Opposing roles of the dorsolateral and dorsomedial striatum in the acquisition of skilled action sequencing

The shift in control from dorsomedial to dorsolateral striatum during skill and habit formation is well established, but whether striatal subregions orchestrate this shift co-operatively or competitively remains unclear. Cortical inputs have also been implicated in the shift towards automaticity. Do cortical inputs mirror their downstream striatal targets across this transition? We addressed these questions using a five-step heterogeneous action sequencing task that is optimally performed by automated chains of actions. By optimising automatic responding, we discovered that loss of function in the dorsomedial striatum accelerated acquisition. In contrast, loss of function in the dorsolateral striatum impeded acquisition of sequencing, demonstrating functional opposition within the striatum. Unexpectedly the medial prefrontal cortex was not involved, however the lateral orbitofrontal cortex was critical. These results shift current theories about striatal control of behaviour to a model of competitive opposition, where the dorsomedial striatum acts in a gating role to inhibit dorsolateral-driven behaviour.

Action Grammars of Human Behavior – Measurement, Evolution, and Neural Representation

Human behaviors from tool-making to language are thought to rely on a uniquely evolved capacity for hierarchical action sequencing. Testing this idea will require objective, generalizable methods for measuring the structural complexity of real-world behavior. Here we present a data-driven approach for extracting action grammars from basic ethograms, exemplified with respect to the evolutionarily-relevant behavior of stone tool-making. We analyzed sequences from the experimental replication of ~2.5 Mya Oldowan vs. ~0.5 Mya Acheulean tools, finding that, while using the same “alphabet” of elementary actions, Acheulean sequences are quantifiably more complex and Oldowan grammars are a subset of Acheulean grammars. We illustrate the utility of our complexity measures by re-analyzing data from an fMRI study of stone tool-making to identify brain responses to structural complexity. Beyond specific implications regarding the co-evolution of language and technology, this exercise illustrates the general applicability of our method to investigate naturalistic human behavior and cognition.

Acquisition and Utilization of Mental Imagery Capability in Robotic Action Sequencing Tasks

This work presents a series of neurorobotic models underlying learning in robots. It demonstrates the way in which, during sensorimotor exploration, robots do not just gain knowledge about how to form movement primitives but also obtain a mental imagery capability. Mental imagery plays a key role in these computational models by accelerating learning processes of action sequencing tasks. The first experiment involves permitting a humanoid robot to learn how to retrieve an out-of-reach object using a provided tool. This experiment explores a phenomenon on tool use development found in human infants. In addition, it tests two hypotheses on tool use development. The second experiment extends the domain of robot learning by targeting a useful robotic application. It drives a service robot to learn to acquire knowledge of how to manipulate perceived objects based on the objects’ information and a goal from users. By means of planning, learning the sequence of actions in mind, the robots are able to learn by examining actions’ outcome without really performing actions. This allows the second model to completely exclude parts of overt movements from the training loop. The results confirm that two types of robots can complete their given tasks in a reasonable period of time. The proposed models would benefit robotic applications in terms of online learning.

Spontaneous Melodic Productions of Expert Musicians Contain Sequencing Biases Seen in Language Production

Language production involves complex action sequencing to produce fluent speech in real-time, placing considerable constraints on working memory that lead to sequencing biases in production. Researchers have speculated that these biases may extend beyond language to other human behaviors involving action sequencing, but this claim has not been empirically investigated. Here we provide a strong test of this hypothesis, examining whether biases seen in language production also constrain one of the most complex and spontaneous human behaviors: musical improvisation. Using a large corpus of improvised solo transcriptions from eminent jazz musicians, we test for the existence of an established production bias observed in language production termed easy first—a tendency for more accessible sequences to occur at the beginning of a phrase, allowing incremental planning of more complex phrases. Our analysis shows consistent evidence of easy first in improvised music. We find that the beginning of improvised musical phrases contains 1) more frequently occurring interval patterns, 2) smaller intervals, 3) less interval variety, 4) less pitch variety, and 5) fewer direction changes. There was no easy first bias in a control corpus containing simulated data with the same structure, indicating that the effects are specific to real-time melodic production and not simply due to stylistic conventions. The findings indicate that even expert jazz musicians, known for spontaneous creative performance, reliably retrieve easily-accessed melodic sequences before creating more complex sequences—consistent with an incremental planning strategy employed in language production—suggesting that similar biases constrain the spontaneous production of music and language.

Chemogenetic inhibition in the dorsal striatum reveals regional specificity of direct and indirect pathway control of action sequencing

Animals engage in intricate action sequences that are constructed during instrumental learning. There is broad consensus that the basal ganglia play a crucial role in the formation and fluid performance of action sequences. To investigate the role of the basal ganglia direct and indirect pathways in action sequencing, we virally expressed Cre-dependent Gi-DREADDs in either the dorsomedial (DMS) or dorsolateral (DLS) striatum during and/or after action sequence learning in D1 and D2 Cre rats. Action sequence performance in D1 Cre rats was slowed down early in training when DREADDs were activated in the DMS, but sped up when activated in the DLS. Acquisition of the reinforced sequence was hindered when DREADDs were activated in the DLS of D2 Cre rats. Outcome devaluation tests conducted after training revealed that the goal-directed control of action sequence rates was immune to chemogenetic inhibition—rats suppressed the rate of sequence performance when rewards were devalued. Sequence initiation latencies were generally sensitive to outcome devaluation, except in the case where DREADD activation was removed in D2 Cre rats that previously experienced DREADD activation in the DMS during training. Sequence completion latencies were generally not sensitive to outcome devaluation, except in the case where D1 Cre rats experienced DREADD activation in the DMS during training and test. Collectively, these results suggest that the indirect pathway originating from the DLS is part of a circuit involved in the effective reinforcement of action sequences, while the direct and indirect pathways originating from the DMS contribute to the goal-directed control of sequence completion and initiation, respectively.

Contributions of the basal ganglia to action sequence learning and performance

Animals engage in intricately woven and choreographed action sequences that are constructed from trial-and-error learning. The mechanisms by which the brain links together individual actions which are later recalled as fluid chains of behavior are not fully understood, but there is broad consensus that the basal ganglia play a crucial role in this process. This paper presents a comprehensive review of the role of the basal ganglia in action sequencing, with a focus on whether the computational framework of reinforcement learning can capture key behavioral features of sequencing and the neural mechanisms that underlie them. While a simple neurocomputational model of reinforcement learning can capture key features of action sequence learning, this model is not sufficient to capture goal-directed control of sequences or their hierarchical representation. The hierarchical structure of action sequences, in particular, poses a challenge for building better models of action sequencing, and it is in this regard that further investigations into basal ganglia information processing may be informative.