Several studies have found gender differences in Iowa Gambling Task (IGT) performance in which males typically outperform females. However, the precise mechanism that underlies this effect remains unclear, and prior modeling efforts have been unable to pinpoint specific gender differences in behavioral aspects of the IGT. Our results replicated the behavioral gender difference finding and showed that females select the disadvantageous Deck B more than males. We fit the data with versions of the Expectancy Valence and Prospect Valence Learning models that included a parameter to account for participants’ perseverative tendencies. The addition of this parameter to the models led to a substantial improvement in the fit to the data. An examination of the best fitting parameter differences suggests that females give greater weight to recent events than males, which may lead females to discount the large, infrequent losses given by Deck B more and select that option more frequently

Perseveration, or stickiness parameters have been added to reinforcement-learning (RL) models to capture autocorrelationin choices. Here, we systematically examined whether perseveration terms simply improve a models ability to fit noisein the data, thereby making them overly flexible. We simulated data with basic versions of a Delta and Prediction-ErrorDecay model with no perseveration terms added, and for half of the simulated data sets we added random noise to expectedRL values on each trial. We then performed cross-fitting analyses where the simulated data sets were fit by the basicdata-generating models as well as extended models with perseveration terms added. The addition of perseveration termsimproved model fit, particularly when noise was added to the simulation process. Parameter recovery was generally poorerfor the extended models. These results suggest simpler models may be more useful for prediction and generalization tonovel environments, as well as for theory development.

Frequency of reward and average reward value are two types of reward information we utilize when making decisions between two alternative options. Often, these two pieces of information coincide with the highest value option, however, when a slightly less valuable option is presented more frequently, standard reinforcement learning models such as the Delta model can make incorrect predictions. This paper explores the discrepancy in these predictions by way of simulating relevant behavioral tasks with the Delta model, the Decay model, and a novel Bayesian model based on the Dirichlet distribution. We then compare model predictions to behavioral data from some of the same tasks that were simulated. The Delta model provides a poor fit to the data for each of the three presented tasks when compared to the Decay model and the two Bayesian learning models, because it predicts a bias toward options with higher average reward, while the Decay and Bayesian models predict a bias toward reward frequency. The Decay and Bayesian models show a distinct similarity in prediction and fits to the data for most of the tasks. This is because both models predict a bias toward reward frequency rather than average reward magnitude, despite different computational formalisms. However, we also note some interesting discrepancies between the Decay and Bayesian models which will show that in some cases, the frequency of reward may be more important than the reward value.

Decision-making under uncertainty is pervasive. This worksought to understand the role of working memory (WM) in losssensitivity by utilizing two widely used tasks, the IowaGambling Task (IGT) and the Soochow Gambling Task (SGT),and manipulating WM with a dual-task paradigm. Wehypothesized that WM load would reduce attention to both lossvalue and frequency in the decision-making tasks. To betterdelineate the psychological processes underpinning choicebehavior, we developed an Expectancy-Frequency-Perseveration (EFP) model which parsimoniously capturesthree critical factors driving choices: expected value,frequency of gains and losses, and perseveration. Behavioraland computational modeling results indicate that WM loadcompromised performance in the IGT due to reduced attentionto loss value but enhanced performance in the SGT because ofdiminished attention to loss frequency. Our findings suggestthat WM heightens attention to losses, but that greater attentionis given to loss frequency than loss value.

The Delta and Decay rules are two learning rules used to update expected values in reinforcement learning (RL) models. The delta rule learns average rewards, whereas the decay rule learns cumulative rewards for each option. Participants learned to select between pairs of options that had reward probabilities of .65 (option A) versus .35 (option B) or .75 (option C) versus .25 (option D) on separate trials in a binary-outcome choice task. Crucially, during training there were twice as AB trials as CD trials, therefore participants experienced more cumulative reward from option A even though option C had a higher average reward rate (.75 versus .65). Participants then decided between novel combinations of options (e.g, A versus C). The Decay model predicted more A choices, but the Delta model predicted more C choices, because those respective options had higher cumulative versus average reward values. Results were more in line with the Decay model’s predictions. This suggests that people may retrieve memories of cumulative reward to compute expected value instead of learning average rewards for each option.

Reward deficit models of addiction posit weaknesses in reward sensitivity to be promotive of substance dependence, while the externalizing spectrum model views substance problems as arising in large part from a general disinhibitory liability. The current study sought to integrate these perspectives by testing for separate and interactive associations of disinhibition and reward dysfunction with interview-assessed substance use disorders (SUDs). Community and college adults (N = 199) completed a scale measure of trait disinhibition and performed a gambling-feedback task yielding a neural index of reward sensitivity, the 'Reward Positivity' (RewP). Disinhibition and blunted RewP independently predicted SUDs, and also operated synergistically, such that participants - in particular, men - with high levels of disinhibition together with blunted RewP exhibited especially severe substance problems. Though limited by its cross-sectional design, this work provides new information about the interplay of disinhibition, reward processing, and gender in SUDs and suggests important directions for future research.