Reduce cognitive load with proper information design
Working memory is the temporary retention of a small amount of information in a readily accessible form while we are accomplishing cognitive tasks (Baddeley, 1983; Oberauer, 2003; Barrouillet et al., 2007). It allows us to process and manipulate information in our heads. It plays a predominant role in our cognitive activities such as planning, comprehension, logic reasoning, attention control, decision-making, and problem-solving (Swanson et al., 2009; Cowan, 2014).
The limited capacity and crucial status of working memory have made it a precious resource for human cognition. Working memory overload may represent a bottleneck in information processing and human-computer interaction, creating anxiety, hampering human performance, and becoming the source of human errors (Baddeley, 1992; Eysenck & Calvo; 1992; Byrne & Bovair, 1997; Vincent, 2003). A product that imposes a particularly heavy load on a user may cause stress and demotivation, negatively impacting their working memory (Norman, 2013). Designers therefore should seek opportunities to offload the processing burden from working memory via effective design, reconsideration of the work process, and performance support.
This design review focuses on examining working memory and the influence of load on human performance and user experience via a case study of AT&T website design. This review first discusses the general characteristics and subcomponents of working memory as well as the impact of cognitive load on human emotions and behavior. It then delves deeper into how designing a product that eases cognitive load can reduce anxiety, increase pleasure and motivation, and support a user’s performance.
I. Characteristics and Subcomponents of Working Memory
1.1 Working Memory in Human Memory Stages
After the stimuli enter our sensory memory, a small amount of the sensory information that exceeds the minimal threshold moves to short-term memory (STM) and is stored for brief periods. The information is then held in working memory through rehearsal while being coded phonetically, visually, or semantically. We process and manipulate the information in working memory to perform complex cognition activities. The new information linked to our existing knowledge or semantic network is then moved from working memory to long-term memory (LTM) through accommodation, tuning, or restructuring. It is integrated and stored in LTM as part of our knowledge structure and can help us make sense of new things through activation and interchange with working memory (Atkinson & Shiffrin, 1968; Baddeley & Hitch, 1974; Winkler & Cowan, 2005).
1.2 The Limited Capacity of Working Memory
Many theories about working memory in modern cognitive psychology have been following and extending the pioneering proposal of Baddley and Hitch (1974) that working memory is a system with limited capacity; it is involved in a dynamic process that requires storing and manipulating the information simultaneously. As the limited resource is shared between storage and manipulation, a trade-off must inevitably be made in the human cognitive process. As the concurrent mental load increases and reaches capacity, one’s reasoning, comprehension, and learning performance decreases (Baddley, 1992). This limited capacity proposal is supported by various empirical studies where participants’ cognitive activities slowed down or even abandoned their task at hand (Anderson et al., 1996; Hester & Garavan, 2005; Barrouillet et al., 2007). Given the limited capacity of working memory, it is crucial that design is intuitive and helps the users focus on their goal instead of focusing on the unnecessary actions or mental efforts to take to figure out how to navigate the design.
1.3 The Time-Constrained and Highly Volatile Nature of Working Memory
Information in working memory is ephemeral; it has a short duration of around 20 seconds. Unless one makes an intentional mental effort to rehearse and repeat it for an extended time, working memory will decay or get replaced (Cowan, 1992 & 1995; Towse & Hitch, 1995). Barrouillet et al. (2004), for example, found that processing and maintenance of the main working memory tasks require attention and that memory traces will decay as soon as attention is switched away. Just as it takes effort to hold the mental load given the capacity constraints of working memory, attending to and rehearsing information also takes effort, which can cause cognitive overwhelm (Barrouillet et al., 2004). Design, therefore, should reduce the mental load for users (especially novice users) when possible and allow users to swiftly make sense of the design before their working memory decays or gets replaced.
Upon landing on the AT&T “Compare Wireless Plans” page below (Figure 1), users, especially those novice users with limited literacy or tech literacy skills, may immediately get overwhelmed by the high load of information on display and feel clueless about where to start reading or compare the information. Their working memory resources need to be allocated to store the information while processing the information temporarily. It takes time for them to absorb the information, and chances are, by the time they absorb, they already forget what they need to compare against. Technical terminology such as “UHD,” “Hotspot”, “streaming” may create a barrier for the users to comprehend and likely cause them to stop reading and leave the site.
Figure 1 “Compare Wireless Plans” page
To reduce a user’s cognitive load and accommodate their limited capacity in working memory, AT&T may take a few options: 1) think twice about who the audience (advanced or novice) is and what level of technical detail they may need in this section when one first lands on this page; 2) distill the essential factors people often care about when choosing a plan and only include these; 3) ensure symmetry visually in the comparison chart for the same feature (e.g., display “streaming” in a symmetrical manner across all 3 plans); and 4) consider building a more straightforward choice architecture by restructuring the grid by extracting the essential features into a separate column and replace text with symbols. This will reduce the load of reading (considering the unique challenge of information online in a non-linear format), facilitate one’s decision-making, and give users a sense of ownership and empowerment.
The limited capacity, time-constrained nature, coupled with its interaction with LTM mentioned earlier, has made working memory highly volatile and subject to interference. Working memory can be easily disrupted by interruptions (i.e., stimuli that required attention) or distractions (i.e., irrelevant stimulus) (Clapp et al., 2010). Working memory is also easily influenced or overwritten by proactive intervention, e.g., categories and meaning in prior learning or LTM (Baddley et al., 2009). From this perspective, design should effectively leverage meaningful schema that one can retrieve from their prior knowledge or LTM, so that it can help reduce mental efforts, facilitate working memory operations, and allow one to conduct cognitive activities more effectively and effortlessly.
The schemata (e.g., information architecture) and categorization of the AT&T site are less intuitive than they can be. It is very difficult for one to make sense of how the site is structured by applying their existing semantic framework stored in LTM. For example, it is counterintuitive that items such as “Wireless” in Figure 2a and “Prepaid,” and “Bundles” in Figure 2b, which belong to different categories, are grouped with the rest of the items on the menu (e.g., “Internet,” “TV,” “Home phone,” and “Smart technology”). Similarly, in submenus, “Phones & devices” and “accessories” belong to different categories. Counterintuitive architecture, naming, and categorization conflict with how people usually organize information in prior learning. As a result, it is next to impossible for one to apply their semantic framework to the navigation menu and find what they are looking for. Figuring out where to find in submenu may unnecessarily drain a user’s mental effort in this case.
AT&T may consider optimizing the user experience by organizing the items into meaningful categories and patterns closer to how users would typically understand and categorize information (see, e.g., Verizon categorization in Figure 3). Exercises such as card sorting may help reveal the schemata users have and help build a clear, straightforward, and easy-to-understand information architecture and visual hierarchy, which allow one to leverage their pre-attentive processing and tap into their LTM.
Figure 2a Menu and submenu of “Wireless”
Figure 2b Menu and submenu of “Prepaid”
Figure 3 Categorization and grouping of Verizon site
1.4 The Subcomponents of Working Memory
In line with the efforts to understand the characteristics of working memory, a series of conceptual models were proposed to explore the structure of working memory. One of the most influential and often discussed models is the three-component model by Baddley (1992), i.e., the phonological loop, the visuospatial sketchpad, and central executive, which is highly correlated to the 3 characteristics discussed earlier.
The phonological loop includes two components: (1) the phonological store, which has limited capacity and can only hold information for a few seconds; and (2) the articulatory rehearsal process, which is charge of rehearsal so that items can be kept in the phonological store from decaying (Baddley, 1992; Goldstein, 2011). The visuospatial sketchpad helps the brain to hold visual and spatial information through visual imagery, i.e., creating in our mind the visual images in the absence of a physical visual stimulus (Baddley, 1992; Goldestein, 2011). It’s worth mentioning that intervention, e.g., irrelevant information such as irrelevant sound and visual stimulus may disrupt the operation or rehearsal of the phonological loop and visuospatial sketchpad; working memory, therefore, can be vulnerable to “noises” and get easily overwritten (Baddley et al., 1984; Brooks, 1968). From this perspective, it is essential to avoid designs cluttered with unnecessary or irrelevant information, icons, images, and the like for the user. This can protect users from getting distracted from what they are trying to achieve and prevent draining their precious mental resources.
Finally, the central executive serves as the traffic control of the working memory system. It pulls information from long-term memory (LTM) through activation and coordinates with the visuospatial sketchpad and phonological loop. Serving as an attentional-controlling system, the central executive determines how attention is divided or switched between tasks and blocks interference by inhibiting distractors. Therefore, the central executive is the most crucial component and is where the major work of working memory occurs (Baddeley, 1996; Engle et al., 1999). What is worth mentioning is the central executive is also where anxiety resides. People with math anxiety waste their central executive resources by attending to their anxiety (Ashcraft & Krause, 2007). Anxiety also impairs one’s central executive functioning in completing non-verbal tasks (Eysenck et al., 2005). Therefore, it is imperative that a design is easy to understand and allows the users to dedicate their central executive resources fully to accomplishing the user goals and achieving their goals in a relatively effortless manner. In doing so, we can reduce users’ anxiety, fear, or worries while conducting mental activities, avoiding negative emotions constricting their working memory.
The episodic buffer in Baddeley’s (2000) revised working memory model creates a multimodal code, which can integrate information from the phonological loop and the visuospatial sketchpad, and from LTM, into a unitary episodic representation.
The header or menu bar of AT&T site is cluttered with icons and items, which may create mental chaos and overstimulation for the user. A few items are redundant (e.g., “Deals” and “Wireless deals”) and may be irrelevant for the users to achieve their goal. The high density of icons and menu items, coupled with unclear categorization, increases the user’s mental load. The many visual stimuli create unnecessary burden for the central executive to store, process, or inhibit distractors temporarily all at once. They may also disrupt the operation or rehearsal of the phonological loop and visuospatial sketchpad. As a result, users are distracted from what they are about to do, and their mental resources may get drained quickly.
AT&T may consider reducing the number of icons and items by distilling the absolute necessities for the user when they first land the site (e.g., keep the top menu bar only) and slowly add load by showcasing the information and icons based on the user’s goal. For example, icons and items related to “Prepaid” show up if one clicks on “Prepaid,” indicating their interest in learning more information.
Figure 4 Menu of the AT&T main page
II. Cognitive Load and its Impact
Cognitive load has important implications on optimizing working memory, so one can dedicate limited capacity of their working memory to the highest-value path to achieve their goal in a short period of time. Cognitive load refers to the amount of mental resources needed to conduct cognitive activities in working memory (Baddley, 1992). As mentioned earlier, the cognitive capacity in working memory is limited. When the amount of effort required exceeds the capacity, it will exert a heavy cognitive load on the users and may hamper learning and cause errors. Based on the tradeoff principle mentioned earlier, if our brain is dedicated to absorbing more information than it can handle, and it’s preventing you from achieving your goal. Many research studies indicated the negative linear effect of cognitive load on the length of a list that one can recall (e.g., Barrouillet et al., 2011). As the memory load increases, reasoning takes longer (Kyllonen & Christal, 1990; Süß et al., 2002). High-load situations may cause significant user errors and create unnecessary anxiety and frustration (Byrne & Bovair, 1997; Vincent, 2003).
It is therefore paramount to manage the user’s cognitive load through effective design, reconsideration of the work process, and performance support. Chunking, for example, is an effective way to reduce the load on one’s working memory (Thalmann et al., 2019). Chunking describes the fact that a collection of elements that are strongly associated are combined into larger, meaningful units. By taking advantage of the existing schema, semantic network, mental model, and other knowledge framework stored in one’s LTM, chunking “expands” the capacity of one’s working memory by freeing up mental resources used to interpret new information. For example, by arranging a sequence of unrelated words into a meaningful sentence or phrases, one can expand their memory span of recalling these words (Butterworth et al., 1990; Erricsson et al., 1980; Cowan, 2001). Designers therefore should apply “chunking” technique when possible, organize information by well-known principles, and create a perceptual experience in a natural and intuitive way.
Users may experience a high cognitive load when landing on the account overview page below (Figure 5). It is clustered with CTA (call-to-action) with inconsistent design elements, for example, different colors, sizes, see (1), (2), and (3). The information in (4) is chunked into smaller units. However, the units are not meaningful and intuitive enough to detect the patterns quickly.
AT&T may consider reorganizing the chunks by placing strongly associated items in proximity with each other (e.g., information related to add a line and add a new line can be placed right next to each other) and further optimize the chunks by 1) creating grouping and patterns based on critical user goals, such as “Make a Payment,” “Contact Support,” “Upgrade Devices,” “Add a Line,” “Benefits”, etc.; and 2) placing pictures in close proximity to the text. This will allow users to quickly identify the patterns, associate the meaning to them, and split their cognitive load between the phonological loop and visuospatial sketchpad.
Figure 5 Account Overview page
III. Anxiety, Motivation, and their Impact on Human Performance
Cognition and affect together influence how human understand and interact with the world. While cognition interprets and makes sense of what is happening, affect evaluates and judges through reflecting upon our experience, which informs us of what action to take moving forward (Norman et al., 2003). Notably, the affective system not only controls the muscles of our body but changes how our brain functions through chemical neurotransmitters (Norman, 2004). Negative affect may lead one to more focused, depth-first processing and can sometimes cause tunnel vision. Positive affect, on the contrary, tend to lead one to a widely spread processing and allow one to enhance their creativity when in a pleasurable state (Norman et al., 2003).
Anxiety, for example, impairs a user’s efficiency and adversely impacts their performance. With anxiety, one may narrow their thought processes and restrict their creative and imaginative problem solving. Eysenck and Calvo (1992) found that performance deficits caused by generalized anxiety may be prominent in exactly those tasks that tap the limited capacity of working memory. Anxiety also causes worry, which further leads to a reduction in one’s storage and processing capacity of the working memory available for a concurrent task (Humphreys & Revelle, 1984; Sarason, 1988). Motivation, on the other hand, allows a user to extend their biological and cognitive capabilities and overcome limitations that would otherwise adversely affect performance (see e.g., Humphreys & Revelle, 1984).
In view of the significant impact of affect in human behavior and performance, AT&T designers should seek opportunities to offload processing burden from working memory and deliver a positive, motivating product experience that will stimulate the operation of a user’s cognitive system.
IV. Conclusion
Working memory plays a significant and indispensable role in our reasoning, decision making, action planning, and creative problem-solving. With its limited capacity, time constraints, and volatile nature, working memory is a precious resource that needs to be utilized in an optimal way to dedicate their cognitive capabilities fully to activities crucial to their goal. In doing so, design is more likely to positively affect and drive up a user’s performance.
Designers should thoughtfully design to manage the user’s cognitive load in working memory, be mindful of the user’s emotional state, and accommodate the limitation of the user’s working memory through practical offloading, reconsidering the work process, and providing performance support. In the case of the AT&T site, we see how designers may adopt design principles to reduce one’s cognitive load, facilitate one’s complex mental activities via intuitive design, decluttered interface, and effectively chunking information in well-known organizing principles. This can increase the product’s usability, improve a users’ performance, increase adoption, and shift to an “experience economy” by crafting a product that people will love.
References
Anderson, J. R., Reder, L. M., & Lebiere, C. (1996). Working memory: Activation limitations on retrieval. Cognitive Psychology, 30(3), 221–256. https://doi.org/10.1006/cogp.1996.0007
Ashcraft, M. H., & Krause, J. A. (2007). Working memory, math performance, and math anxiety. Psychonomic Bulletin & Review, 14(2), 243–248. https://doi.org/10.3758/BF03194059
Atkinson, R. C., & Shiffrin, R. M. (1968). Chapter: Human memory: A proposed system and its control processes. In Spence, K. W., & Spence, J. T. The psychology of learning and motivation (Volume 2). New York: Academic Press. pp. 89–195
Baddeley, A. D., & Hitch, G. (1974). Working memory. In G.H. Bower (Ed.), The psychology of learning and motivation: Advances in research and theory (Vol. 8, pp. 47–89). New York: Academic Press.
Baddeley, A. (1992). Working Memory. Science, 255(5044), 556–559. http://www.jstor.org/stable/2876819
Baddeley, A. (2007). Working memory, thought, and action. Oxford University Press. https://doi.org/10.1093/acprof:oso/9780198528012.001.0001
Baddeley, A. (2012) Working Memory: Theories, Models, and Controversies. Annual Review of Psychology, 63, 1–29. http://dx.doi.org/10.1146/annurev-psych-120710-100422
Baddeley, A. D., Lewis, V., & Vallar, G. (1984). Exploring the articulatory loop. The Quarterly Journal of Experimental Psychology A: Human Experimental Psychology, 36A(2), 233–252. https://doi.org/10.1080/14640748408402157
Baddeley, A., Papagno, C., & Vallar, G. (1988). When long-term learning depends on short-term storage. Journal of memory and language, 27(5), 586–595.
Baddeley, A. D., Eysenck, M., & Anderson, M. C. (2009). Memory. New York: Psychology Press.
Barrouillet, P., Bernardin, S., & Camos, V. (2004). Time Constraints and Resource Sharing in Adults’ Working Memory Spans. Journal of Experimental Psychology: General, 133(1), 83–100. https://doi.org/10.1037/0096-3445.133.1.83
Barrouillet, P., Bernardin, S., Portrat, S., Vergauwe, E., & Camos, V. (2007). Time and cognitive load in working memory. Journal of experimental psychology. Learning, memory, and cognition, 33(3), 570–585. https://doi.org/10.1037/0278-7393.33.3.570
Barrouillet, P., Portrat, S., & Camos, V. (2011). On the law relating processing to storage in working memory. Psychological review, 118(2), 175–192. https://doi.org/10.1037/a0022324
Brooks, L. R. (1968). Spatial and verbal components of the act of recall. Canadian Journal of Psychology/Revue canadienne de psychologie, 22(5), 349–368. https://doi.org/10.1037/h0082775
Butterworth, B., Shallice, T., & Watson, F. L. (1990). Short-term retention without short-term memory. In G. Vallar & T. Shallice (Eds.), Neuropsychological impairments of short-term memory (pp. 187–213). Cambridge University Press. https://doi.org/10.1017/CBO9780511665547.011
Byrne, M.D., & Bovair, S. (1997). A Working Memory Model of a Common Procedural Error. Cogn. Sci., 21, 31–61.
Cowan, N. (1992). Verbal memory span and the timing of spoken recall.
Journal of Memory and Language, 31, 668 –684.
Cowan, N. (1995). Attention and memory: An integrated framework.
Oxford, England: Oxford University Press
Clapp, W. C., Rubens, M. T., & Gazzaley, A. (2010). Mechanisms of working memory disruption by external interference. Cerebral cortex (New York, N.Y.: 1991), 20(4), 859–872. https://doi.org/10.1093/cercor/bhp150
Cowan, N. (1992). Verbal memory span and the timing of spoken recall. Journal of Memory and Language, 31, 668–684.
Cowan N. (2001). The magical number 4 in short-term memory: a reconsideration of mental storage capacity. The Behavioral and brain sciences, 24(1), 87–185. https://doi.org/10.1017/s0140525x01003922
Cowan N. (2014). Working Memory Underpins Cognitive Development, Learning, and Education. Educational psychology review, 26(2), 197–223. https://doi.org/10.1007/s10648-013-9246-y
Daneman, M., & Carpenter, P. A. (1980). Individual differences in working memory and reading. Journal of Verbal Learning & Verbal Behavior, 19(4), 450–466. https://doi.org/10.1016/S0022-5371(80)90312-6
Engle, R. W., Kane, M. J., & Tuholski, S. W. (1999). Individual differences in working memory capacity and what they tell us about controlled attention, general fluid intelligence, and functions of the prefrontal cortex. In A. Miyake & P. Shah (Eds.), Models of working memory: Mechanisms of active maintenance and executive control (pp. 102–134). Cambridge University Press. https://doi.org/10.1017/CBO9781139174909.007
Ericsson, K. A., Chase, W. G., & Falloon, F. (1980). Acquisition of a memory skill. Science, 208, 1181–1182.
Goldstein, E. Bruce. Cognitive Psychology: Connecting Mind, Research, and Everyday Experience (p. 451). Cengage Learning. Kindle Edition.
Eysenck, M. W., & Calvo, M. G. (1992). Anxiety and performance: The processing efficiency theory. Cognition and Emotion, 6(6), 409–434. https://doi.org/10.1080/02699939208409696
Eysenck, M. W., Payne, S., & Derakshan, N. (2005). Trait anxiety, visuospatial processing, and working memory. Cognition and Emotion, 19(8), 1214–1228. https://doi.org/10.1080/02699930500260245
Goldstein, E. B. (2011). Cognitive psychology: Connecting mind, research, and everyday experience. Australia: Wadsworth Cengage Learning.
Just, M. A., & Carpenter, P. A. (1992). A capacity theory of comprehension: Individual differences in working memory. Psychological Review, 99(1), 122–149. https://doi.org/10.1037/0033-295X.99.1.122
Halford, G. S., Wilson, W. H., & Phillips, S. (1998). Processing capacity defined by relational complexity: implications for comparative, developmental, and cognitive psychology. The Behavioral and brain sciences, 21(6), 803–864. https://doi.org/10.1017/s0140525x98001769
Hester, R., & Garavan, H. (2005). Working memory and executive function: The influence of content and load on the control of attention. Memory & Cognition, 33(2), 221–233. https://doi.org/10.3758/BF03195311
Humphreys, M. S., & Revelle, W. (1984). Personality, motivation, and performance: A theory of the relationship between individual differences and information processing. Psychological Review, 91(2), 153–184. https://doi.org/10.1037/0033-295X.91.2.153
Kyllonen, P. C., & Christal, R. E. (1990). Reasoning ability is (little more than) working-memory capacity?! Intelligence, 14(4), 389–433. https://doi.org/10.1016/S0160-2896(05)80012-1
Norman, D.A., Ortony, A., & Russell, D.M. (2003). Affect and machine design: Lessons for the development of autonomous machines. IBM Syst. J., 42, 38–44.
Norman, D. A. (2004). Emotional design: Why we love (or hate) everyday things. New York: Basic Books.
Norman, D. A. (2013). The design of everyday things. MIT Press.
Oberauer, K., Süß, H.-M., Wilhelm, O., & Wittman, W. W. (2003). The multiple faces of working memory: Storage, processing, supervision, and coordination. Intelligence, 31(2), 167–193. https://doi.org/10.1016/S0160-2896(02)00115-0
Sarason, I. G. (1988). Anxiety, self-preoccupation and attention. Anxiety Research, 1(1), 3–7. https://doi.org/10.1080/10615808808248215
Süß, H.-M., Oberauer, K., Wittmann, W. W., Wilhelm, O., & Schulze, R. (2002). Working-memory capacity explains reasoning ability — and a little bit more. Intelligence, 30(3), 261–288. https://doi.org/10.1016/S0160-2896(01)00100-3
Swanson, H. L., Xinhua Zheng, & Jerman, O. (2009). Working memory, short-term memory, and reading disabilities: a selective meta-analysis of the literature. Journal of learning disabilities, 42(3), 260–287. https://doi.org/10.1177/0022219409331958
Thalmann, M., Souza, A. S., & Oberauer, K. (2019). How does chunking help working memory?. Journal of experimental psychology. Learning, memory, and cognition, 45(1), 37–55. https://doi.org/10.1037/xlm0000578
van den Broek, P. & Helder, A. (2017). Cognitive processes in discourse comprehension: Passive processes, reader-initiated processes, and evolving mental representations. Discourse Processes, 54, 360–372.
Vicente, K. J. (2003). The human factor: Revolutionizing the way people live with technology. Toronto: A.A. Knopf Canada.
Winkler, I., & Cowan, N. (2005). From sensory to long-term memory: evidence from auditory memory reactivation studies. Experimental psychology, 52(1), 3–20. https://doi.org/10.1027/1618-3169.52.1.3
