Working Memory, Cognitive Load, and Test Anxiety

A photograph of the back of a student carrying a backpack as they walk down an aisle of books in a library.

The brain never sleeps. It is always processing, and therefore can experience cognitive overload. What the body senses goes through the section of the brain called the thalamus, and the thalamus then sends what was sensed to the working memory part of the brain. Working memory can hold information for only a short time, so it is also called short term memory. Working memory can connect to the long term memory sections of the brain where information can be stored indefinitely or retrieved for immediate use. Teachers and instructional designers should be concerned with overloading student’s working memory during instruction. An overloaded working memory can lead to anxiety which can lead to test anxiety and impact students’ overall learning performance.

Working Memory

Working memory is what you are using at this very moment to process this text. Working memory is the only memory where you can monitor, update, and manipulate information; it is active and carries out on-going tasks. Working memory is for temporary storage and manipulation of information. Information from sensory memory passes into working memory, where it is either processed or discarded. What drives the working memory system is a component called the central executive. It is like the boss of working memory and directs data to two subcomponents: one deals with the manipulation of speech based information and the other deals with the manipulation of visual images (Baddeley, 1986; 2000). What is already in permanent long-term memory remains unused until it is needed for working memory.


Processed information is stored in long term memory in knowledge structures called “schemas.” Schemas are neural networks that can hold huge amounts of information into connected networks, which working memory can retrieve and process as a single unit. Working memory can process only a limited number of elements at a time, but the elements can be very large, complex, and sophisticated schemas. Thus, schema development reduces working memory load (Kirschner, 2002).

Cognitive Load

Working memory does have a limited capacity. Cognitive load theory assumes a limited working memory connected to an unlimited long-term memory (Baddeley, 1986). Good instructional design accounts for the limits of students’ working memory load. How a student performs is limited by the inherent nature of the material, referred to as intrinsic cognitive load, and by the manner in which the material is presented, referred to as germane cognitive load. Germane cognitive load is required for the construction and storage of schemata into long-term memory (Kirschner, 2002).

Test construction also must take into consideration cognitive load because the test taker is retrieving information from long term memory into working memory for making decision about how to answer. Students can experience overload when test questions do not match what was taught or have unclear or unclear instructions. Cognitive load can also be greatly increased by poorly constructed test items that violate basic item writing rules. 

Anxiety and Test Anxiety

Anxiety is considered to be composed of at least two dimensions: worry/apprehension and arousal/emotionality (Nitschke et al., 2001). Moran (2016) in a meta study states, “Overall, there appears to be a great deal of agreement that anxiety should be associated with impaired performance and working memory tasks” and that “most theories propose that the anxiety/WMC [working memory capacity] relationship can be attributed to interference or competition between anxiety-related processes and task-related processes” (p. 4).  

Moran explains further, “anxiety appears to interfere with inhibitory control over attention. This impaired inhibition, in turn, leaves anxious individuals less capable of preventing irrelevant information from gaining access to WM [working memory]and reducing the WMC available for task performance” (p. 14). Therefore, a reduction in anxiety would be expected to improve working memory.


During learning students need to have enough time to process what is in their working memory so it can be passed into long term memory or connected with what is already in their long term memory. When so much information is presented at once that students’ working memory capacity becomes overwhelmed, then a significant amount of the information will be lost.

The MindTools website (2018) makes the following recommendations for designing instruction that reduces the demands on working memory so students can learn more effectively:

  • Adapt instruction to reflect the level of expertise of the students you are teaching because the more expertise the student has, the more information the student has in their schemas. It doesn’t matter how complex a schema is – it counts as a single item in working memory.
  • Reduce the gap between the students’ current knowledge and the desired goal. If the gap is too large, the student’s working memory becomes overloaded. The best approach is to break instruction into parts to reduce the cognitive load.
  • To reduce the cognitive load when the student is presented multiple sources of information, such as diagrams, labels, and text, place the labels into the diagrams rather than placing them underneath or to one side. For example, a box that is part of a diagram with the words “part one” under the box splits the reader’s attention and increases cognitive load, whereas placing the “part one” inside the box reduces the cognitive load. 
  • To reduce the cognitive load caused by split-attention when speaking to students, try to remove any extraneous sources of noise, such as music playing in the background.

Take advantage of auditory and visual channels in working memory by replacing some of the visual information with auditory information, thus reducing the cognitive load on the students’ visual working memory by also using auditory working memory, which has its own memory space. A Mayer and Moreno study (1998) found that students learned most effectively when an animation was accompanied by narration rather than using added on-screen text.


Baddeley, A. D. (1986). Working memory. Oxford, UK: Oxford University Press.

Baddeley, A. D. (2000). The episodic buffer: A new component of working memory? Trends in Cognitive Sciences, 4(11): 417-423.

Kirschner, P.A. (2002). Cognitive load theory: Implications of cognitive load theory on the design of learning. Learning and Instruction, 12: 1-10.

Mayer, R.E. & Moreno, R. (1998). A split-attention effect in multimedia learning: Evidence for dual processing systems in working memory. Journal of Psychology, 90(2), Jun, 312-320.

Moran, T.P. (2016). Anxiety and working memory capacity: A meta-analysis and narrative review. Psychological Bulletin, 142I(8), 831-864.

MindTools. (2018). Cognitive Load Theory. Retrieved from

Nitschke, J. B., Heller, W., Imia, J. C., McDonald, R. P., & Miller, G. A. (2001). Distinguishing dimensions of anxiety and depression. Cognitive Therapy and Research, 25, 1–22.