Task Structure Modulates Flexibility-Stability Interplay: Cognitive Computational Mechanisms of the List-Wide Proportion Switch Effect

Read full text Summarise with VIRAT
Authors: Linghe Li, Ling Wang
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

1. Research Purpose This study aims to investigate how the structural features of the task environment—specifically, the probability of task switches (operationalized as list-wide proportion switch)—shape the dynamic relationship between cognitive flexibility and cognitive stability. Two core functions of cognitive control—flexibility (the ability to switch between tasks) and stability (the ability to maintain current task goals and resist interference)—have long been debated regarding how they interact. Classical theories posit a trade-off relationship between these two functions. However, recent studies using the list-wide proportion switch paradigm have found that when individuals are placed in a high-switch-probability environment, they exhibit not only reduced switch costs (enhanced flexibility) but also reduced congruency effects (enhanced stability). This finding challenges simple trade-off models, yet the underlying cognitive computational mechanisms remain unclear. By employing computational modeling, this study systematically investigates how task-switch probability modulates the interplay between flexibility and stability, aiming to reveal the computational mechanisms by which the cognitive system configures itself adaptively based on environmental statistics. 2. Research Questions and Hypotheses This study addresses the following core question: How does the cognitive system dynamically adjust its internal configuration based on task-switch probability? Is this adjustment a global strategic shift or a selective optimization of specific processes? Based on this, we propose the following competing mechanistic hypotheses: Hypothesis Core Idea Behavioral Prediction Global Strategy Hypothesis High switch probability makes individuals globally faster and less cautious, reducing decision threshold Both switch costs and congruency effects decrease Process-Specific Hypothesis High switch probability specifically enhances task-set reconfiguration efficiency, increasing drift rate on switch trials Only switch costs decrease; congruency effects remain unchanged Hybrid Hypothesis Both mechanisms operate simultaneously; multiple parameters are modulated Both switch costs and congruency effects change, but in distinct patterns 3. Methods 3.1 Participants We will recruit 20 healthy right-handed adults (aged 18-30 years) with normal or corrected-to-normal vision and no history of neurological or psychiatric disorders. Sample size was determined a priori using G*Power (α = 0.05, 1-β = 0.90, medium effect size). 3.2 Experimental Design A 2 × 2 × 3 within-subjects factorial design will be employed: Task Sequence: Repeat vs. Switch Response Congruency: Congruent vs. Incongruent Switch Proportion: 30% switch, 50% switch, 70% switch (administered on separate days, order counterbalanced) Each condition consists of 6 blocks of 320 trials, totaling 1,920 trials per condition. Practice blocks precede each condition. 3.3 Task Paradigm A cued task-switching paradigm will be used, with the cue and stimulus presented simultaneously (CSI = 0 ms) to maximize reactive control demands. Each trial consists of: fixation (500 ms) → cue and stimulus presented together (max 2000 ms) → blank screen (1000 ms). Task rules: Color Task: Classify the stimulus as warm (red/yellow) or cool (green/blue) Shape Task: Classify the stimulus as curved ("@", "&") or straight ("#", "=") The task cue is presented as a frame (solid frame vs. dashed frame) surrounding the stimulus. Participants respond using the left and right CTRL keys, with mapping counterbalanced across participants. 3.4 Computational Modeling: Diffusion Model of Conflict We will use the Diffusion Model of Conflict as the unified analytical framework. DMC decomposes behavioral data into the following cognitive parameters: Drift Rate (v): The efficiency of information accumulation, reflecting the speed of evidence accumulation during the decision process. Higher values indicate greater information processing efficiency. Decision Boundary (a): The amount of evidence required to make a decision, reflecting response caution. Higher values indicate more cautious decision-making (slower but more accurate responses). Starting Point (z): The offset of the decision starting point, fixed at a/2, meaning the starting point is located at the midpoint of the decision boundary. Starting Point Variability (sz): The trial-to-trial variability of the starting point, reflecting the fluctuation in decision starting point. Non-Decision Time (Ter): The time spent on peripheral processing such as stimulus encoding and motor execution. This parameter reflects the efficiency of preparation and execution processes, including task-set reconfiguration and response selection. Non-Decision Time Variability (st): The trial-to-trial variability of non-decision time, reflecting the stability of peripheral processing. Automatic Activation Peak (A): The peak of the gamma function for automatic activation, reflecting conflict strength. Higher values indicate stronger automatic influence of the distractor dimension on the decision process, and thus greater demand for conflict monitoring. Shape Parameter (alpha): The shape parameter of the gamma function for automatic activation, fixed at 2. Peak Time (tau): The time at which the gamma function for automatic activation reaches its peak, reflecting the temporal dynamics of automatic activation. 3.5 Model Fitting Model fitting will be performed using the Simplex parameter estimation method (Nelder-Mead algorithm). Each participant will be fitted independently to obtain individual-level parameter estimates. Subsequently, parameters will be subjected to 2 (Task Sequence) × 3 (Switch Proportion) repeated-measures ANOVA to test the modulatory effects of switch proportion. 4. Expected Outcomes 4.1 Behavioral Level Based on existing literature, we expect: Reduced switch costs under high switch proportion conditions (enhanced flexibility) Stable congruency effects across switch proportion conditions (stability independent of flexibility modulation) The interaction pattern between task sequence and congruency may vary with switch proportion 4.2 Computational Level Through model comparison, we expect to identify the winning model, thereby revealing the specific mechanism of adaptive regulation 4.3 Theoretical Contribution This study will address the long-standing debate on the relationship between flexibility and stability at the mechanistic level, revealing how the cognitive system dynamically configures itself based on environmental statistics, and provide a computational framework for understanding cognitive control dysfunction in clinical populations.

Fields of Study & Subjects

Neural and Behavioral Psychology Studies Cognitive Abilities and Testing Neurobiology of Language and Bilingualism

Publication details

Publication date
1 January 2026
Language
EN
Document type
Preprint