Bergerot, C. | Search bioRxiv

bioRxiv2023

Moderate confirmation bias enhances collective decision-making

Humans tend to give more weight to information confirming their beliefs than to information that disconfirms them. Nevertheless, this apparent irrationality has been shown to improve individual decision-making under uncertainty. However, little is known about this bias impact on collective decision-making. Here, we investigate the conditions under which confirmation bias is beneficial or detrimental to collective decision-making. To do so, we develop a Collective Asymmetric Reinforcement Learning (CARL) model in which artificial agents observe others actions and rewards, and update this information asymmetrically. We use agent-based simulations to study how confirmation bias affects collective performance on a two-armed bandit task, and how resource scarcity, group size and bias strength modulate this effect. We find that a confirmation bias benefits group learning across a wide range of resource-scarcity conditions. Moreover, we discover that, past a critical bias strength, resource abundance favors the emergence of two different performance regimes, one of which is suboptimal. In addition, we find that this regime bifurcation comes with polarization in small groups of agents. Overall, our results suggest the existence of an optimal, moderate level of confirmation bias for collective decision-making. AUTHOR SUMMARYWhen we give more weight to information that confirms our existing beliefs, it typically has a negative impact on learning and decision-making. However, our study shows that moderate confirmation bias can actually improve collective decision-making when multiple reinforcement learning agents learn together in a social context. This finding has important implications for policymakers who engage in fighting against societal polarization and the spreading of misinformation. It can also inspire the development of artificial, distributed learning algorithms. Based on our research, we recommend not directly targeting confirmation bias but instead focusing on its underlying factors, such as group size, individual incentives, and the interactions between bias and the environment (such as filter bubbles).

animal behavior and cognition↗

bioRxiv2023

Homeostatic Reinforcement Theory Accounts for Sodium Appetitive State- and Taste- Dependent Dopamine Responding.

Seeking and consuming nutrients is essential to survival and maintenance of life. Dynamic and volatile environments require that animals learn complex behavioral strategies to obtain the necessary nutritive substances. While this has been classically viewed in terms of homeostatic regulation, where complex nutrient seeking behaviors are triggered by physiological need, recent theoretical work proposed that such strategies are a result of reinforcement learning processes. This theory also proposed that phasic dopamine (DA) signals play a key role in signaling potentially need-fulfilling outcomes. To examine potential links between homeostatic and reinforcement learning processes, we focus on sodium appetite as sodium depletion triggers state and taste dependent changes in behavior and DA signaling evoked by sodium-related stimuli. We find that both the behavior and the dynamics of DA signaling underlying sodium appetite can be accounted for by extending principles of homeostatic regulation into a reinforcement learning framework (HRRL). We first optimized HRRL-based agents to model sodium-seeking behavior measured in rats. Agents successfully reproduced the state and the taste dependence of behavioral responding for sodium as well as for lithium and potassium salts. We then show that these same agents can account for the regulation of DA signals evoked by sodium tastants in a taste and state dependent manner. Our models quantitatively describe how DA signals evoked by sodium decrease with satiety and increase with deprivation suggesting that phasic DA signals and sodium consumption are down regulated prior to animals reaching satiety. Lastly, our HRRL agents also account for the behavioral and neurophysiological observations that suggest mice cannot distinguish between sodium and lithium containing salts. Our HRRL agents exhibited an equal preference for sodium versus lithium containing solutions, and underestimated the nutritional value of sodium when lithium was concurrently available. We propose that animals use orosensory signals as predictors of the internal impact of the consumed good and our results pose clear targets for future experiments. In sum, this work suggests that appetite-dirven behavior may be driven by reinforcement learning mechanisms that are dynamically tuned by homeostatic need.

neuroscience↗