2017, July 2). William James on the Psychology of Habit. Brain Pickings. https://www.brainpickings.org/2012/09/25/william-james-on-habit/
- LeMieux, J. (2018, March 16). Our Brain Switches Between Habits And Choices. Here’s How. American Council on Science and Health. https://www.acsh.org/news/2018/03/16/our-brain-switches-between-habits-and-choices-heres-how-12692
- Material. (2020, October 14). The behavioral science of habit formation. https://materialplus.io/behavioral-science/the-behavioral-science-of-habit-formation/
- Moni, R. (2019, February 18). Reinforcement Learning algorithms — an intuitive overview. Medium. https://medium.com/@SmartLabAI/reinforcement-learning-algorithms-an-intuitive-overview-904e2dff5bbc