Master of Integrative Biology Neuroscience Specialization UE 5NB04 INFORMATION PROCESSING IN THE PREFRONTAL CORTEX FOR DECISION MAKING Sandrine Duverne Paris, Dec 02, 2019
Decision making A SERIES OF INTERWEAVED CHOICES IN EVER-CHANGING ENVIRONMENT
Perspective of cognitive sciences Decision making refers to a set of cognitive processes that results in the selection of a belief or a course of action among several alternative possibilities.
Characteristics of decision making Decisions are : either automatic/uncontrolled or voluntary/controlled. either irrational/intuitive/emotional or rational/thoughtful/effortful Based on both knowledge (subjective values, preferences, beliefs) and information in the immediate environment (context, instructions, rewards). dynamic and temporally intermixed.
How the brain makes decision THE PREFRONTAL CORTEX ORCHESTRATES DECISION MAKING
The prefrontal cortex A very large part of the brain that plays a key role in high- level cognitive processes involved in goal-driven behaviors
Cross-species evolution of the PFC • PFC grey matter in humans is up to 1.9-fold greater than in macaques and 1.2-fold greater than in chimpanzees • Subcortical white matter underlying the PFC in humans is 2.4-fold greater than in macaques and 1.7-fold greater than in chimpanzees (Donahue et al., 2018) Human cognitive uniqueness should therefore focus less on the frontal lobes in isolation and more on distributed neural networks. (Barton & Venditti, 2013)
Two approches of PFC functions in decision making MODULAR APPROACH CIRCUIT-BASED APPROACH localization distribution Goal-directed choices can be Goal-directed choices emerge from subdivided into a set of discrete repeated computations that are component processes with serial and distributed across many brain regions localized neural implementation. that perform similar computations. Hunt & Hayden, Nature Rev Neurosci, 2017
Tools and techniques Connectivity modelling Temporal resolution Spatial resolution static functional connectivity EEG, MEG fMRI, PET scans dynamic functional connectivity
The modular approach of PFC functions MODULES OF INFORMATION PROCESSING
Localization of cognitive functions in the brain An old hypothesis recycled to study the organization of cognitive functions in the brain. Fodor, 1983 Gall & Surzheim , 1810
Modularity of the mind Modules are : Domain specific: dedicated to a specific The mind is composed of type of process or computation composed of neural structures Encapsulated: each module refer to (modules) with distinct other modules evolutionarily-developed Associated with a specific neural functions. structure Their output converge and are centralized in a central system Fodor, The Modularity of Mind , 1983
Decision making in the modular approach A series of component processes : evaluation of options Each component process is assumed to correspond to comparison of option values discrete neural computations selection of an appropriate action implemented in distinct plan neural modules . monitoring of the outcome of the choice Hunt & Hayden, Nature Rev Neurosci, 2017
Three main functional axes Lateral PFC Each PFC axis Control system operates along distinct Dorso-Medial PFC Motivational system dimensions of information Ventral PFC processing Valuation system
COGNITIVE CONTROL IN LATERAL PFC Hierarchical processing of information
Cognitive Control Set of processes that allow information processing and behavior to vary adaptively from moment to moment depending on current goals and environmental constrains Attention Memory Problem solving • Selective • Action selection • Working attention memory • Conflict • Inhibitory resolution • Episodic attention memory • Error detection • Shifting • Prospective • Planning attention memory
Hierarchical organization of PFC functions The frontal and posterior parts of the brain are two networks that are hierarchically organized: • Frontal system = motor memory • Dorsal system = perceptual memory PFC functions are hierarchically organized in an anterior-posterior axis to process abstract to more concrete information. Fuster, Trends NeuroSci , 1997
The notion of task set Task sets : a configuration of cognitive processes that is actively maintained for subsequent task performance Dorsolateral PFC functions are organized to selectively represent, update and implement a specific task Competition between neural representations of task sets during task switching Facilitate task performance by establishing context-dependent rules Sakai, Ann Rev Neurosci , 2003 Collins & Frank, PLoS Comput Biol , 2016
Cascade of cognitive control processing Abstraction subsets of actions are linked with contecxtual cues to form task sets • Sensory control : select actions based on stimulus-response associations • Contextual control : select task set based on contextual cues Temporal processing ensemble of task sets are maintained and temporally extended • Episodic control : select ensembles of task set based on temporal episode (frequencies of cues) • Branching control : monitor alternative task sets = not currently performed but can be quickly reactivated Koechlin & Summerfield, Trends Cog Sci , 2007
Take home message Cognitive control in lateral PFC • Hierarchical organization • Contextual control : abstraction of task set based on contextual cues • Temporal control : maintenance of task sets during episodes with repeated cues
VALUATION SYSTEM IN VENTRAL PFC Reward-based choices
Valuation system Set of processes that allow information processing and behavior to vary adaptively from moment to moment depending on available rewards in the environment . Reward Subjective values Social cognition • Part of the • Emotion • Expected values dopaminergic and reward- • Self-estim and self- system with direct probabilities representation projection from associated with • Theory of mind the ventral stimuli striatum • Common currency across stimuli
Ventromedial PFC & value-based choices 4-arm bandit choice task Reward-related activations Markov Decision Process to maximize expected gain by choosing vmPFC activations linearly increase with rewads and choices between competing choices associated with stockastic rewards Daw et al., Nature 2006
Dopaminergic system and reinforcement learning Dopaminergic loops Reinforcement learning and dopaminergic loops Doya, HFSPJ 2007
PFC vs Striatum MODEL-FREE DECISION IN STRIATUM MODEL-BASED DECISION IN PFC • Estimation of cache values with • Estimation of the state and reward temporal-difference learning transitions • Actions are selected on the basis of • A knowledge of action-outcome stored value representations relationships is used to anticipate long- run consequences of candidate actions • Long-term values are estimated directly • Long-term reward probabilities are from experience estimated by iteratively searching through a decision tree Daw et al, Nature Neurosci 2005
Model-free and model-based choices Decision tree paradigm State and reward prediction error State prediction correct signal s equential two-choice Markov decision in striatum vs lateral PFC in striatum and vmPFC State-based > standard RL Glascher et al., Neuron 2010 Hampton et al., J Neurosci 2006
Take home message Valuation system in ventral PFC • Reward dependent • Value-based model of the environment • Signature of choice value • abstract model of task structure to guide behavioral choices : combines model-free striatal signals and model-based lateral PFC signals
MOTIVATIONAL CONTROL IN DORSO-MEDIAL PFC A key interface between reward and control systems
Dorso-medial PFC functions Mediating role of the dorsal anterior cingulate cortex (dACC) as building the links between rewards and actions Reward-based Cognitive control Effort decisions • Error monitoring • Volatility of rewards • Sustained effort • Conflict detection • Regulate alertness • Unchosen task values • Motivational • Speed-accuracy control • Foraging tradre-off • Task switching • Choice difficulty • Regulation of • Preference signals cognitive control
Dorsal ACC & Cognitive control Error monitoring Conflict detection Hierarchical motivation van Schie et al, 2004 Holroyd & Coles, 2002 Carter & van Veen, 2008 Kouneiher et al, 2009
Dorsal ACC & value-based choices Volatility of reward Foraging ACC learns the value of information in an uncertain world ACC encodes the average value of the foraging environment and cost of foraging Kolling et al., Science , 2012 Behrens et al., Nature Neurosci 2007
Dorsal ACC as selector or detector of alternative choices ACC activations scale with choice difficulty ACC activations signal the need to switch and depart from the default choice Shenhav et al., Nature Neurosci 2014 Donoso et al., Science 2014
Take home message Motivational control in Dorso- medial PFC • Interface between cognitive control and valuation system • Signature of the need to increase cognitive control • Signature of difficulty to depart from default choice
SUMMARY OF THE MODULAR APPROACH Strenghts and limits
Recommend
More recommend
