Understanding and modeling consumers' stylistic taste such as "sporty" is crucial for creating designs that truly connect with target audiences. However, capturing taste during the design process remains challenging because taste is abstract and subjective, and preference data alone provides limited guidance for concrete design decisions. This paper proposes an integrated human-centered computational framework that links subjective evaluations (e.g., perceived luxury of car wheels) with domain-specific features (e.g., spoke configuration) and computer vision-based measures (e.g., texture). By jointly modeling human-derived (consumer and designer) and machine-extracted features, our framework advances aesthetic assessment by explicitly linking model outcomes to interpretable design features. In particular, it demonstrates how perceptual features, domain-specific design patterns, and consumers' own interpretations of style contribute to aesthetic evaluations. This framework will enable product teams to better understand, communicate, and critique aesthetic decisions, supporting improved anticipation of consumer taste and more informed exploration of design alternatives at design time.

People learn from experience, but with considerable individual differences in the degree and type of behavioral adjustments resulting from a given experience. Error driven learning rules provide an elegant framework for explaining both learning behavior and its neural signatures; however, implementing them requires carving the world into so-called “latent states”, that serve as substrates for learning, meaning that the same learning algorithm can produce different sorts of learning given different state representations. Recent theoretical and behavioral work hints that individual differences in learning may reflect differences in how individuals carve their environment into states, with some individuals combining multiple temporal contexts into a single state and others separating these contexts into individuated latent states. Here, we develop a behavioral paradigm and modeling framework to test this idea directly and show in a large cohort of human participants that individuals can be classified into groups according to whether and how they carve temporal contexts into latent states. These behavioral phenotypes impact continual learning, specifically the degree to which individuals avoid interference at context changes or are able to reuse information when encountering a familiar context. We tested whether these behavioral phenotypes related to individual differences in underlying brain connectivity, as measured by resting state-fMRI, but found no evidence for such a relationship. Taken together, this work suggests that learning differences across individuals are attributable to differences in underlying state representations that are not predicted by underlying resting state brain connectivity.

Human decision making can be challenging to predict because decisions are affected by a number of complex factors. Adding to this complexity, decision-making processes can differ considerably between individuals, and methods aimed at predicting human decisions need to take individual differences into account. Behavioral science offers methods by which to measure individual differences (e.g., questionnaires, behavioral models), but these are often narrowed down to low dimensions and not tailored to specific prediction tasks. This paper investigates the use of representation learning to measure individual differences from behavioral experiment data. Representation learning offers a flexible approach to create individual embeddings from data that are both structured (e.g., demographic information) and unstructured (e.g., free text), where the flexibility provides more options for individual difference measures for personalization, e.g., free text responses may allow for open-ended questions that are less privacy-sensitive. In the current paper we use representation learning to characterize individual differences in human performance on an economic decision-making task. We demonstrate that models using representation learning to capture individual differences consistently improve decision predictions over models without representation learning, and even outperform well-known theory-based behavioral models used in these environments. Our results propose that representation learning offers a useful and flexible tool to capture individual differences.

Phone applications to track vehicle information have become more common place, providing insights into fuel consumption, vehicle status, and sustainable driving behaviors. However, to test what resonates with drivers without deep vehicle integration requires a proper research instrument. We built DriveStats: a reusable library (and encompassing an mobile app) to monitor driving trips and display related information. By providing estimated cost/emission reductions in a goal directed framework, we demonstrate how information utility can increase over the course of a 10 day diary study with a group of North American participants. Participants were initially interested in monetary savings reported increased utility for emissions-related information with increased app usage and resulted in self-reported sustainable behavior change. The DriveStats package can be used as a research probe for a plurality of mobility studies (driving, cycling, walking, etc.) for supporting mobile transportation research.

The heterogeneity of outcomes in behavioral research has long been perceived as a challenge for the validity of various theoretical models. More recently, however, researchers have started perceiving heterogeneity as something that needs to be not only acknowledged but also actively addressed, particularly in applied research. A serious challenge, however, is that classical psychological methods are not well suited for making practical recommendations when heterogeneous outcomes are expected. In this article, we argue that heterogeneity requires a separation between basic and applied behavioral methods, and between different types of behavioral expertise. We propose a novel framework for evaluating behavioral expertise and suggest that selective expertise can easily be automated via various machine learning methods. We illustrate the value of our framework via an empirical study of the preferences towards battery electric vehicles. Our results suggest that a basic multiarm bandit algorithm vastly outperforms human expertise in selecting the best interventions.

Prior beliefs are central to Bayesian accounts of cognition, but many of these accounts do not directly measure priors. More specifically, initial states of belief heavily influence how new information is assumed to be utilized when updating a particular model. Despite this, prior and posterior beliefs are either inferred from sequential participant actions or elicited through impoverished means. We had participants play a version of the game “Plinko”, to first elicit individual participant priors in a theoretically agnostic manner. Subsequent learning and updating of participant beliefs was then directly measured. We show that participants hold a variety of priors that cluster around prototypical probability distributions that in turn influence learning. In follow-up experiments we show that participant priors are stable over time and that the ability to update beliefs is influenced by a simple environmental manipulation (i.e. a short break). This data reveals the importance of directly measuring participant beliefs rather than assuming or inferring them as has been widely done in the literature to date. The Plinko game provides a flexible and fecund means for examining statistical learning and mental model updating.

We consider the problem of aligning a large language model (LLM) to model the preferences of a human population. Modeling the beliefs, preferences, and behaviors of a specific population can be useful for a variety of different applications, such as conducting simulated focus groups for new products, conducting virtual surveys, and testing behavioral interventions, especially for interventions that are expensive, impractical, or unethical. Existing work has had mixed success using LLMs to accurately model human behavior in different contexts. We benchmark and evaluate two well-known fine-tuning approaches and evaluate the resulting populations on their ability to match the preferences of real human respondents on a survey of preferences for battery electric vehicles (BEVs). We evaluate our models against their ability to match population-wide statistics as well as their ability to match individual responses, and we investigate the role of temperature in controlling the trade-offs between these two. Additionally, we propose and evaluate a novel loss term to improve model performance on responses that require a numeric response.

Behavior change interventions are important to coordinate societal action across a wide array of important applications, including the adoption of electrified vehicles to reduce emissions. Prior work has demonstrated that interventions for behavior must be personalized, and that the intervention that is most effective on average across a large group can result in a backlash effect that strengthens opposition among some subgroups. Thus, it is important to target interventions to different audiences, and to present them in a natural, conversational style. In this context, an important emerging application domain for large language models (LLMs) is conversational interventions for behavior change. In this work, we leverage prior work on understanding values motivating the adoption of battery electric vehicles. We leverage new advances in LLMs, combined with a contextual bandit, to develop conversational interventions that are personalized to the values of each study participant. We use a contextual bandit algorithm to learn to target values based on the demographics of each participant. To train our bandit algorithm in an offline manner, we leverage LLMs to play the role of study participants. We benchmark the persuasive effectiveness of our bandit-enhanced LLM against an unaided LLM generating conversational interventions without demographic-targeted values.

We present a demonstration system that combines off-the-shelf capacitive screens with easy-to-produce stencils to facilitate the rapid iteration of tactile in-cabin user interfaces. Designers can use this tool to experiment with the position and layout of interactive components; they can 3D print or manually cut stencils out of common material and mount them to the display surface with low-tack glue to create a tactile experience that simulates tactile controls.

With the rising popularity of electrified vehicles, emphasis has been placed on encouraging charging with renewable energy and maximizing battery longevity to improve vehicle sustainability. Many mobile applications offer tools to suggest charging times with more sustainable renewable energy and charging strategies that preserve battery health. However, these options often result in longer, less convenient charging times for drivers. Here we conducted three charging scenario studies to identify factors that influence willingness to wait for sustainable charging. Participants selected between faster but less sustainable charging options and slower charging options that either reduce charging emissions or improve battery longevity. We find people’s willingness to wait for green energy is influenced by situational factors; further we find that information and battery longevity interventions can increase willingness to wait for sustainable charging. Finally, we provide design recommendations to promote sustainably in charging behaviors.

Narrative story generation has gained emerging interest in the field of large language models. The present paper aims to compare stories generated by an LLM only (non-interleaved) with those generated by interleaving human-generated and LLM-generated text (interleaved). The study’s hypothesis is that interleaved stories would perform better than non-interleaved stories. To verify this hypothesis, we conducted two tests with roughly 500 participants each. Participants were asked to rate stories of each type, including an overall score or preference and four facets—logical soundness, plausibility, understandability, and novelty. Our findings indicate that interleaved stories were in fact less preferred than non-interleaved stories. The result has implications for the design and implementation of our story generators. This study contributes new insights into the potential uses and restrictions of interleaved and non-interleaved systems regarding generating narrative stories, which may help to improve the performance of such story generators.

From ride-hailing to car rentals, consumers are often presented with eco-friendly options. Beyond highlighting a "green" vehicle and CO2 emissions, CO2 equivalencies have been designed to provide understandable amounts; we ask which equivalencies will lead to eco-friendly decisions. We conducted five ride-hailing scenario surveys where participants picked between regular and eco-friendly options, testing equivalencies, social features, and valence-based interventions. Further, we tested a car-rental embodiment to gauge how an individual (needing a car for several days) might behave versus the immediate ride-hailing context. We find that participants are more likely to choose green rides when presented with additional information about emissions; CO2 by weight was found to be the most effective. Further, we found that information framing-be it individual or collective footprint, positive or negative valence-had an impact on participants' choices. Finally, we discuss how our findings inform the design of effective interventions for reducing car-based carbon-emissions.

Electrification is an important first step toward reducing the greenhouse emissions of passenger vehicles. However, how drivers drive, charge, and operate their electrified vehicles can have a large impact on their emissions, particularly for Plug-in Hybrid Electric vehicles (PHEVs) that combine all-electric driving with an internal combustion engine. In this paper, we investigate how and why drivers use their PHEVs and uncover design opportunities for interfaces that can support the efficient use of PHEVs. We used a mixed-method approach combining quantitative, qualitative, and concept elicitation methods with PHEV owners in the US. While past findings indicate that PHEV drivers are not motivated to charge regularly, our work contradicts this with evidence of (1) regular charging with home infrastructure, (2) high cost sensitivity, and (3) preference for driving in all-electric mode. Our results indicate that the most critical problem is inadequate user support for navigating poor charging infrastructure.

Auto racing is one of the riskiest sports with potentially significant physical, affective and mental strains on athletes. However, auto racing can offer valuable knowledge and skills that are transferable to regular road driving. The sport poses unique physiological challenges, but they have not been extensively studied, especially as they relate to novice motorsport drivers. In this paper, we report an exploratory study of the physiological arousal experience of drivers in the social context of learning racetrack driving. We find that novice drivers improve speed before steering performance, and that performance is primarily linked to phasic measures of arousal. We discuss these findings and how they inform the design of multimodal affectively-aware driving systems and interventions.

Designers often struggle to sufficiently explore large design spaces, which can lead to design fixation and suboptimal outcomes. Here we introduce DesignAID, a generative AI tool that supports broader design space exploration by first using large language models to produce a range of diverse ideas expressed in words, and then using image generation software to create images from these words. This innovative combination of AI-based capabilities allows human-computer pairs to rapidly create a diverse set of visual concepts without time-consuming drawing. In a study with 87 crowd-sourced designers, we found that designers rated the automatic generation of images from words as significantly more inspirational, enjoyable, and useful than a conventional baseline condition of image search using Pinterest. Surprisingly, however, we found that automatically generating highly diverse ideas had less value. For image generation, the high diversity condition was somewhat better in inspiration but no better in the other dimensions, and for image search it was significantly worse in all dimensions.

Visualizations are common methods to convey information but also increasingly used to spread misinformation. It is therefore important to understand the factors people use to interpret visualizations. In this paper, we focus on factors that influence interpretations of scatter plots, investigating the extent to which common visual aspects of scatter plots (outliers and trend lines) and cognitive biases (people's beliefs) influence perception of correlation trends. We highlight three main findings: outliers skew trend perception but exert less influence than other points; trend lines make trends seem stronger but also mitigate the influence of some outliers; and people's beliefs have a small influence on perceptions of weak, but not strong correlations. From these results we derive guidelines for adjusting visual elements to mitigate the influence of factors that distort interpretations of scatter plots. We explore how these guidelines may generalize to other visualization types and make recommendations for future studies.

Ridesharing is a popular choice for personal transportation needs. Although more ecologically-friendly than single-occupancy vehicles, there is an opportunity to further reduce CO2 emissions by offering green choices. Here we examine whether providing people with information about CO2 emissions nudges them to make more eco-friendly rideshare decisions. Our study tested what kind of information works best to inform people about carbon emissions, comparing direct CO2 values with more relatable carbon equivalents (e.g., trees). We conducted an online study with 1000 participants who picked between regular and eco-friendly ride options that detailed various carbon-output equivalency interventions (e.g., pounds of coal, number of smartphones charged, etc.). We found that participants are more likely to choose a green ride when presented with information about direct CO2 emissions than when presented with carbon-equivalencies. This study aims to inform future information-based interventions more broadly, beyond the context of ridesharing.

Adopting electric vehicles (EVs) is an important step towards meeting climate change targets. Despite the increased availability of electric vehicles (EVs), many individuals are unfamiliar with the environmental and cost savings and how their driving behaviors might change (e.g., where and how to charge) when switching from a conventional fuel vehicle. While behavioral science research can identify what factors are barriers to EV adoption, there is a struggle to identify interventions that can help mitigate these barriers. We introduce EV Life, a mobile app for showing a counterfactual view of people’s automotive behaviors which introduces two functions. First, the app monitors a person’s driving trips in their current vehicle and provides a counterfactual dashboard that highlights what their trip would be like with an EV, including information about cost savings, reduction in carbon emissions, and charging locations. Second, the app provides a research platform for testing interventions for belief change using rule based or machine learning notification delivery.

Full text Figures and data Side by side Abstract Editor's evaluation Introduction Results Discussion Methods Data availability References Decision letter Author response Article and author information Metrics Abstract Inhibition is crucial for brain function, regulating network activity by balancing excitation and implementing gain control. Recent evidence suggests that beyond simply inhibiting excitatory activity, inhibitory neurons can also shape circuit function through disinhibition. While disinhibitory circuit motifs have been implicated in cognitive processes, including learning, attentional selection, and input gating, the role of disinhibition is largely unexplored in the study of decision-making. Here, we show that disinhibition provides a simple circuit motif for fast, dynamic control of network state and function. This dynamic control allows a disinhibition-based decision model to reproduce both value normalization and winner-take-all dynamics, the two central features of neurobiological decision-making captured in separate existing models with distinct circuit motifs. In addition, the disinhibition model exhibits flexible attractor dynamics consistent with different forms of persistent activity seen in working memory. Fitting the model to empirical data shows it captures well both the neurophysiological dynamics of value coding and psychometric choice behavior. Furthermore, the biological basis of disinhibition provides a simple mechanism for flexible top-down control of the network states, enabling the circuit to capture diverse task-dependent neural dynamics. These results suggest a biologically plausible unifying mechanism for decision-making and emphasize the importance of local disinhibition in neural processing. Editor's evaluation This novel theoretical work outlines a unifying architecture for decision-making via disinhibition. The model clearly links observations across multiple empirical studies and highlights how characteristics from previous decision models can be effectively integrated into a single mechanism. This will be of interest to a wide variety of neuroscientists who work across levels of analysis. https://doi.org/10.7554/eLife.82426.sa0 Decision letter Reviews on Sciety eLife's review process Introduction Inhibition is an essential component in neural network models of decision-making. In standard decision models, pools of option-selective excitatory neurons compete in a winner-take-all (WTA) selection process via feedback inhibition (Roach et al., 2023; Wang, 2002; Wong and Wang, 2006). Generally, such inhibition is thought to be homogeneous and non-selective, with a single pool of inhibitory neurons receiving broad excitation, and in turn inhibiting excitatory neurons. However, more recent empirical findings suggest that inhibitory neurons interact with the decision circuit in a more structured manner. Inhibitory neurons active in decision-making exhibit choice-selective activity on par with excitatory neurons in the frontal cortex (Allen et al., 2017), parietal cortex (Allen et al., 2017; Najafi et al., 2020), and striatum (Gage et al., 2010) in contrast to the non-selective or broadly tuned inhibition seen in visual cortex during stimulus representation (Bock et al., 2011; Chen et al., 2013; Hofer et al., 2011; Kerlin et al., 2010; Liu et al., 2009; Niell and Stryker, 2008; Sohya et al., 2007). At an anatomic level, inhibitory interneurons also exhibit a remarkable diversity in morphology, connectivity, and physiological functions (Kepecs and Fishell, 2014; Markram et al., 2004; Tremblay et al., 2016). A prominent circuit motif observed in these anatomical studies is local disinhibition in which vasoactive intestinal peptide (VIP)-expressing interneurons inhibit the neighboring interneurons expressing somatostatin (SST) or parvalbumin (PV) that inhibit dendritic or perisomatic areas in pyramidal neurons, thus locally disinhibiting the activities of the pyramidal neurons in the neighboring area (Chiu et al., 2013; Fino and Yuste, 2011; Fu et al., 2014; Karnani et al., 2014; Karnani et al., 2016; Lee et al., 2013; Letzkus et al., 2011; Pfeffer et al., 2013; Pi et al., 2013; Urban-Ciecko and Barth, 2016). Here, we explore the computational implications of that motif in decision-making. While disinhibitory circuit motifs have been implicated in cognitive processes including learning, attentional selection, and input gating (Fu et al., 2014; Letzkus et al., 2011; Wang and Yang, 2018), how disinhibition functions in decision-making circuits is unknown. Local circuit inputs to the VIP neurons suggest that disinhibition may be a key mechanism for generating the mutual competition necessary for option selection in decision-making. In addition, given the existence of long-range inputs (Kepecs and Fishell, 2014; Lee et al., 2013; Pfeffer et al., 2013; Pi et al., 2013; Schuman et al., 2021) and neuromodulatory inputs (Alitto and Dan, 2012; Fu et al., 2014; Pfeffer et al., 2013; Prönneke et al., 2020; Rudy et al., 2011; Tremblay et al., 2016) to the VIP neurons, local disinhibition has been proposed to play a particular role in dynamic gating of circuit activity; such gating may be essential in decision circuits underlying flexible behavior, mediating top-down control of network function (Fu et al., 2014; Kamigaki, 2019; Lee et al., 2013; Letzkus et al., 2011; Pi et al., 2013; Schuman et al., 2021; Zhang et al., 2014). Here, we hypothesize that disinhibition controls a transition between information processing states, allowing a single decision-making circuit to both represent the values of alternatives and select a single best option amongst those alternatives. Value representation is prominent in the early stage of a decision. Integrated decision variables combine outcome information such as expected gain and probability of realization. Neural firing rates in numerous decision-related brain areas vary with the integrated option values, including the frontal (Kiani et al., 2014; Kim and Shadlen, 1999; Padoa-Schioppa, 2013; Padoa-Schioppa and Conen, 2017; Pastor-Bernier and Cisek, 2011; Roesch and Olson, 2003; Thura and Cisek, 2014; Thura and Cisek, 2016; Yamada et al., 2018) and parietal (Andersen and Buneo, 2002; Churchland et al., 2008; Dorris and Glimcher, 2004; Hanks et al., 2014; Kiani et al., 2008; Kiani et al., 2014; Louie and Glimcher, 2010; Platt and Glimcher, 1999; Roitman and Shadlen, 2002; Rorie et al., 2010; Shadlen and Newsome, 2001; Sugrue et al., 2004) cortices and basal ganglia (Ding and Gold, 2010; Ding and Gold, 2012; Ding and Gold, 2013; Thura and Cisek, 2017). Recent research shows more specifically that neural value coding is contextual in nature, with the value of a given option represented relative to the value of available alternatives (Churchland et al., 2008; Kira et al., 2015; Louie et al., 2011; Louie et al., 2013; Louie et al., 2014; Pastor-Bernier and Cisek, 2011; Rorie et al., 2010; Strait et al., 2014; Yamada et al., 2018). Furthermore, this relative value coding employs a divisive normalization-like representation (Hunt et al., 2012; Louie et al., 2011; Louie et al., 2015; Yamada et al., 2018), a canonical computation prevalent in sensory processing and thought to implement efficient coding principles (Carandini et al., 1999; Carandini and Heeger, 1994; Carandini and Heeger, 2012; Heeger, 1992; Heeger, 1993; Schwartz and Simoncelli, 2001; Silver, 2010) and temporal adaptation (Chau et al., 2020; Heeger, 1992; Louie et al., 2013; Louie et al., 2015; Steverson et al., 2019; Webb et al., 2014). Option selection and categorical choice occur when the decision process progresses beyond simple representation. A common and powerful neural mechanism for this categorical choice is WTA competition (Wickens et al., 2007; Wilson, 2007). WTA dynamics are widely observed in multiple brain regions: the neural firing rate representing the chosen option or action target increases in concert with selection (often reaching an activity threshold at choice), while firing rates representing the other unchosen option are suppressed (Churchland et al., 2008; Gold and Shadlen, 2007; Hanes and Schall, 1996; Hanks et al., 2014; Lo et al., 2015; Lo and Wang, 2006; Roitman and Shadlen, 2002; Rorie et al., 2010; Shadlen and Newsome, 2001; Wang, 2002; Wong and Wang, 2006). The wide prevalence of WTA dynamics in decision-related neural activities suggests that it is a general feature of biological choice. Existing models have identified core circuit motifs that produce either normalized value representation or WTA selection (Figure 1). For normalized value representation, dynamic circuit-based models emphasize a crucial role for both lateral and feedback inhibition (Lofaro et al., 2014; Louie et al., 2014). In the dynamic normalization model (DNM), paired excitatory and inhibitory neurons represent each choice option (Figure 1A); feedforward excitation delivers value inputs, lateral connectivity mediates contextual interactions, and feedback inhibition drives divisive scaling. This simple differential equation model emphasizes the crucial role of lateral connectivity and feedback inhibition in driving empirically observed divisive scaling and contextual interactions (Figure 1B). Figure 1 Download asset Open asset Standard circuit motifs and neural dynamics in existing decision-making models. (A) Dynamic normalization model (DNM). Each pair of excitatory (R) and inhibitory (G) units corresponds to an option in the choice set, with R receiving value-dependent input V and providing output. Lateral interactions implement a cross-option gain control that produces normalized value coding. Panel adapted from Figure 1 from Louie et al., 2014 (B) DNM predicted dynamics replicate empirical contextual value coding. The example task involves orthogonal manipulation of both option values. R1 activity increases with the direct input value V1 (array framed in red) but is suppressed by the contextual input V2 (array framed in blue), consistent with value normalization. (C) Recurrent network model (RNM). The network consists of excitatory pools with self-excitation (1 and 2) and a common pool of inhibitory neurons (I). Panel adapted from Wong and Wang, 2006. (D) RNM predicted dynamics generate winner-take-all selection. The example task involves motion discrimination of the main direction of a random dot motion stimulus with varying coherence (c’) levels (left). Model activity (right) under two different levels of input coherence (0 and 51.2%) predicts different ramping speeds to the decision threshold and generates a selection even with equal inputs. © 2014, Louie et al. Panel B has been reproduced from Figure 5A from Louie et al., 2014 (published under a CC BY-NC-SA 3.0 license). It is not covered by the CC-BY 4.0 license and further reproduction of this panel would need permission from the copyright holder. © 2006, Society for Neuroscience. Panel D (right) is reproduced from Figure 2 from Wong and Wang, 2006 with permission from Society for Neuroscience. It is not covered by the CC-BY 4.0 licence and further reproduction of this panel would need permission from the copyright holder. For WTA selection, the predominant class of decision models (recurrent network models, hereafter RNM) proposes a central role for recurrent connectivity (Houck and Person, 2014; Ito, 2002; Ito, 2006; Ito, 2008; Llinás, 1975; Sathyanesan et al., 2019; Sillitoe and Joyner, 2007) and non-selective feedback inhibition (Wickens et al., 2007; Wilson, 2007; Figure 1C). RNMs capture psychophysical and neurophysiological results in perceptual (Furman and Wang, 2008; Wang, 2002; Wong et al., 2007; Wong and Wang, 2006) and economic (Hunt et al., 2012; Jocham et al., 2012; Rustichini and Padoa-Schioppa, 2015; Soltani and Wang, 2006) choices, recapitulating much of the complex nonlinear dynamics of empirical neurons (Figure 1D). The competitive nature of the RNM generates attractor states which maintain continued activity even in the absence of stimuli, consistent with persistent spiking activity associated with working memory during delay intervals (Brunel and Wang, 2001; Compte et al., 2000; Constantinidis et al., 2018; Furman and Wang, 2008; Hart and Huk, 2020; Lo and Wang, 2006; Macoveanu et al., 2006; Murray et al., 2017; Tegnér et al., 2002; Wang et al., 2013; Wang, 1999, Wang, 2002; Wong and Wang, 2006). While sequential valuation and selection processes may occur independently, electrophysiological evidence shows sequentially coexisting value coding and WTA signals in many prominent decision-related circuits. When decisions are framed as action selection, such integrated representation of values exists primarily in frontoparietal areas tightly linked to motor action commitment. In the control of eye movements, valuation and selection dynamics coexist in multiple brain regions including the lateral intraparietal (LIP) cortex (Louie and Glimcher, 2010; Roitman and Shadlen, 2002; Rorie et al., 2010; Shadlen and Newsome, 2001; Sugrue et al., 2004), the frontal eye fields (Ding and Gold, 2012; Kim and Shadlen, 1999; Roesch and Olson, 2003), and the superior colliculus (Basso and Wurtz, 1997; Basso and Wurtz, 1998; Horwitz et al., 2004; Horwitz and Newsome, 1999; Zhang et al., 2021). In these areas, neural activity initially represents the relevant decision variables but shifts to encode the selected saccade after a WTA-like interval. Similar activity emerges in parallel circuits controlling arm movements, including the parietal reach region (Kubanek et al., 2015; Rajalingham et al., 2014; Snyder et al., 1997), dorsal premotor cortex (Cisek and Kalaska, 2005; Pastor-Bernier and Cisek, 2011; Thura and Cisek, 2016), and primary motor cortex (Thura and Cisek, 2014). Notably, when examined, contextual value coding during a decision typically arises after the initial absolute value coding (Louie et al., 2014; Pastor-Bernier and Cisek, 2011; Rorie et al., 2010), consistent with a local normalization process; these dynamics suggest that normalized value coding is not simply inherited from upstream regions and support coexisting within-region normalization and selection computations. Despite electrophysiological evidence for sequentially coexisting relative value coding and WTA signals in prominent decision-related circuits, no current model integrates both properties within a single circuit. The DNM cannot capture late-stage choice dynamics because it lacks a mechanism for WTA competition. Similarly, RNMs typically neither exhibit contextual value coding nor predict contextual choice patterns (Wang, 2012) due to the lack of structured lateral inhibition. Here, we propose that disinhibition is a biologically plausible solution to unify these key features of decision-making into a single circuit. We develop and characterize a biological circuit consisting of three neuronal types which critically include a form of local disinhibition. This model hybridizes the architectural features of divisive gain control and recurrent self-excitation used in existing models but utilizes disinhibition rather than the commonly assumed pooled inhibition to implement competition. We find that the disinhibition-based model unifies multiple characteristics of decision activity including normalized value coding, WTA choice, and working memory. A top-down gating signal operating via this disinhibition enables the model to switch between the states of value representation and WTA selection and to reproduce decision activity in a range of experimental paradigms with diverse task timing and activity dynamics. These findings suggest that local disinhibition provides a robust, biologically plausible integration of normalization and WTA selection in a single-circuit architecture. Results Local disinhibition decision model To develop an integrated circuit model of decision-making, we systematically tested a series of models incorporating disinhibitory motifs and the core elements of existing models, namely divisive gain control, recurrent excitation, and mutual competition (Figure 2—figure supplement 1; see Methods Motifs tested and compared for normalized coding and WTA choice for the analysis details). This analysis identified local disinhibition as the crucial component that can integrate mutual competition and value normalization within the existing circuit architecture of DNM. In the rest of this paper, outside of the methods and supplementary figures, we focus on this local disinhibition decision model (hereafter LDDM) that emerged from our detailed examination of potential models. In the LDDM (Figure 2A), as in the DNM, option-specific excitatory R units receive value inputs and interact via widespread lateral inhibition. However, the LDDM also includes an option-specific disinhibitory D unit that receives input from its associated excitatory R unit and locally inhibits the inhibitory G unit in the local circuit. In this way, disinhibition biased by different value inputs can serve to selectively release local circuit gain control, generating an unbalanced gain control between local and opponent circuits and leading to a WTA competition. In this model, the network thus shifts from value coding to WTA competition regimes in response to the onset of disinhibition (controlled by the coupling strength between R and D). With zero or weak R-D coupling, the circuit preserves normalized value coding consistent with the DNM; with strong R-D coupling, the circuit switches to a state of WTA selection (Figure 2B). Inhibitory units, as a result, dynamically switch from a non-selective response pattern to a selective response pattern (G and D units in Figure 2B) driven by local disinhibition. This flexible onset of disinhibition is modeled after biological findings, which show that activation of disinhibition in cortical circuits arises from exogenous, long-distance projections (Fu et al., 2014; Kamigaki, 2019; Lee et al., 2013; Pi et al., 2013; Zhang et al., 2014; Figure 2C). This form of top-down control allows for flexibility in the relative timing of the valuation and selection processes, consistent with neural and behavioral data in different task paradigms (see Gated disinhibition provides top-down control of choice dynamics). Figure 2 with 1 supplement see all Download asset Open asset Local disinhibition decision model (LDDM) and its biological plausibility. (A) LDDM extends the dynamic normalization model (DNM) by incorporating a disinhibitory D unit to mediate the local disinhibition of the associated excitatory R unit; strength of R to D coupling is controlled by the parameter β presumed via an external top-down control. Vi , α, and ω indicate the corresponding input value to each option, self-excitation of R unit, and the coupling weights from R to G unit, respectively. (B) The network phase transition between representation and choice under gated disinhibition. With the disinhibitory module silent, the network performs dynamic divisive normalization on R units and predicts non-selective inhibition via G units; after the disinhibitory module is triggered via an external top-down control signal, the network switches to a winner-take-all competition dynamic. The circuit predicts selective inhibition after disinhibition is triggered. (C) Biological basis of disinhibition. Disinhibition provides a mechanism for dynamic gating of circuit states. Vasoactive intestinal peptide (VIP)-expressing interneurons typically inhibit somatostatin (SST) and parvalbumin (PV)-positive interneurons, resulting in a disinhibition of pyramidal neurons. VIP neurons receive local, long-range, and neuromodulatory input, providing different potential mechanisms to modulate local circuit dynamics. © 2014, Springer Nature. Panel C is reproduced from Figure 3 from Kepecs and Fishell, 2014 with permission from Springer Nature. It is not covered by the CC-BY 4.0 licence and further reproduction of this panel would need permission from the copyright holder. Activity dynamics of the LDDM are described by a set of differential equations: (1) τRdRidt=-Ri+Vi+αRi+BR1+Gi, (2) τGdGidt=-Gi+∑j=1NωijRj+BG-Di, (3) τDdDidt=-Di+βRi. where i=1, …, N designates choice alternatives, each of which is represented by an R unit receiving selective input Vi and non-selective baseline input BR. τR, τG, and τD are the time constants for the R, G, and D units. The weights ωij represent the coupling strength between excitatory units Rj and inhibitory (gain control) units Gi, with each G unit driven by a weighted sum of excitatory inputs from all R units and a non-selective baseline input BG and inhibited by its local Di; the parameter α reflects the strength of recurrent self-excitation on R units. Finally, β weights the coupling strength between the excitatory Ri and the disinhibitory Di units and is presumed to be under external (task-triggered) control. Dynamic divisive normalization preserved in the LDDM We first examine whether the LDDM retains the dynamics of divisively normalized value coding seen empirically and in the DNM (Lofaro et al., 2014; Louie et al., 2014). As discussed above, during the initial option evaluation, the disinhibitory units are silent (β=0); therefore, the sole difference between the LDDM and the DNM is recurrent excitation (controlled by α). Example activity traces in Figure 3B show that the LDDM preserves characteristic early-stage dynamics and contextual modulation seen in both empirical data (Figure 3C) and the original DNM (Lofaro et al., 2014; Louie et al., 2011; Louie et al., 2014). Immediately after stimulus onset, R1 activities replicate the transient peak observed in a wealth of studies (Andersen and Buneo, 2002; Churchland et al., 2008; Gnadt and Andersen, 1988; Louie et al., 2011; Louie et al., 2014; Platt and Glimcher, 1999; Rorie et al., 2010; Sugrue et al., 2004). Furthermore, the network settles to equilibrium displaying relative value coding: R1 activity increases with V1 and decreases with V2, reflecting a contextual representation of value (Figure 3B, R1 activity across V1 inputs [upper panel] and V2 inputs [bottom panel]). Figure 3 Download asset Open asset Normalized value coding in the local disinhibition decision model (LDDM). (A) In this example, the LDDM receives a set of two input values with varying V1 (framed in red) and V2 (framed in blue). (B) Example LDDM dynamics show relative value coding. R1 activity shows a transient peak before a sustained period of coding. Increasing V1 increases R1 activity but increasing V2 decreases R1 activity. (C) Value coding dynamics recorded in monkey parietal cortex. The model prediction we showed is consistent with the empirical observation. Panel is adapted from Figure 1B and D from Louie et al., 2011. (D) Phase plane analysis of the system under equal (left), weakly unequal (middle), and extremely unequal (right) inputs. The nullclines of R1 (solid) and R2 (dashed) indicating the equilibrium state of the individual units intersect at a unique and stable equilibrium point with divisively normalized coding. Taking advantage of its simplified mathematical form, we analytically evaluated the LDDM by conducting phase plane analyses. We found that it represents each set of input values (V1,…,VN) as one unique and stable equilibrium point in its output space (R1,…,RN) when β=0. Specifically, we solved for the equilibrium state of each R unit by setting each differential equation (Equations 1–3) to zero, which defines the nullcline of each R unit as a function of the activity of the complementary R unit, visualized in Figure 3D. The nullclines of R1 (solid) and R2 (dashed) intersect at a unique equilibrium point, regardless of whether input values are equal or unequal (see different panels for examples of different inputs). This point indicates that the dynamical system, when receiving any positive inputs, can maintain a unique equilibrium where every unit maintains a steady level of activity. Linearization analysis around this point suggests that this point is attractive: given any initial values to the system, the activities of the units will converge into the unique equilibrium point for the network (see Methods Equilibria and stability analysis of the LDDM for mathematical proof). The steady state of neural activity at equilibrium (noted as Ri*) reflects divisive normalization (Equation 4), as in the original DNM (Lofaro et al., 2014; Louie et al., 2014). The only difference between the LDDM and the DNM at equilibrium is the introduction of a constant in the denominator (BG-α) representing baseline gain control and recurrent excitation; this change rescales the activity magnitudes but preserves normalized value coding. (4) Ri*=Vi+BR1+BG-α+∑j=1NωijRj* We next verified that the normalized value coding produced by the LDDM cannot be implemented by standard RNM models. Figure 4A compares the activity of R1* as a function of both value inputs (V1 and V2) in the LDDM (left panel), the original DNM (middle panel), and the RNM (right panel). Both the LDDM and the DNM exhibit R1* activities (indicated by color) that monotonically increase with input V1 but decrease with V2, with a slightly steeper V2 dependence in the LDDM versus the DNM model depending on the rescaling of α. In contrast, strong WTA dynamics in the RNM implement categorical (choice) coding rather than relative value representation, with high or low coding of input values (right panel). Figure 4 with 2 supplements see all Download asset Open asset Quantitative comparison of contextual value coding across the local disinhibition decision model (LDDM), dynamic normalization model (DNM), and recurrent network model (RNM) models. (A) Comparison between the LDDM (left), the DNM (middle), and the RNM (right) in value coding. The LDDM and the DNM show normalized value coding. The neural activity of R1 (indicated by color) increases with the direct input V1 but decreases with the contextual input V2. The LDDM shows slightly stronger contextual modulation than the DNM but qualitatively replicated normalized value coding. The RNM shows a qualitatively different pattern consistent with winner-take-all (WTA) competition. Within the regime of WTA competition (V1 and V2 within a reasonable scale), R1 activity is high when V1 > V2 and low when V1 < V2. (B) Fitting the models to a trinary choice dataset shows that the LDDM (left panel) performed slightly better than the DNM (middle panel) in capturing the neural activities responding to values inside (Vin) and outside (Vout) of the receptive field. Fitting the RNM to the dataset does not capture the neural activities as well as the LDDM (and DNM; right panel). To quantitatively test value normalization, we fit the models to observed firing rates of monkey LIP neurons under varying reward conditions (Louie et al., 2011). In the empirical data (Figure 4B, dots), LIP activity increases with the reward (water quantity) associated with the target inside the neuronal response field (Vin) and decreases with the summed rewards of targets outside the response field (Vout). The fitting results show that the DNM captures the rescaled firing rates very well with only two free parameters (baseline input BR = 70.92 and an arbitrary scaling parameter Rmax; see Methods; middle panel in Figure 4B, R2=0.9640). The LDDM with an additional parameter BG-α introduced by self-excitation and baseline gain control fitted slightly better than the DNM (BR =71.53, BG-α=3.82; see Methods; left panel in Figure 4B, R2=0.9646; parameter recovery analysis shows that the LDDM is highly robust in the data fitting, Figure 4—figure supplement 1). Note that fitting to the current dataset is not able to differentiate the contributions of α and BG to the neural dynamics (see proof in Methods); thus more empirical data will be needed to draw conclusions about the role of recurrent self-excitation in value coding. However, we do show below that self-excitation is critical for generating persistent activities (see section: Disinhibition controls point versus line attractor dynamics in persistent activity). We found that fitting the standard RNM with its standard four parameters (see Methods) cannot capture the pattern of neural activity as well as the LDDM and DNM (right panel in Figure 4B; R2=0.8920). This small but clear difference in performance between model classes arises from the difference between divisive (DNM and LDDM) and subtractive (RNM) types of inhibition, with subtractive inhibition failing to capture the concave contextual effects predicted by divisive models. Furthermore, fitting the RNM to the data results in a parameter regime that can no longer generate WTA competition; instead, the model predicts mean firing rates in a low-activity regime with a maximum value of 3.5 Hz (Figure 4—figure supplement 2). These results suggest that RNM models cannot simultaneously support both normalized value coding and WTA selection regimes. Local disinhibition drives WTA competition A key question is whether the LDDM can also produce WTA competition. Given the architecture of the LDDM, local disinhibition is hypothesized to break the symmetry between option-specific R-G sub-circuits, enabling a competitive interaction between sub-circuits. To examine whether this competition produces WTA selection, we simulated model activity in a reaction-time version of a motion discrimination task, a standard perceptual decision-making paradigm in non-human primates (Churchland et al., 2008; Roitman and Shadlen, 2002). The task contains two stages of processing: the pre-motion stage with only the choice targets presented and the motion stage presenting a random-dot motion stimulus simultaneously with a go signal