The Influence of Worker’s Motivation on Intellectual Concentration by ACT-R Cognitive Models

Abstract

IntroductionThe purpose of this study is to examine the worker’s motivation effect on intellectual concentration by an experiment and a cognitive process simulation. By utilizing the Adaptive Control of Thought-Rational (ACT-R) concept, it helps to understand the cognitive models of the different motivational condition. ACT-R is one of the prevalent cognitive architectures that mainly differ the cognitive process into production modules and declarative modules.Several studies have been investigating the human cognitive performance in respect of others factors such as working environments. Inspirited by previous studies, this study has explored the human motivational factors that might influence intellectual concentration. Additionally, exploring the differences in workers’ individual problem-solving strategy is one of the alluring factors in this study given the differences in motivational conditions.MethodsFive participants were recruited for the experiment the range of their ages between 18-22 years old as a pilot study. The simple summation mathematical task was given during the experiments and shows two numbers on the left side and right side of the monitor screen. They were asked to answer the problem by pressing the numerical keyboard and the answering time was automatically recorded on the experiment program. The eye gaze movements were recorded and analyzed during the experiments. In this study, the participants performed the task under two different motivational conditions. The first condition was set to create the urge to finish the task immediately which generates high motivation to finish the task as fast as they can. The other condition was set in a neutral condition which the participants could perform at their own pace without a rush. From this experiment, approximately 3000 answering time data and 900.000 frames of the eye gaze movement were gathered and analyzed.ResultsThe different motivational conditions influenced the user’s task performance, showing a statistical difference in the user’s answering time in a high motivational and neutral conditions (p < 0.01). The distribution of the answering time follows the log-normal distribution shapes shown by both conditions.The participants performed different strategies in the cognitive task. Following the ACT-R modules, in general, the identified eye gaze patterns are divided into 2 patterns. The first pattern follows the production rules: (1)find-left-number, (2)attend-left-number, (3)encode-left-number, (4)retrieve-left-number, (5)find-right-number, (6)attend-right-number, (7)encode-right-number, (8)waiting-for-start-back, (9)attend-start-back, summation, and (10)keyboard-click. On the other hand, the second pattern follows the production rules: (1)find-left-number, (2)attend-left-number, (3)encode-left-number, (4)retrieve-left-number, (5)find-right-number, (6)encode-right-number, (7)summation, keyboard-click, (8)waiting-for-start-back, and (9)attend-start-back.In the different motivational conditions, the pattern’s appearance is altered. For participants 1, 3, and 5 the second patterns were dominant during the high motivational condition. Some irregular patterns arouse in the last 5 minutes of the task. This condition might due to the degradation of the user’s cognitive performance when the user feels tired and it might be the time duration that affected their cognitive performance. The results of this study might enlighten future research on cognitive performance by proposing a new method of eye gaze pattern identification on ACT-R cognitive models.