Current limits of experimental research into habits and future directions
Introduction
Habits are of great interest to the scientific community with a PubMed search revealing that in 2016, nearly 5000 publications contained the word ‘habit’ in the title or abstract. While the definition of habit varies, researchers in the field of associative learning distinguish between goal-directed behavior that is sensitive to the current motivational value of outcomes and habitual behaviors that are stamped in through repetition and are directly triggered by the environment [1, 2]. The outcome devaluation paradigm was developed as a means of assessing whether observed behavior is goal-directed or habitual [3, 4]. Following devaluation of an outcome (through, e.g. satiation on a particular food) responding for that outcome is assessed (see Figure 1, left panel). If behavior is goal-directed then responding for the devalued outcome will be reduced. Persistent responding for the devalued outcome, however, is typically regarded as evidence that responses are being directly triggered by the context, via stimulus-response (S-R) associations. This choice test is performed in extinction (without any reinforcement) in order to prevent novel learning and examine instead the integration of the new outcome value with existing response-outcome (R-O) associations. The outcome devaluation paradigm has been used to systematically investigate the parameters that affect the transition of control from goal-directed to habitual behavior in animals, and the neurobiological substrates of these two processes. Although the inflexibility and efficiency of habits is often mentioned in the same breath, this body of research remains quite separate from that of (motor) skill learning (e.g. [5]) where automaticity is measured as, for example, decreased RTs with repetition. The following sections will highlight some of the translational research from animals to humans using the outcome devaluation paradigm and some of the challenges that have arisen in this field, particularly in the hunt for the behavioral and neurobiological signatures of habits in humans.
Section snippets
Parameters leading to insensitivity to outcome devaluation
Seminal studies in the 1980s showed that even rats could flexibly adjust responding depending on the current incentive value of the consequences of their actions [3, 4]. Adams and colleagues demonstrated that moderately trained rats immediately reduced responding for a food outcome that had been paired with lithium chloride (inducing illness), whereas extensively trained rats displayed no such reduction in responding for the devalued food [3, 4, 6]. In line with dual-process theory, subsequent
Demonstrations of habits
Insensitivity to devaluation — for example, in certain psychopathologies — could result from excessive habit formation or weak goal-directed control or a combination of the two. Most of the behavioral paradigms currently at our disposal (such as the slips-of-action task) measure the relative balance between the two systems but do not allow us to distinguish between them (see: [31•] for a similar discussion of habits in the domain of compulsivity). Although the sequential-decision task includes a
Alternative ways to study habits
Translational research is difficult — especially because human participants use different (explicit) strategies that could potentially prevent the behavioral expression of habit formation. Some studies have measured insensitivity to outcome devaluation using a concurrent choice test in which participants are simply asked to choose between two outcomes of differing value and, crucially, in the absence of any discriminative stimuli. Using this very simple procedure where the two outcomes (differing
Conclusion
For nearly 40 years the outcome devaluation paradigm has provided a clear and coherent measure of assessing whether behavior is goal-directed or habitual. Researchers in the field of associative learning, behavioral neurosciences and clinical psychology have used this paradigm to explore the conditions under which behavior is shifted from one to the other forms of control and the distinct neurobiological networks that support both types of learning. However, doubts remain as to the extent to
Funding
Funded by the Netherlands Organization for Scientific Research (NWO; 016.145.382).
Conflict of interest statement
Nothing declared.
References and recommended reading
Papers of particular interest, published within the period of review, have been highlighted as:
• of special interest
References (66)
- et al.
Neuroimaging studies of practice-related change: fMRI and meta-analytic evidence of a domain-general control network for learning
Cogn Brain Res
(2005) - et al.
Reduced goal-directed action control in autism spectrum disorder
Autism Res
(2016) - et al.
Loss of lateral prefrontal cortex control in food-directed attention and goal-directed food choice in obesity
NeuroImage
(2017) - et al.
General habit propensity relates to the sensation seeking subdomain of impulsivity but not obesity
Front Behav Neurosci
(2016) - et al.
Characterizing a psychiatric symptom dimension related to deficits in goal-directed control
eLife
(2016) - et al.
Stress-induced changes in human decision-making are reversible
Transl Psychiatry
(2012) - et al.
Tryptophan depletion promotes habitual over goal-directed control of appetitive responding in humans
Int J Neuropsychopharmacol
(2015) - et al.
Model-based control in dimensional psychiatry
Biol Psychiatry
(2017) - et al.
Dopamine enhances model-based over model-free choice behavior
Neuron
(2012) Relations between Pavlovian-instrumental transfer and reinforcer devaluation
J Exp Psychol Anim Behav Process
(2004)
Choice and goal-directed behavior in preschool children
Learn Behav
Appetitive Pavlovian-instrumental transfer: a review
Neurosci Biobehav Rev
Contextual control of chained instrumental behaviors
J Exp Psychol Anim Learn Cogn
Goal-directed and habit-like modulations of stimulus processing during reinforcement learning
J Neurosci Off J Soc Neurosci
Investigating habits: strategies, technologies and models
Front Behav Neurosci
Instrumental conditioning
Instrumental conditioning revisited: updating dual-process theory
Instrumental responding following reinforcer devaluation
Q J Exp Psychol Sect B
Actions and habits: the development of behavioural autonomy
Philos Trans R Soc Lond B Biol Sci
Variations in the sensitivity of instrumental responding to reinforcer devaluation
Q J Exp Psychol Sect B
The psychological and physiological mechanisms of habit formation
Corticostriatal connectivity underlies individual differences in the balance between habitual and goal-directed action control
J Neurosci
Functional neuroimaging of avoidance habits in obsessive-compulsive disorder
Am J Psychiatry
Behavioral and neuroimaging evidence for overreliance on habit learning in alcohol-dependent patients
Transl Psychiatry
Enhanced habit formation in Gilles de la Tourette syndrome
Brain J Neurol
Carrots and sticks fail to change behavior in cocaine addiction
Science
Corticostriatal control of goal-directed action is impaired in schizophrenia
Biol Psychiatry
Slave to habit? Obesity is associated with decreased behavioural sensitivity to reward devaluation
Appetite
Stimulus control over action for food in obese versus healthy-weight individuals
Front Psychol
An investigation of habit learning in Anorexia Nervosa
Psychiatry Res
Are nonclinical obsessive–compulsive symptoms associated with bias toward habits?
Psychiatry Res
Socially evaluated cold pressor stress after instrumental learning favors habits over goal-directed action
Psychoneuroendocrinology
Reliance on habits at the expense of goal-directed control following dopamine precursor depletion
Psychopharmacology (Berl)
Cited by (76)
Acute stress facilitates habitual behavior in female rats
2024, Physiology and BehaviorInfluence of social-cultural practices on the adoption of Renewable Energy Technologies (RETs) in Uganda
2023, Renewable Energy FocusMaking habits measurable beyond what they are not: A focus on associative dual-process models
2022, Neuroscience and Biobehavioral ReviewsAn fMRI meta-analysis of the role of the striatum in everyday-life vs laboratory-developed habits
2022, Neuroscience and Biobehavioral ReviewsCitation Excerpt :An issue here could be that in some of these studies behaviours established in the laboratory that are presumed to be habits fail to meet formal requirements. For example, several recent human experimental studies failed to demonstrate an effect of training duration on insensitivity to outcome-devaluation test (de Wit et al., 2018; Pool et al., n.d.; Watson and de Wit, 2018). In such studies it is possible that more trials may be needed to establish stimulus-response associations that are fully automatic and can persist after devalued outcome challenges.