Forgetting Phenomena in the Iowa Gambling Task: a New Computational Model Among Diverse Participants

Tiancheng Yang¹Chenghan Xie²Xuehe Wang^1*

• ¹Sun Yat-sen University, Guangzhou, China
• ²The Chinese University of Hong Kong, Shatin, Hong Kong Region, China

The Iowa Gambling Task (IGT) is a widely used paradigm for evaluating decision-making and executive functioning in healthy individuals and various clinical populations. Although several existing computational models have been proposed to quantify different aspects of decision-making behavior in the IGT, these models rarely incorporate the forgetting phenomenon, which is critical for understanding dynamic decision-making processes.In this paper, we introduce a novel computational framework-the Exploitation and Exploration with Forgetting (EEF) model, which explicitly integrates a dynamic forgetting parameter into the decision-making process and considers participants' initial preferences for each deck. The EEF model captures not only exploitation and exploration behaviors but also the role of information decay in guiding decisions. Using a dataset of 504 participants, the empirical analyses show that the EEF model provides a better fit than prior models, and its performance is validated through parameter recovery, model recovery, and simulation analyses. Moreover, our study introduces two new behavioral metrics, Sequential Exploration Decay (SED) and Forgetting Interval (FI), which quantify the influence of forgetting on decision-making patterns and validate the model's effectiveness across diverse participant groups. Finally, we extend the EEF model to examine the effects of age and gambling frequency on forgetting and decision strategies, providing novel insights into cognitive changes across the lifespan and in different behavioral contexts. Our findings highlight the importance of forgetting in complex decision-making environments, promoting further research and refinement of computational models to better understand human cognitive dynamics.