The role of conjunctive representations in prioritizing and selecting planned actions

What Is Cognitive Control?

The last two decades have seen major advances in cognitive control research. In this paper, I provide an overview of this research. I next make a case that it might benefit from more reflection on its theoretical foundation. I end by suggesting that action theory might be of use with this.

Cognitive Control

Humans and other primates have a remarkable ability to perform a wide range of tasks and behaviors, even novel ones, in order to achieve their goals. Further, they are able to shift flexibly among these behaviors as the contexts demand. Cognitive control is the function at the base of this remarkable behavioral generativity and flexibility. The present review provides a survey of current research on cognitive control focusing on two of its primary features within a control systems framework: (a) the ability to select new behaviors based on context and (b) the ability to monitor ongoing behavior and adjust accordingly. Throughout, the review places an emphasis on how differences in the content and structure of task representations affect these core features of cognitive control.

The representation of abstract goals in working memory is supported by task-congruent neural geometry

Successful goal-directed behavior requires the maintenance and implementation of abstract task goals on concrete stimulus information in working memory. Previous working memory research has revealed distributed neural representations of task information across cortex. However, how the distributed task representations emerge and communicate with stimulus-specific information to implement flexible goal-directed computations is still unclear. Here, leveraging electroencephalography (EEG) and functional magnetic resonance imaging (fMRI) in human participants along with state space analyses, we provided converging evidence in support of a low-dimensional neural geometry of goal information congruent with a designed task space, which first emerged in frontal cortex during goal maintenance and then transferred to posterior cortex through frontomedial-to-posterior theta coherence for implementation on stimulus-specific representations. Importantly, the fidelity of the goal geometry was associated with memory performance. Collectively, our findings suggest that abstract goals in working memory are represented in an organized, task-congruent neural geometry for communications from frontal to posterior cortex to enable computations necessary for goal-directed behaviors.

A transient high-dimensional geometry affords stable conjunctive subspaces for efficient action selection

Flexible action selection requires cognitive control mechanisms capable of mapping the same inputs to different output actions depending on the context. From a neural state-space perspective, this requires a control representation that separates similar input neural states by context. Additionally, for action selection to be robust and time-invariant, information must be stable in time, enabling efficient readout. Here, using EEG decoding methods, we investigate how the geometry and dynamics of control representations constrain flexible action selection in the human brain. Participants performed a context-dependent action selection task. A forced response procedure probed action selection different states in neural trajectories. The result shows that before successful responses, there is a transient expansion of representational dimensionality that separated conjunctive subspaces. Further, the dynamics stabilizes in the same time window, with entry into this stable, high-dimensional state predictive of individual trial performance. These results establish the neural geometry and dynamics the human brain needs for flexible control over behavior.

Practice Reshapes the Geometry and Dynamics of Task-tailored Representations

Extensive practice makes task performance more efficient and precise, leading to automaticity. However, theories of automaticity differ on which levels of task representations (e.g., low-level features, stimulus-response mappings, or high-level conjunctive memories of individual events) change with practice, despite predicting the same pattern of improvement (e.g., power law of practice). To resolve this controversy, we built on recent theoretical advances in understanding computations through neural population dynamics. Specifically, we hypothesized that practice optimizes the neural representational geometry of task representations to minimally separate the highest-level task contingencies needed for successful performance. This involves efficiently reaching conjunctive neural states that integrate task-critical features nonlinearly while abstracting over non-critical dimensions. To test this hypothesis, human participants (n = 40) engaged in extensive practice of a simple, context-dependent action selection task over 3 days while recording EEG. During initial rapid improvement in task performance, representations of the highest-level, context-specific conjunctions of task-features were enhanced as a function of the number of successful episodes. Crucially, only enhancement of these conjunctive representations, and not lower-order representations, predicted the power-law improvement in performance. Simultaneously, over sessions, these conjunctive neural states became more stable earlier in time and more aligned, abstracting over redundant task features, which correlated with offline performance gain in reducing switch costs. Thus, practice optimizes the dynamic representational geometry as task-tailored neural states that minimally tesselate the task space, taming their high-dimensionality.

A Transient High-dimensional Geometry Affords Stable Conjunctive Subspaces for Efficient Action Selection

Flexible action selection requires cognitive control mechanisms capable of mapping the same inputs to different output actions depending on the context. From a neural state-space perspective, this requires a control representation that separates similar input neural states by context. Additionally, for action selection to be robust and time-invariant, information must be stable in time, enabling efficient readout. Here, using EEG decoding methods, we investigate how the geometry and dynamics of control representations constrain flexible action selection in the human brain. Participants performed a context-dependent action selection task. A forced response procedure probed action selection different states in neural trajectories. The result shows that before successful responses, there is a transient expansion of representational dimensionality that separated conjunctive subspaces. Further, the dynamics stabilizes in the same time window, with entry into this stable, high-dimensional state predictive of individual trial performance. These results establish the neural geometry and dynamics the human brain needs for flexible control over behavior.

Flexible gating between subspaces in a neural network model of internally guided task switching

Behavioral flexibility relies on the brain's ability to switch rapidly between multiple tasks, even when the task rule is not explicitly cued but must be inferred through trial and error. The underlying neural circuit mechanism remains poorly understood. We investigated recurrent neural networks (RNNs) trained to perform an analog of the classic Wisconsin Card Sorting Test. The networks consist of two modules responsible for rule representation and sensorimotor mapping, respectively, where each module is comprised of a circuit with excitatory neurons and three major types of inhibitory neurons. We found that rule representation by self-sustained persistent activity across trials, error monitoring and gated sensorimotor mapping emerged from training. Systematic dissection of trained RNNs revealed a detailed circuit mechanism that is consistent across networks trained with different hyperparameters. The networks' dynamical trajectories for different rules resided in separate subspaces of population activity; the subspaces collapsed and performance was reduced to chance level when dendrite-targeting somatostatin-expressing interneurons were silenced, illustrating how a phenomenological description of representational subspaces is explained by a specific circuit mechanism.

Task switch costs scale with dissimilarity between task rules

Cognitive flexibility enables humans to voluntarily switch tasks. Task switching requires replacing the previously active task representation with a new one, an operation that typically results in a switch cost. Thus, understanding cognitive flexibility requires understanding how tasks are represented in the brain. We hypothesize that task representations are cognitive map-like, such that the magnitude of the difference between task representations reflects their conceptual differences: The greater the distinction between the two task representations, the more updating is required. This hypothesis predicts that switch costs should increase with between task dissimilarity. To test this hypothesis, we use an experimental design that parametrically manipulates the similarity between task rules. We observe that response time scales with the dissimilarity between the task rules. The findings shed light on the organizational principles of task representations and extend the conventional binary task-switch effect (task repeat vs. switch) to a theoretical framework with parametric task switches. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Flexible gating between subspaces in a neural network model of internally guided task switching

Behavioral flexibility relies on the brain's ability to switch rapidly between multiple tasks, even when the task rule is not explicitly cued but must be inferred through trial and error. The underlying neural circuit mechanism remains poorly understood. We investigated recurrent neural networks (RNNs) trained to perform an analog of the classic Wisconsin Card Sorting Test. The networks consist of two modules responsible for rule representation and sensorimotor mapping, respectively, where each module is comprised of a circuit with excitatory neurons and three major types of inhibitory neurons. We found that rule representation by self-sustained persistent activity across trials, error monitoring and gated sensorimotor mapping emerged from training. Systematic dissection of trained RNNs revealed a detailed circuit mechanism that is consistent across networks trained with different hyperparameters. The networks' dynamical trajectories for different rules resided in separate subspaces of population activity; the subspaces collapsed and performance was reduced to chance level when dendrite-targeting somatostatin-expressing interneurons were silenced, illustrating how a phenomenological description of representational subspaces is explained by a specific circuit mechanism.

Prospection of Potential Actions during Visual Working Memory Starts Early, Is Flexible, and Predicts Behavior

For visual working memory to serve upcoming behavior, it is crucial that we prepare for the potential use of working-memory contents ahead of time. Recent studies have demonstrated how the prospection and planning for an upcoming manual action starts early after visual encoding, and occurs alongside visual retention. Here, we address whether such "output planning" in visual working memory flexibly adapts to different visual-motor mappings, and occurs even when an upcoming action will only potentially become relevant for behavior. Human participants (female and male) performed a visual-motor working memory task in which they remembered one or two colored oriented bars for later (potential) use. We linked, and counterbalanced, the tilt of the visual items to specific manual responses. This allowed us to track planning of upcoming behavior through contralateral attenuation of β band activity, a canonical motor-cortical EEG signature of manual-action planning. The results revealed how action encoding and subsequent planning alongside visual working memory (1) reflect anticipated task demands rather than specific visual-motor mappings, (2) occur even for actions that will only potentially become relevant for behavior, and (3) are associated with faster performance for the encoded item, at the expense of performance to other working-memory content. This reveals how the potential prospective use of visual working memory content is flexibly planned early on, with consequences for the speed of memory-guided behavior.SIGNIFICANCE STATEMENT It is increasingly studied how visual working memory helps us to prepare for the future, in addition to how it helps us to hold onto the past. Recent studies have demonstrated that the planning of prospective actions occurs alongside encoding and retention in working memory. We show that such early "output planning" flexibly adapts to varying visual-motor mappings, occurs both for certain and potential actions, and predicts ensuing working-memory guided behavior. These results highlight the flexible and future-oriented nature of visual working memory, and provide insight into the neural basis of the anticipatory dynamics that translate visual representations into adaptive upcoming behavior.

The role of conjunctive representations in prioritizing and selecting planned actions

For flexible goal-directed behavior, prioritizing and selecting a specific action among multiple candidates are often important. Working memory has long been assumed to play a role in prioritization and planning, while bridging cross-temporal contingencies during action selection. However, studies of working memory have mostly focused on memory for single components of an action plan, such as a rule or a stimulus, rather than management of all of these elements during planning. Therefore, it is not known how post-encoding prioritization and selection operate on the entire profile of representations for prospective actions. Here, we assessed how such control processes unfold over action representations, highlighting the role of conjunctive representations that nonlinearly integrate task-relevant features during maintenance and prioritization of action plans. For each trial, participants prepared two independent rule-based actions simultaneously, then they were retro-cued to select one as their response. Prior to the start of the trial, one rule-based action was randomly assigned to be high priority by cueing that it was more likely to be tested. We found that both full action plans were maintained as conjunctive representations during action preparation, regardless of priority. However, during output selection, the conjunctive representation of the high-priority action plan was more enhanced and readily selected as an output. Furthermore, the strength of the high-priority conjunctive representation was associated with behavioral interference when the low-priority action was tested. Thus, multiple alternate upcoming actions were maintained as integrated representations and served as the target of post-encoding attentional selection mechanisms to prioritize and select an action from within working memory.