Cognitive Skills & Otter.Ai
Human Factors in Information Design, Bentley University
HF700: Human Factors Foundations
Dr. WM Gribbons
November 23, 2020
Metacognition is a notion of self-awareness, which early philosophers investigated and attributed with the maxim “know thyself” (Metcalfe, 2008). This phenomenon refers to how we think about thinking. Humans are uniquely equipped with their metacognitive ability to conduct “higher order thinking that involves active control over the cognitive processes engaged in learning (Livingston, 2003).” From an evolutionary perspective, this “self-reflective capability is nevertheless a special mental capability and a phenomenological experience that is special to humans” and “confer[s] humans an ability to escape from being stimulus bound and allow self-control of their learning and actions (Metcalfe, 2008).” In turn, designers examine these metacognitive processes to extrapolate actionable designs that aid and guide human comprehension and progress trajectory.
This paper will review the primary components and effects of metacognitive processes. Human performance and metacognitive skill areas intersect at metacognitive components and skills, literacy and reading psychology, and reading styles and strategies. A design case of an online transcription service examines metacognitive properties of reading psychology and strategies.
Metacognitive Components & Skills
Metacognition, operates in parallel to primary cognitive activity and translates into three main processes: metacognitive knowledge, metacognitive monitoring, and metacognitive control (Dunlosky & Metcalfe, 2008). These processes consist of goal setting, self assessment, monitoring progress, and validating feedback. And it indicates “the ease or difficulty of recall and thought generation and the fluency with which new information can be processed,” allowing us to examine the minutiae of processes in thinking about thinking (Schwarz, 2004).
The first component of metacognition is metacognitive knowledge, “one's stored knowledge or beliefs about oneself and others as cognitive agents, about tasks, about actions or strategies, and about how all these interact to affect the outcomes of any sort of intellectual enterprise” (Flavell, 1979). To initiate the processes, an individual formulates a goal based upon this metacognitive knowledge, declarative knowledge pertaining to “facts, beliefs, and episodes that you can state verbally (recall from long-term memory),” and simultaneously constructs a query to achieve the intended outcome. This benchmark enables appraisal of ability, influenced by individual cultural perception (Dunlosky & Metcalfe, 2008). With a goal now in mind, an individual must set a strategy to monitor their thinking via evaluating feedback cues of progress, strategy adjustment, and performance. Metacognitive monitoring, referring to “assessing or evaluating the ongoing progress or current state of a particular cognitive activity,” is the highest level of cognitive monitoring, which acts to access and evaluate the evolution around our thinking processes (Dunlosky & Metcalfe, 2008). The command center of the mind directs attention to given strategies to approach the goal. This iterative process allows an individual to evaluate investment costs for appropriate reallocation of attention. When a task is deemed no longer supportive of the ultimate goal, an individual will exercise metacognitive control to evaluate and reconsider their strategy. Metacognitive control allows an individual to alter any existing queries by “regulating an ongoing cognitive activity, such as stopping the activity, deciding to continue it, or changing it in midstream (Dunlosky & Metcalfe, 2008).” Metacognition is connected to working memory and must be considered within the context of load as these actions run parallel to primary cognitive activity. However, humans are limited in capability to identify and manipulate large quantities of information to determine the optimal procedural alternative, which yields potential unavoidable error despite best intentions.
Metacognition and performance are correlated and mediated by self-efficacy; an individual who is highly aware and motivated, equipped with a firm belief in their capabilities and strong metacognitive strategies and skills, often performs better (Coutinho, 2008). Expert performers are highly engaged monitoring strategy and decision making. Experts, nimble and adept, are highly aware of their performance level and execute significant metacognitive ability to validitate ongoing queries of rich strategic arrays (Schraw, 1994). These high performers operate with a pre-existing set of heuristics in their minds while novices must execute a task and form these analytical structures, which requires significant effort and may compound the complexity of the task (Geurten, 2018). Novices with limited cognitive abilities may require additional application support to perform a task. It is likely that any given audience exhibits a wide variation of cognitive ability and some individuals may even experience cognitive disabilities. As a result, we can expect significant variability in human performance and proficiency. However, designers foster productive interaction by supporting metacognitive processes.
Humans employ metacognitive skills in several primary areas: processing priorities, reading psychology, psychology of search, reading styles, literacy, cognitive learning styles, and decision-making. Metacognitive skills allow the mind to monitor, direct, and guide attention to words on the screen based on visual or textual hierarchy and contrast. Designers reallocate attention in metacognitive processes to accomplish tasks with greater efficiency.
Literacy & Reading Psychology
Connections between factors of metacognitive skill areas enable processing enhanced performance across reading styles. Efficiency in reading style is critical as an individual cannot devote the time or mental effort to read everything thoroughly beginning to end. Reading psychology and reading styles align. A significant portion of the United States population is functionally illiterate. This inequity points to accidental and systematic economic and opportunistic disparities that compound the need for universal design and accessibility (Redish, 2000). Further, as we age, we tend to lose the ability of cognitive resources yet are simultaneously met with an increasing number of complex textual documents regarding financial health decisions. To address the discrepancy across the American literacy landscape, the plain language movement set standards of readability and structure for accessible writing (Mazur, 2000; Jarrett, Redish, Summers, & Meloncon, 2013).
Inexperienced readers, oftentimes children or poor readers, must focus significant attention on basic components such as individual characters (Gibson, 1969). Thus, reading psychology is directly impacted by font features such as ascender recognition, line length, type size, line spacing, and justification. As a baseline, to ensure optimal reading passage legibility text should be displayed in common serif typefaces to best differentiate characters (Sanocki & Dyson, 2012). Sufficient letter spacing ensures character legibility, which these readers must identify and then coalesce to form a word. For a new reader, much cognitive resource is allocated to recognizing each individual letter, while an expert quickly identifies and combines familiar letterforms into a meaningful word (Sanocki & Dyson, 2012). While these text factors are critical to comprehension, developmental psychologists and reading researchers have since shifted their focus to reading factors of, “awareness, monitoring, and strategy use for text-processing (Garner, 1987).” A mature reader overlooks their automatic identification of characters; they deemphasize individual characters and prioritize recognizing word shape, perimeter and outline (Fiset, Blais, Éthier-Majcher, Arguin, Bub, Gosselin, 2008). An expert then advances to assign meaning to words in a sentence and ultimately surmise meaning from a passage. When a resource is well written, the resultant is more conducive to detection and evaluation of content as opposed, as the reader expends less load on clarification and allocates more time to comprehension (Gibson, 1969).
In regard to characteristics of reading competency, a poor reader that displays lower literacy may be a new or young reader or an individual with a reading disability or learning incompetence (Jarrett, Redish, Summers, & Meloncon, 2013). Oftentimes vocabulary can be a good benchmark for reader capability. A poor reader/learner tends to be impulsive, is limited by their working memory, and exhibits lack of self-monitoring as they struggle to adjust reading strategies. They are less tolerant to variation, novel structural formats, and irregular justification (Redish, 2000). A poor reader is unable to detect structure and constantly operates at a high state of anxiety, compounded by the knowledge that they are likely to struggle. Here, a style guide may combat some anxiety by providing a known framework for readers to monitor and recognize, apply and interpret a resource (Mazur, 2000). Content designers often overlook the poor readers, and by failing to support deficiencies, we exaggerate existing difficulties. For the poor readers, complex text may be an insurmountable barrier, while it is simply a nuisance for expert counterparts. In contrast, a good reader/learner imposes their own sense of logic and employs strong working memory and self-monitoring. These readers benefit from design variation and structural formats and are tolerant of lack of structure. This is representative of reflective reading strategies that employ adaptive processing.
Reading Styles & Strategies
Reading strategies require connections across fields of text. These strategies are sampling techniques that are driven by goals and efficiency which are especially critical given the text context. Online reading and its nonlinear nature requires that a reader allocate auxiliary monitoring load to glean meaning from a resource (Duggan, Payne, 2011). One strategy is to skim, in which a reader intends to find the general gist of a passage. Scanning is when a reader is quickly seeking for specific information. When a reader wants to scan with the goal of assigning meaning to specific items (Day, 1993). The ability of a reader to construct meaning, form connections, and detect structure diminishes as working memory is devoted to understanding the resource within this environment format (Ali, Wahid, Samsudin, Idris, 2013). As such, there is a higher probability of error in online formats and to circumvent some of this error content should be written as small modules of information assumed to be randomly accessed.
To expand upon the concept of information search, Information Foraging Theory seeks to model how strategies of information collection and consumption are adapted within the evolving circumstance of the environment (Pirolli & Card, 1999). It follows the notion that a hunter follows a scent to obtain its prey. Similarly, a reader “forages” for key words and phrases—titles, abstracts, names—that are related to an activated network with the assumption that “time allocation and information filtering and enrichment activities in environments in which information is encountered in clusters (Pirolli & Card, 1999).” A heightened state of attention drives an intense and intentional “identification of information value from proximal cues (Pirolli & Card, 1999).” A reader then evaluates the material based on “decisions about the selection and pursuit of information items” to determine appropriate attention allocation given potential investment (Pirolli & Card, 1999). For compatibility with Information Foraging Theory, design for search uses hierarchy to direct attention across an information display. Organization is critical to legibility and comprehension; sequencing, transitions, and connections all contribute to a productive reading experience. Users may construct an initial query and review key components of a resource such as headers, footers, headlines, and formatting to determine how to efficiently prioritize and process material on the page.
Design Review
Otter is an online transcription service, used in a variety of industries, to import audio files into relatively comprehensible text. The process of upload is simple as a user finds and selects their digital files for transcription that is available through the online portal (Figure 01, 02, 03, 04). The value of transcription across varying professions is incalculable, offering lasting documentation of exact words and phrasing. However, often transcriptions are lengthy and may require post-processing, review, or summary. Depending upon the purpose of the text, Otter’s platform must support variations of analysis and cater to both receptive and critical reading styles. While we are not formally trained in receptive reading, individuals do develop an understanding and approach to reading that is consistent with a search query strategy by reading for holistic comprehension. In contrast, conceptive or critical reading is performed for evaluative purposes of a resource and requires thorough investment of time and cognitive load. Once a reader receptively finds the appropriate area of material within their text, they are able to more closely review potentially rich content that aligns with their goal.
To cater to receptive reading, Otter has formalized the notion of Information Foraging Theory in its searchable key words at the top of the text resource. This feature enables readers easier access to critical points by performing a manual search to reduce the area of the searchable visual field. By eliminating this task, readers and analysts are able to quickly allocate attention to content.
Figure 01: Home page, Otter.ai
Figure 02: Import dialog, Otter.ai
In regard to general readability and text content features, Otter uses a ragged right text of appropriate font size without justification so as to ease readability by consistency in word spacing. Further, there is no hyphenation, which often elicits attention loss from readers. These factors contribute to readability. This point is critical, as Otter is primarily responsible for displaying an imported text transcription, of which Otter has limited content control. Further, the transcription clearly indicates breaks in speakers, providing shorter paragraphs for visual relief.
Additional design actions and practices ensure reader comprehension, some of which are employed in the Otter platform and some of which are not yet developed perhaps due to technology limitations. Design directives recommend starting with learning objectives or an abstract for additional anchors of identification. Otter users would likely benefit from a manual entry for an abstract/description or a platform generated one to supplement users in their search for information.
Conclusion
To conclude, humans are uniquely capable of self-reflective thinking via metacognitive processes, operating in parallel to primary cognitive activity. This phenomenon equips content designers to guide and enable users through the process of goal setting, self assessment, monitoring progress, and validating feedback. Understanding reading styles and reading psychology enables designers to create more compatible experiences and systems that people are more easily able to engage with.
Figure 03: Home page, Otter.ai
Figure 04: Import dialog, Otter.ai
Figure 05: Transcription page (top), Otter.ai
Figure 04: Transcription page (middle), Otter.ai
References
Ali, A. Z. M., Wahid, R., Samsudin, K., & Idris, M. Z. (2013). Reading on the Computer Screen: Does Font Type Have Effects on Web Text Readability?. International Education Studies, 6(3), 26-35.
Coutinho, S. (2008). Self-efficacy, metacognition, and performance. North American Journal of Psychology, 10(1).
Day, R. R. (1993). New Ways in Teaching Reading. New Ways in TESOL Series: Innovative Classroom Techniques. TESOL, 1600 Cameron Street, Suite 300, Alexandria, VA 22314.
Duggan, G. B., & Payne, S. J. (2011, May). Skim reading by satisficing: evidence from eye tracking. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (pp. 1141-1150).
Dunlosky, J., & Metcalfe, J. (2008). Metacognition. Sage Publications.
Fiset, D., Blais, C., Éthier-Majcher, C., Arguin, M., Bub, D., & Gosselin, F. (2008). Features for Identification of uppercase and lowercase letters. Psychological Science, 19, 1161–1168. doi:10.1111/j.1467-9280.2008.02218.x
Flavell, J. H. (1979). Metacognition and cognitive monitoring: A new area of cognitive–developmental inquiry. American psychologist, 34(10), 906.
Garner, R. (1987). Cognition and literacy.Metacognition and reading comprehension. Ablex Publishing.
Geurten, M., Meulemans, T., & Willems, S. (2018). A closer look at children’s metacognitive skills: The case of the distinctiveness heuristic. Journal of Experimental Child Psychology, 172, 130-148.
Gibson, E. J. (1969). Principles of perceptual learning and development. Englewood Cliffs, NJ: Prentice-Hall.
Jarrett, C., Redish, J., Summers, K., & Meloncon, L. (2013). Designing for people who do not read easily. Rhetorical accessability: At the intersection of technical communication and disability studies, 39-66.
Livingston, J. A. (2003). Metacognition: An Overview.
Mazur, B. (2000). Revisiting plain language. TECHNICAL COMMUNICATION-WASHINGTON-, 47(2), 205-211.
Metcalfe, J. (2008). Evolution of metacognition. Handbook of metamemory and memory, 29-46.
Pirolli, P., & Card, S. (1999). Information foraging. Psychological review, 106(4), 643.
Redish, J. C. G. (2000). What is information design?. Technical communication, 47(2), 163-166.
Sanocki, T., & Dyson, M. C. (2012). Letter processing and font information during reading: Beyond distinctiveness, where vision meets design. Attention, Perception, & Psychophysics, 74(1), 132-145.
Schraw, G., & Dennison, R. S. (1994). Assessing metacognitive awareness. Contemporary educational psychology, 19(4), 460-475.
Schwarz, N. (2004). Metacognitive experiences in consumer judgment and decision making. Journal of Consumer Psychology, 14(4), 332-348.
