AHAdrian M. Haith

Adrian M. Haith

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Adrian M. Haith has a total of 58 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Bridging the divide in motor learning research

    by Eric Grießbach, Alfredo Hernandez-Inostroza, Rouwen Cañal-Bruland, Adrian M. Haith

    Motor learning underlies our ability to acquire and perfect movements across all domains of human behaviour. Its study encompasses two distinct but related endeavours: basic research on how we learn movement skills and applied research on how to enhance skill acquisition in real-world contexts. Despite their natural complementarity, these approaches developed historically in two largely independent communities with different disciplinary roots. One is closely associated with computational neuroscience and focuses predominantly on basic research. The other comes from experimental psychology and movement sciences with a stronger focus on applied research. The division between these communities is reflected in the experimental tasks they use, the theoretical frameworks they develop and the cultural norms that shape their training and communication. In this Perspective, we scrutinize this divide, review emerging efforts towards better integration and identify future opportunities. We contend that bridging these gaps can build a more comprehensive science of motor learning.

  • Open Access

    Policy-Gradient Reinforcement Learning as a General Theory of Practice-Based Motor Skill Learning

    by Adrian M Haith

    Mastering any new skill requires extensive practice, but the computational principles underlying this learning are not clearly understood. Existing theories of motor learning can explain short-term adaptation to perturbations, but offer little insight into the processes that drive gradual skill improvement through practice. Here, we propose that practice-based motor skill learning can be understood as a form of reinforcement learning (RL), specifically, policy-gradient RL, a simple, model-free method that is widely used in robotics and other continuous control settings. Here, we show that models based on policy-gradient learning rules capture key properties of human skill learning across a diverse range of learning tasks that have previously lacked any computational theory. We suggest that policy-gradient RL can provide a general theoretical framework and foundation for understanding how humans hone skills through practice.

  • Open Access

    Policy-Gradient Reinforcement Learning as a General Theory of Practice-Based Motor Skill Learning

    by Adrian M. Haith

    Mastering any new skill requires extensive practice, but the computational principles underlying this learning are not clearly understood. Existing theories of motor learning can explain short-term adaptation to perturbations, but offer little insight into the processes that drive gradual skill improvement through practice. Here, we propose that practice-based motor skill learning can be understood as a form of reinforcement learning (RL), specifically, policy-gradient RL, a simple, model-free method that is widely used in robotics and other continuous control settings. Here, we show that models based on policy-gradient learning rules capture key properties of human skill learning across a diverse range of learning tasks that have previously lacked any computational theory. We suggest that policy-gradient RL can provide a general theoretical framework and foundation for understanding how humans hone skills through practice.


  • Dissociable habits of response preparation versus response initiation

    by Yue Du, Adrian M. Haith

    Behaviours we repeat often tend to become habitual. The process of habit formation in humans is commonly studied in the context of learning an arbitrary association between stimuli and responses. Assessing whether or not this association becomes habitual usually involves testing either whether participants can withhold certain responses or whether participants can generate different responses to certain stimuli. However, studies using these two methods have yielded conflicting findings. Here we propose that this discrepancy arises because these approaches target distinct forms of habit associated with different components of action control: response initiation and response preparation, respectively. Through a series of experiments (total n = 215), along with computational models, we show that these two approaches indeed measure dissociable forms of habit. Our results illustrate that a given behaviour can become habitual in multiple, qualitatively different ways, with important implications for how habits can best be promoted or eliminated.


  • Top Cited Papers
    Open Access

    Fundamental processes in sensorimotor learning: Reasoning, refinement, and retrieval

    by Jonathan S Tsay, Hyosub E Kim, Samuel D McDougle, Jordan A Taylor, Adrian Haith, Guy Avraham, John W Krakauer, Anne Ge Collins, Richard B Ivry

    Motor learning is often viewed as a unitary process that operates outside of conscious awareness. This perspective has led to the development of sophisticated models designed to elucidate the mechanisms of implicit sensorimotor learning. In this review, we argue for a broader perspective, emphasizing the contribution of explicit strategies to sensorimotor learning tasks. Furthermore, we propose a theoretical framework for motor learning that consists of three fundamental processes: reasoning, the process of understanding action–outcome relationships; refinement, the process of optimizing sensorimotor and cognitive parameters to achieve motor goals; and retrieval, the process of inferring the context and recalling a control policy. We anticipate that this ‘3R’ framework for understanding how complex movements are learned will open exciting avenues for future research at the intersection between cognition and action.


  • Open Access

    Seemingly "habitual” responses can be caused by both habits and strategic action selection

    by Yue Du, Adrian Haith

    After we have practiced a task extensively, we often struggle to adapt to changes in task requirements, committing frequent habitual “slips of action” where we revert to our originally learned actions. Habitual action selection of this kind is often attributed to an obligatory stimulus-response association which competes with goal-directed responses. An alternative interpretation, however, is that, when task requirements change, we adjust by strategically building on our existing learning. That is, we may first retrieve the action we originally practiced, then switch this to a different action if needed. This retrieve-then-switch strategy may simplify learning and reduce cognitive effort under the altered task, particularly if changes from the original task are minor. To dissociate these two potential mechanisms underlying apparent habitual behavior, we recruited two groups of participants to practice a novel visuomotor association. We assessed habitual responses either by switching the responses for two out of five cues (partial-remapping group), in which case a retrieve-then-replace strategy may be advantageous, or by completely permuting the responses for all five cues (full-remapping group), in which case there would be no benefit to using a retrieve-then-replace strategy. We observed clear habitual behavior in both groups, confirming that participants do form an obligatory habit through practice. We also observed, however, that habitual responses were exacerbated in the partial-remapping group, supporting the view that participants adopted a retrieve-then-switch strategy in this case. Our findings support existing views of habits, but also demonstrate that measures of habitual behavior can be confounded by participant strategies.


  • Action initiation and action inhibition follow the same time course when compared under matched experimental conditions

    by Yue Du, Alexander D. Forrence, Delaney M. Metcalf, Adrian M. Haith

    The time courses of initiating an action and canceling an action have largely been studied in isolation, and little is known about their relationship. Here, we show that when measured under comparable conditions the speeds of action initiation and action inhibition are the same. This finding raises the possibility that these two functions may be more closely related than previously assumed, with potentially important implications for their underlying neural basis.


  • Open Access

    Switching between newly learned motor skills

    by Kahori Kita, Yue Du, Adrian M. Haith

    Studies of cognitive flexibility suggest that switching between different tasks can entail a transient switch cost. Here, we asked whether analogous switch costs exist in the context of switching between different motor skills. We tested whether participants could switch between a newly learned skill associated with a novel visuomotor mapping, and an existing skill associated with an intuitive mapping. Participants showed increased errors in trials immediately following a switch between mappings. These errors were attributable to persisting with the pre-switch policy, rather than imperfect implementation or retrieval of the post-switch policy. A subset of our participants further learned a second new skill. Switching between these two novel skills was initially very challenging, but improved with further training. Our findings suggest that switching between newly learned motor skills can be challenging, and that errors in the context of switching between skills are primarily attributable to perseveration with the wrong control policy.

  • Open Access

    Habitual responses are retrieved from memory, not triggered reflexively

    by Yue Du, Adrian Haith

    Habits are inflexible patterns of behavior that are so well engrained we often can’t help ourselves from performing them. Habitual actions are widely thought to be directly triggered by specific cues via an automatic, reflex-like stimulus–response association, especially if these responses are generated very rapidly. Here we propose, alternatively, that habitual actions might be retrieved from memory. A hallmark of retrieval-based action selection is that the time needed to generate a habitual action should depend on the number of possible responses required – in contrast to more direct and reflexive mechanisms, where the time needed to respond should be invariant. Through a series of experiments, we tested the nature of habitual behavior in a visuomotor association task in human participants. After extensively practicing a novel association, participants developed robust habits. The speed at which these habitual actions were selected increased with the number of potential stimuli/responses, consistent with memory retrieval and inconsistent with a direct and reflexive stimulus–response association. Our results question a commonly-held view of habitual behavior, and help to clarify how the concepts of habit and automaticity relate to one another.


  • Open Access

    Continuous motor skills as flexible control policies: a video game study

    by David M. Huberdeau, Adrian M. Haith, John W. Krakauer

    Many motor skills consist of continuous sequential actions, such as a tennis serve. It is currently unclear how these surprisingly understudied behaviors are learned, with the leading hypothesis being that sequences of single actions become “chunked” into larger single executable units. Under this hypothesis, continuous sequential actions should become more task-specific and less generalizable with practice. To test this, we developed a video game that requires participants to hold a tablet with both hands and steer a virtual car (the “ant car”) along a curving track. We tested participants’ ability to generalize their skill to a probe track that required a different sequence of turns. Across days of practice, task success increased, and movement variability decreased. On the probe track, movement quality at the level of kinematics fully generalized but performance at the level of task success showed a consistent decrement. To address this apparent paradox, we empirically derived the control policy participants used at their maximal skill level on the training track. Notably, this policy was fully transferred to the probe track, but there were more instances of momentary deviations from it (lapses), which explains the worse performance despite equivalent skill. We conclude that continuous motor skills are acquired through learning of a flexible control policy that maps states onto actions and not through chunking or automatizing of a specific sequence of actions.