| In today’s society,probability reasoning ability is very important for people to reason and make decisions on major events,such as health care,insurance investment and so on.However,psychological research shows that people often have cognitive bias and thinking errors in probabilistic reasoning.In order to explain the reasons for these deviations and fallacies,researchers have put forward many theories,and the most representative ones are the Ecological Rationality Framework and the Nested Set Theory ".Among them,the Ecological Rational Framework" emphasizes that people will perform better in probabilistic reasoning as long as the problem can be put forward in a way consistent with the real environment,nature and evolutionary ecology;while the Nested Set Theory holds that a clear problem set structure can effectively activate the analysis system,which can analyze and reason with the help of working memory resources to obtain the correct answer.In other words,the Ecological Rationality Framework emphasizes ecological adaptation and heuristic algorithms,while the Nested Set Theory emphasizes the role of the analysis system and working memory resources.It can be seen that the core of the dispute between the two is the role of working memory in probabilistic reasoning.So,does working memory work in the process of probabilistic reasoning? And how do working memory resources,sub components and their functions work? Existing studies have not been systematically discussed.Recently,the Tripartite Model of Mind proposed by Stanovich(2015)divides the human mind into autonomous mind,algorithmic mind and reflective mind,and believes that analytical processing occurs at the level of reflective mind and algorithmic mind.The model emphasizes that "human beings are cognitive misers".When analytical processing fails to successfully suppress the inferior autonomous response of heuristic processing,it will lead to thinking fallacies.Then,working memory,the general cognitive ability and cognitive resources of the algorithmic mind,how to exercise the suppression function of poor response to the autonomous mind in the reasoning process,and how to guide the metacognitive process of algorithm operation,control the expenditure of cognitive resources,and then make correct reasoning under the action of reflective mind? Based on the previous probabilistic reasoning research and the "triple processing mind" model,this study attempts to explore whether working memory resources and sub components of working memory play a role in probabilistic reasoning,at which stage,and how to play a role,so as to reveal the mechanism of working memory.In Study 1,two parallel experiments were conducted to explore the performance of working memory resources in different difficulty probabilistic reasoning tasks under empirical paradigm and text paradigm;Study 2 also explored the role of sub-components of working memory in probabilistic reasoning under text paradigm and experience paradigm through two parallel experiments;In Study 3,on the basis of study 1 and 2,we further investigated the co-regulatory role of inhibitory function and metacognitive ability in the process of working memory affecting probabilistic reasoning;In the fourth study,eye movement technology was used to investigate the fixation characteristics of subjects on probabilistic information in the process of probabilistic reasoning,and the mediating role of this feature in the relationship between working memory span and probabilistic reasoning performance.At the same time,the co-regulatory role of metacognitive ability and inhibition function in this mediating process was investigated.Specifically,study 1 investigated the performance differences of subjects with high and low working memory span in probabilistic reasoning tasks with different problem difficulties under different paradigms.Among them,experiment 1 took 118 college students as subjects under the condition of text paradigm,grouped the working memory span through the letter span task,and investigated the performance differences of subjects with different working memory span in solving probabilistic reasoning problems(PRS)and Bayesian reasoning problems.Experiment 2 took urn-ball task as the empirical paradigm to investigate the performance differences of 118 college students with different working memory span when solving different difficult tasks.The results show that no matter which paradigm,the reasoning performance of individuals with a high working memory span is significantly better than that of individuals with a low working memory span,and the performance in simple tasks is significantly better than that in difficult tasks.The difference is that in the text paradigm,there is no difference between the subjects with high and low working memory span in reasoning performance in simple tasks,and the subjects with high working memory span have better probability reasoning performance than those with low working memory span in difficult tasks;In the empirical paradigm,there was no difference between subjects with high and low working memory span in difficult tasks.In simple tasks,subjects with high working memory span had better probability reasoning performance than those with low working memory span.Study 2 investigated the role of working memory sub-components in probabilistic reasoning under different paradigms.Among them,Experiment 3 used the dual-task paradigm under the text paradigm to suppress the sub-components of working memory,speech circuit,visuospatial sketchpad and central executive system of 124 college students through pronunciation task,percussion task and random interval generation task respectively,and investigated the difference of probability reasoning performance between the experimental group and the control group;The difference between experiment 4 and Experiment3 is that 124 college students were investigated under the empirical paradigm,and the others are the same as experiment 3.The results show that:(1)under the text paradigm,the probabilistic reasoning performance of the three sub component suppression groups is lower than that of the control group,and the probabilistic reasoning performance of the phonological loop suppression group is lower than that of the central execution system suppression group and the visuospatial sketchpad suppression group;(2)Under the empirical paradigm,the probabilistic reasoning performance of the phonological loop suppression group was significantly lower than that of the control group,the central executive system suppression group and the visuospatial sketchpad suppression group.There was no difference between the control group and the central executive suppression group and the visuospatial sketchpad suppression group.Study 3 examined the mechanism of working memory in the process of probabilistic reasoning.In sub-study 1,116 college students were tested by color word Stroop task and letter span task to explore the regulatory role of central executive system inhibition function in the relationship between working memory span and probabilistic reasoning performance;On the basis of sub-study 1,in order to further investigate the role of reflective mind in guiding algorithm operation and controlling cognitive resource expenditure,sub-study 2 measured college students’ metacognitive ability with the metacognitive scale for 115 college students,and systematically detected the co-regulatory effect of metacognitive ability and inhibition function on the relationship between working memory and probabilistic reasoning performance.The results showed that:(1)inhibition played a regulatory role in the relationship between working memory span and probabilistic reasoning performance.The higher the inhibition function of subjects,the greater the impact of working memory span on probabilistic reasoning performance;(2)Metacognitive ability and inhibitory function play a common regulatory role in the relationship between working memory span and probabilistic reasoning performance.Not only the higher the metacognitive ability of the subjects,the greater the impact of working memory span on the performance of probabilistic reasoning;Moreover,metacognitive ability regulates the regulation of inhibitory function,that is,the higher the metacognitive ability,the greater the regulation of inhibitory function.Study 4 revealed the subjects’ attention characteristics to information in the process of probabilistic reasoning and its role in the influence of working memory span on the performance of probabilistic reasoning.Taking the saccade count as the eye movement characteristic index and 40 college students as subjects,their metacognitive ability was measured by metacognitive scale,their inhibitory function was measured by Stroop Color-Word task,and their probabilistic reasoning scores were measured by probabilistic reasoning questionnaire(PRS)and Bayesian questions.The results showed that the saccade count,the fixation characteristic index of subjects’ probabilistic reasoning information,played a partial mediating role in the impact of working memory span on probabilistic reasoning performance.Working memory inhibition and metacognitive ability jointly regulate the first half of the mediating model path and the direct path.The higher the inhibitory function and metacognitive ability,the greater the effect of working memory span on the saccade count;Metacognitive ability also regulates the regulation of inhibitory function.The higher the metacognitive ability,the greater the regulatory effect of the inhibitory function.Based on the above,this study draws the following conclusions:(1)Whether text paradigm or experience paradigm,working memory span can affect college students’ probabilistic reasoning performance.This shows that the process of probabilistic reasoning needs the participation of working memory resources.(2)In text paradigm,phonological loop plays the largest role in probabilistic reasoning,followed by the central execution system and visual space template.This shows that working memory mainly plays a role in the information representation stage of probabilistic reasoning;In the empirical paradigm,the visual space panel plays the largest role in probabilistic reasoning,and the phonological loop and central execution system hardly play a role.It shows that working memory mainly plays a role in the information representation stage in the process of probabilistic reasoning,and hardly plays a role in the solution stage.(3)Inhibition function plays a moderating role between working memory span and probability reasoning performance of college students;Metacognitive ability can regulate the function of cross-level regulation and inhibition;Metacognitive ability and inhibitory function jointly regulate the effect of working memory span on probabilistic reasoning.This shows that inhibition function can adjust the working memory resources are applied to probabilistic reasoning.The stronger the inhibition function,the more working memory resources applied to probabilistic reasoning.This effect of inhibition function is affected by metacognitive ability.(4)Working memory span affects the performance of probabilistic reasoning,which is mediated by individual eye movement characteristics and saccade count of probabilistic information.Working memory inhibition and metacognitive ability jointly regulate the effect of working memory span on probabilistic information attention characteristics and the effect of working memory span on probabilistic reasoning performance.That is,the individual’s working memory span affects his fixation and acquisition of probability information.The higher the working memory span,the fewer the saccade count in probability reasoning,and the lower the degree of neglect of information,which leads to the difference of probability performance;Inhibitory function and metacognitive ability jointly regulate the impact of working memory span on the fixation and acquisition of probabilistic information. |
PDF Full Text Request