Stable Representations of Decision Variables for Flexible Behavior

Bari, Bilal A.; Grossman, Cooper D.; Lubin, Emily; Rajagopalan, Adithya E.; Cressy, Jianna I.; Cohen, Jeremiah Y.

doi:10.1016/j.neuron.2019.06.001

Cited by 145 publications

(316 citation statements)

References 89 publications

Supporting

Mentioning

281

Contrasting

Unclassified

Order By: Relevance

“…In our experiments, we noted that the latency to initiate trials increased as the cumulative volume of reward received during a given session increased (Fig.S2A). As such, we used this value as a proxy of task engagement – a reflection of the animals’ motivation or attention towards the current task (Bari et al, 2019; Hamid et al, 2016). We observed no significant discrepancies in this measure between Nrxn1α wildtype and KO mice across varied reward environments (Fig.2A).…”

Section: Resultsmentioning

confidence: 99%

“…Recent evidence suggests choice value can also modulate the vigor with which selected actions are performed (Bari et al, 2019; Hamid et al, 2016). If inefficient choice patterns of Nrxn1α observe changes KOs result from disrupted value encoding, one might expect to also observe changes in how local reward environment modulates vigor of action performance.…”

Section: Resultsmentioning

confidence: 99%

“…The execution of goal-directed behaviors engages numerous cognitive and motor processes – reward-relevant cues from the external world must be accurately represented, internal motivational states must be weighed and these diverse input variables integrated by circuits that control eventual motor output (Doya, 2008). Cortical brain regions including the orbital frontal cortex (OFC), medial prefrontal cortex (mPFC) and anterior cingulate cortex (ACC) represent aspects of reward value and history (Bari et al, 2019; Bartra et al, 2013; Euston et al, 2012; Noonan et al, 2011; Padoa-Schioppa and Conen, 2017; Rushworth et al, 2011; Rushworth et al, 2012), primary sensory cortices and midline thalamic nuclei represent reward-associated environmental signals (Parker et al, 2019; Znamenskiy and Zador, 2013), and motor thalamic nuclei ensure the smooth performance of actions (Diaz-Hernandez et al, 2018). Furthermore, the flexible adaptation of action-associated value signals is necessary in dynamically changing environments and is supported by error-monitoring signals within the ACC and basolateral amygdala, as well as reward prediction errors encoded by striatal-targeting midbrain dopaminergic neurons (McGuire et al, 2014; Schultz et al, 1997; Ullsperger et al, 2014; Yacubian et al, 2006).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Disruption of Nrxn1α within excitatory forebrain circuits drives value-based dysfunction

Alabi¹,

Robinson²,

Fortunato³

et al. 2019

Preprint

View full text Add to dashboard Cite

Goal-directed behaviors, complex action sequences that maximize reward, are essential for normal function and are significantly impaired across neuropsychiatric disorders. Despite extensive associations between genetic mutations and these brain disorders, the mechanisms by which candidate genes contribute to goal-directed dysfunction remains unclear, owing to challenges in (1) describing aspects of reward processing that drive goal-directed dysfunction, (2) localizing these deficits to specific brain circuits and (3) relating changes in physiology to behavioral alterations. Here we examined mice with mutations in Neurexin1α, a presynaptically-localized adhesion molecule with widespread neuropsychiatric disease association, in value-based decision-making paradigms. We found that Neurexin1α knockout animals exhibited blunted choice bias towards outcomes associated with greater benefits. Mutant mice were similarly impaired in avoiding costlier, benefit-neutral actions. Analysis of trial-by-trial choice data via reinforcement learning models suggested these behavioral patterns were driven largely by deficits in the updating and representation of choice values. Employing conditional gene ablation and region-specific Cre-recombinase strains, we revealed that Neurexin1α disruption within forebrain excitatory projection neurons, but not thalamic populations, recapitulated most aspects of the whole-brain knockout phenotype. Finally, utilizing in vivo recordings of direct pathway spiny neuron population calcium activity, we demonstrated that selective knockout of Neurexin1α within forebrain excitatory neurons disrupts reward-associated neural signals within striatum, a major site of feedbackbased learning. By relating deficits in value-based decision-making to region-specific Nrxn1α disruption and changes in reward-associated neural activity, we reveal potential neural substrates for the pathophysiology of neuropsychiatric disease-associated cognitive dysfunction. Genotype: p=0.049 Reward: p<.0001 Interaction: p=0.025 Figure 8 Nex-Control (n=7) Nex-Nrxn1α cKO (n=6) Nex-Control (n=7) Nex-Nrxn1α cKO (n=6) Large Reward (t-1) Small Reward (t-1) * ΔQ*temp -1 (t)

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Disruption of Nrxn1α within excitatory forebrain circuits drives value-based dysfunction

Alabi¹,

Robinson²,

Fortunato³

et al. 2019

Preprint

View full text Add to dashboard Cite

show abstract

“…Our finding that reward context information is represented preferentially in intracortically-projecting neurons suggests that corticocortical networks are likely a key player underlying maintenance of this activity. While not explicitly tested, our observations may aid in providing a transcriptomic and projection-based cellular context to the body of literature that has linked PFC to working memory (Spellman et al, 2015;Kamigaki and Dan, 2017;Bolkan et al, 2017;Schmitt et al, 2017) and the representation of decision variables such as value (Bari et al, 2019;Hirokawa et al, 2019), both of which could be in play during our uncued task.…”

Section: Distribution Of Task Information Across Brain Regions and Cementioning

confidence: 88%

Differential encoding in prefrontal cortex projection neuron classes across cognitive tasks

Lui

Nguyen

Grutzner

et al. 2020

Preprint

View full text Add to dashboard Cite

Single-cell transcriptomics has been widely applied to classify neurons in the mammalian brain, while systems neuroscience has historically analyzed the encoding properties of cortical neurons without considering cell types. Here we examine how specific transcriptomic types of mouse prefrontal cortex (PFC) projection neurons relate to axonal projections and encoding properties across multiple cognitive tasks. We found that most types projected to multiple targets, and most targets received projections from multiple types, except PFC→PAG (periaqueductal gray). By comparing Ca 2+ -activity of the molecularly homogeneous PFC→PAG type against two heterogeneous classes in several two-alternative choice tasks in freely-moving mice, we found that all task-related signals assayed were qualitatively present in all examined classes. However, PAG-projecting neurons most potently encoded choice in cued tasks, whereas contralateral PFC-projecting neurons most potently encoded reward context in an uncued task. Thus, task signals are organized redundantly, but with clear quantitative biases across cells of specific molecular-anatomical characteristics.

show abstract

“…Unfortunately, most often, the effects of reward on decision making and sensory processing are studied separately. This makes it especially difficult to discern how reward feedback is integrated to modulate choice and sensory processing, and this integration is crucial for understanding value-based choice in dynamic tasks (Bari et al, 2019;Donahue & Lee, 2015;Farashahi, Donahue, et al, 2017;Farashahi, Rowe, Aslami, Lee, & Soltani, 2017;Lau & Glimcher, 2007;Soltani & Wang, 2006, 2008Sugrue et al, 2004).…”

Section: Introductionmentioning

confidence: 99%

Separable influences of reward on visual processing and choice

Soltani

Rakhshan

Schafer

et al. 2020

Preprint

View full text Add to dashboard Cite

Primate vision is characterized by constant, sequential processing and selection of visual targets to fixate. Although expected reward and reward value are known to influence both processing and selection of visual targets, the relationship between these effects remains unclear. Using a novel paradigm, we simultaneously measured the effects of expected reward on target selection and sensitivity to visual motion in monkeys. Monkeys freely chose between two visual targets and received a juice reward with varying probability for eye movements made to either of them. Targets were stationary apertures of drifting gratings, causing the endpoints of eye movements to these targets to be systematically biased in the direction of motion. We used this motion-induced bias as a measure of sensitivity to visual motion on each trial and utilized multiple measures to examine global and local effects of reward outcomes on choice and sensitivity to motion. Specifically, we used different reinforcement learning models to fit choice behavior and estimate reward values based on the integration of reward outcomes over multiple trials. Moreover, to compare the effects of reward value on choice and sensitivity to motion directly, we considered correlations between each of these variables and integrated reward outcomes on a wide range of timescales. We found that in addition to choice, sensitivity to visual motion was also influenced by reward value. However, choice was determined by the difference in reward values of the two options whereas sensitivity to motion was influenced by the sum of values. Moreover, models that best predicted visual processing and choice used sets of reward values based on different types of reward integration and timescales.Together, our results demonstrate separable influences of reward value on visual processing and choice, and point to the presence of multiple brain circuits for integration of reward outcomes.

show abstract

Stable Representations of Decision Variables for Flexible Behavior

Cited by 145 publications

References 89 publications

Disruption of Nrxn1α within excitatory forebrain circuits drives value-based dysfunction

Disruption of Nrxn1α within excitatory forebrain circuits drives value-based dysfunction

Differential encoding in prefrontal cortex projection neuron classes across cognitive tasks

Separable influences of reward on visual processing and choice

Contact Info

Product

Resources

About