Characterizing human habits in the lab

Abstract

Habits pose a fundamental puzzle for those aiming to understand human behavior. They pervade our everyday lives and dominate some forms of psychopathology but are extremely hard to elicit in the lab. In this Registered Report, we develop novel experimental paradigms grounded in computational models, which suggest that habit strength should be proportional to the frequency of behavior and, in contrast to previous research, independent of value. Specifically, we manipulate how often participants perform responses in two tasks varying action repetition without, or separately from, variations in value. Moreover, we ask how this frequency-based habitization relates to value-based operationalizations of habit and self-reported propensities for habitual behavior in real life. (This is the first stage of a Registered Report in-principle-accepted by Collabra: Psychology.)

Cognitive Computational Model Reveals Repetition Bias in a Sequential Decision-Making Task

“…It is related to stimulus-response (S-R) learning, as repetition facilitates the formation of S-R associations and re-cency effects in value-based decision-making tasks ( Guthrie, 1952; Wood and Rünger, 2016; Watson et al, 2022 ). The repetition bias is also consistent with previous proposals for the role of action repetition in the development of habitual behavior ( Thorndike, 1911; Miller et al, 2019; Schwöbel et al, 2021; Nebe et al, 2024 ). Similarly, behavioral repetition of action sequences has been identified as a way to optimize the trade-off between maximizing reward and a reduction of policy complexity ( Gershman, 2020 ).…”

“…Importantly, the repetition learning mechanism is different from stimulus-response associa- The repetition bias is possibly a prerequisite for the development of habitual behavior (Wood and Rünger, 2016;Miller et al, 2019;Schwöbel et al, 2021;Nebe et al, 2024). This opens up possibilities to use this mechanism and its predictions to investigate the formation of habits.…”

“…Here, repetition significantly affects perceptual decision making by accelerating response times for ambiguous stimuli ( Akaishi et al, 2014 ). Similarly, repetition can alter preferences in value-based decision-making processes ( Nebe et al, 2024 ), suggesting that the influence of repetition extends beyond the direct anticipation or receipt of reward, challenging standard views on the effect of reward on decision making and learning. Most importantly, repetitions are seen as a key element of habit formation ( Wood and Rünger, 2016; Watson et al, 2022 ).…”

Computationally-informed insights into anhedonia and treatment byκ-opioid receptor antagonism

“…A prediction of our findings is that anhedonia may not only manifest as a deficit in perseveration, but may also manifest as a deficit in habit formation. Intriguingly, recent work on the origin of habits has revealed that they are largely divorced from reward [44]. If true, this would highlight a cognitive deficit in anhedonia unrelated to reward processing.…”

Social Media Use is Predictable from App Sequences: Using LSTM and Transformer Neural Networks to Model Habitual Behavior

“…While these findings demonstrate regularities in smartphone user behaviors that can be interpreted as habitual, the behavioral science literature has taken a different approach, often relying on relatively simple behavioral measures (e.g., behavioral frequency, reward devaluation [35,36]) or self-report measures to assess habits (e.g., Self-Report Habit Index [37], Self-Report Behavioral Automaticity Index [38], Unified Theory of Acceptance and Use of Technology questionnaire [39], Response Frequency Measure of Media Habit [40]). These methods have proven useful in past research, but they are also limited in at least two important ways.…”