TASK-SWITCHING EXPERIMENT

Task-Switching

Israel Edery

Brooklyn College

Abstract

In an age when multi-tasking is considered the norm and sometimes even glorified, some wonder whether it is actually possible to transition smoothly and effectively from one thing to another. Studies have indicated that there is a switch-cost that transpires when one changes from one task to another. This means that transitioning frequently between multiple tasks takes up much more energy and time than when simply focusing on one thing. As a continuation to the current body of evidence for the existence of a switch-cost when switching between tasks, we have conducted an experiment that compared the reaction time (RT) of individuals to either repeated/similar tasks, or switched/dissimilar tasks. The data for this experiment was collected subsequent to a group of students participating in an online activity that had them reacting to different types of stimuli presented on the computer screen. The reaction times between the two conditions (switched vs. repeated tasks) were gathered and compared using a paired samples t-test. The results showed a significant decrease in speed and efficacy in the task-switching condition. Based on our results we have concluded that indeed changing frequently between tasks is not as efficient as when one is focused on a singular task.

Introduction

There has been a significant amount of research dedicated to understanding various cognitive functions and how they operate. Findings in this domain are important and consequential because they help us understand, detect, and treat various disorders related to cognitive and executive functions. Likewise, exploring the underlying mechanisms of how the brain operates can help find new ways of maximizing and enhancing our day-to-day lives. One example of why cognition research can be so useful would be with regards to the question of when it is best to make an important decision regarding any given dilemma. Numerous human task-switching studies have shown that decision-making that follows a task switch is slower and less accurate than that which follows a task repetition (Stoet & Snyder, 2007). This idea can be generalized not only to small tasks and decisions but also to larger-scale decision-making. Knowing that decisions are made better while focusing on the subject matter and not while focusing on something completely different is a practical tip that can help guide all people with their decision-making practices on a regular basis.

In the experiment we conducted, our intention was to isolate two distinct but related kinds of cognitive tasks. We wanted to show how in one scenario, the brain manages and handles itself rather smoothly and efficiently, while in a different situation the brain function and performance are delayed and less efficient. More specifically, we wanted to provide causal evidence that doing the same task repeatedly is significantly different than switching between two different or dissimilar tasks. Past studies have shown that subjects’ responses are substantially slower and usually more error-prone immediately after a task switch (Monsell, 2003). The reason for this is due to what is known as a “switch cost” which takes place when the brain detects that it needs to reset in preparation for the new task. This follows the idea that prior to any task switch there is a “task-set” that takes place through the natural reconfiguration process that occur in the brain in anticipation of its upcoming objective (Monsell, 2003). 

To measure the “switch cost” phenomenon, a group of college students performed a digital activity where they had to react to certain stimuli that were similar, and subsequently, other stimuli that were dissimilar. We measured the reaction times (RT) to the different kinds of stimuli and compared the RT of the repeated tasks to the RT of the task-switch stimuli. We hypothesized that repeating two tasks that are similar would take significantly less time than doing two tasks that are not the same. In doing so, we would show that there is indeed a cost that is incurred when the mind switches over from one task to another. If, however, our analysis would indicate no real difference between RT’s in the two conditions then our hypothesis regarding the significance of “switch cost” would be conclusively incorrect. The reasoning for our hypothesis was twofold. Firstly, we have an abundance of research that has already found evidence for a switch cost. Secondly, it seems fairly intuitive that one would pass through any given task with ease - so long as the task is focused on one specific topic. Conversely, if one is switching between tasks it seems quite sensible to assert that he/she would require more energy and his/her performance would be more prone to error and reduction in speed. 

Methods

Participants

For this experiment, data were collected from a diverse group of students attending CUNY Brooklyn College (N = 48) who were taking a course in research methods (PSYC 3540) in the Spring semester of 2022.

Assessments and Measures

Students were asked to participate in a computer activity on the website PsyToolkit.org. The activity involved the student having to press either the letter “b” or “n” in response to the prompts on the computer screen throughout the activity. The decision to press either one of the keys (“b” or “n”) depended on if the prompt was an odd or even number, or if the letter presented on the screen was a consonant or a vowel - this was the task-repeat section. There was also a section when both of these types of prompts (odd/even numbers and consonant/vowel) were presented together - this was the task-switch section. The manner in which the stimuli were presented reflected repeated tasks vs. switched tasks, with switched tasks serving as the independent variable. The reaction time (RT) was encoded and gathered, serving as the dependent variable. The comparisons made were within-subjects, with the duration of the entire activity being approximately 5 minutes.  

Results

The data were gathered and analyzed in Jamovi using a standard paired-samples t-test. A noticeable increase in RT for the task-switch condition is apparent when comparing the means and standard deviations of the two conditions - with the mean RT of the task-repeat condition at 1222 and a SD of 386 compared to the task-switch condition with a much higher mean RT of 1583 and a SD of 463 (Table 1). The paired-samples t-test comparison between the two conditions yielded a statistically significant output (t (47.0) = -7.12, p <.001) (Table 2). Since our alpha for this statistic was placed at .05, the p-value of <.001 allows us to reject the null hypothesis and retain our alternative hypothesis which stated that the task-switch condition would take more time than the repeated measures condition. This also supports the notion of the existence of task-set reconfiguration processes prior to a task-switch.

Discussion

We set out to confirm the switch-cost concept in the task-switching paradigm by having a group of students perform an experiment in which they would have to navigate through repeated and switched tasks. We found that there was a significant cost in the level of performance when the students were required to switch between tasks. Our findings have indeed supported the general consensus which posits that there are higher processing demands associated with task-set reconfiguring than when continuing to perform a repeated task (Monsell, 2003). 

Our current culture often encourages doing many things at once and glorifies the ability to multi-task. Research exploring executive function processes, distribution of attention, and task-switching costs are increasingly salient in a culture of multi-tasking. The need to explore the pros and cons and evaluate the cost-benefit relationships between different modes of work seems to be growing and there is no doubt that psychological research has and should continue to contribute to such discussions. In this day and age, it seems rather intuitive that investing in further research in this domain will be truly beneficial for the public for generations to come.

An aspect of the discussion around task-switching that we have not included in our experiment is the neurological components and processes that give rise to concepts like switch-cost and others. Based on what has been found thus far, it is clear that the prefrontal cortex is involved in many aspects of task-set reconfiguration. One study stated that; we are still waiting for more precise knowledge of which pathways and control loops in the brain are crucial and how they relate to other task settings that are studied in cognitive control (Vandierendonck, Liefooghe, & Verbruggen, 2010). While the details of the current neurological findings with regard to the underlying processes of task-switching are beyond the scope of this paper, I believe that this is certainly a component of the research on this topic that is noteworthy and important. 

Some of the limitations in this experiment might include the fact that it was difficult to monitor each individual while they were performing the task-switching activity. This may have resulted in some of the data being less precise than it could have been. Students were not compensated or given any incentive to do their best on this assessment which, again, may have affected some of the data as well. 

References

  Stoet, & Snyder, L. H. (2007). Extensive practice does not eliminate human switch costs. Cognitive, Affective, & Behavioral Neuroscience, 7(3), 192–197. https://doi.org/10.3758/CABN.7.3.192

  Monsell. (2003). Task switching. Trends in Cognitive Sciences, 7(3), 134–140. https://doi.org/10.1016/S1364-6613(03)00028-7

  Vandierendonck, Liefooghe, B., & Verbruggen, F. (2010). Task Switching: Interplay of Reconfiguration and Interference Control. Psychological Bulletin, 136(4), 601–626. https://doi.org/10.1037/a0019791

Tables and Figures

Table 1

Table 2