Surface and Deep Processing: Cognitive Behaviors of Aha! Moments (Part II)

This article is a continuation of a research entry from the May 24, 2021 edition:

The spectrum of early research on insight ranges from observing changes in behavior and understanding psychological patterns that influence learning (Bühler, 1907; Duncker & Lee, 1945; Wallas, 1926), to the present and how insight is a unique form of learning. There are a number of theories on insight; at present, no one theory dominates interpretation (Kounios & Beeman, 2015; Sternberg 1996). In spite of differences between theories, they share two principles: (a) sudden, conscious change in a person’s representation of a stimulus, situation, event, or problem (Davidson, 1995; Kaplan & Simon, 1990), and (b) the change occurs unexpectedly (Jung-Beeman, et al., 2004; Kounios & Beeman, 2014; Metcalfe, 1986). Further, a strong correlation has been demonstrated between moments of insight and increased engagement in learning, positive boost in mood, and greater likelihood of more moments of insight (Kizilirmak, Da Silva, Imamoglu, & Richardson-Klavehn, 2016; Kounios & Beeman, 2014). Aha! moments have been shown to increase and enhance memory performance (Ash, Jee, & Wiley, 2012; Auble, Franks, & Soraci, 1979; Danek, Fraps, von Müller, Grothe, & Öllinger, 2013; Dominowski & Buyer, 2000; Kizilirmak, Da Silva, Imamoglu, & Richardson- Klavehn, 2016), reliably grounded on insight’s proven ability to, “comprise associative novelty, schema, congruency, and intrinsic reward” (Kizilirmak et al., 2016, p. 1).

The observation and categorization of these moments can also be a source of valuable information for theorists and educators. Crocker and Algina (1986) demonstrate this operationalization in order to, “establish some rule of correspondence between the theoretical construct and observable behaviors that are legitimate indicators” (p. 4). The suddenness of Aha! moments makes observing behavioral changes (and subsequent changes in understanding) more dramatic and pronounced, as opposed to more gradual and deductively reasoned outcomes. Baker, Goldstein, and Heffernan (2010) have observed this distinction by studying the precise moment when understanding changes – graphing the precise moment of learning in humans. Baker et al. (2010) diagram the shift in surface to deep processing by showing the, “differences between gradual learning (such as strengthening of a memory association) and learning given to ‘eureka’ moments, where a knowledge component is understood suddenly” (p. 13).

Figure 5. A Single Student’s Performance on a Specific Knowledge Concept (Baker et al., 2010, p. 13)

Baker et al. explain that, “entering a common multiple” (left, Figure 5) results in a “spiky” graph, indicating eureka learning, while “identifying the converted value in the problem statement of a scaling problem” (right, Figure 5) results in a relatively smooth graph, indicating more gradual learning (p. 14).

Another important implication to consider is that deep processing seems to create greater investment in learning, along with more positive outcomes for students. Dolmans, Loyens, Marcq, and Gijbels (2016) have reviewed 21 different studies that reported on surface and deep processing strategies in relation to problem-based learning, and concluded that students using deep processing strategies use, “the freedom to select their own resources to answer the learning issues, which gives them ownership over their learning” (p. 1097). This ownership suggests a strong link between intrinsic and autonomous motivation, resulting in stronger and longer-lasting outcomes. Dolmans et al. also report that surface learning strategies with problem-based learning had a similar negative effect, stating:

a high perceived workload will more likely result in surface approaches to studying and might be detrimental for deep learning. Students who perceive the workload as high in their learning environment are more likely to display a lack of interest in their studies as well as exhaustion. This is particularly true for beginning [problem-based learning] students. (p. 1097)

““If we get the deep processing, we almost always get the surface, but with much richer and rewarding outcomes!””

— J. Littlejohn, Elementary School Math Instructor

The meta-analysis concluded by affirming these positive deep processing outcomes do not come at the cost of the various surface processing benefits (p. 1097). Deep processing strategies employed by learners have also been shown to boost long-term recall of information and wider conceptual understanding. Jensen, McDaniel, Woodard, and Kummer (2014) report that learners who utilized deep processing learning strategies while preparing for high-level assessments (i.e., problem solving, analysis, and evaluation) performed better than students that did not, and these students retained a, “deep conceptual understanding of the material and better memory for the course information” (p. 307). Jensen et al. (2014) have found that this higher level of cognitive processing and understanding also made transfer-appropriate processing more likely. This conclusion is supported by similar research conducted on learners using deep processing strategies and motivated by deeper conceptual understanding (Carpenter, 2012; Fisher & Craik, 1977; McDaniel, Friedman, & Bourne, 1978; McDaniel, Thomas, Agarwal, McDermott, & Roediger, 2013). Students using transfer-appropriate processing outcomes showed improved mastery and conceptual development greater than surface strategies and beyond the at-hand assessment; the gains were greater in current work and also in future assessments utilizing deep processing strategies. This developed processing strategy offers learners the greatest advantage in future outcomes. Studying Aha! moments in learning makes understanding surface processing and shifts into deep processing more probable, and the transfer-appropriate advantages more common, offering teachers a tremendous perspective into how to best develop pedagogy.

Surface and Deep Processing: Cognitive Behaviors of Aha! Moments

Marton and Säljö (1976) have produced a hierarchy for different levels of learning in order to determine, “processes and strategies for learning to be examined” (p. 1). Marton and Säljö were motivated to understand more than what a student may have learned, they were interested in understanding how much was learned. Marton and Säljö are the first to have produced surface-level and deep-level processing definitions and formulated research outcomes.

Hattie and Purdie (1998) sum up varying studies on both styles of processing and showed that surface approaches in learning, “involves minimum engagement with the task and typically focuses on memorization” (p. 4), with a key indicator being that learners reflect little or not at all on concepts, while a deep-level thinking approach “involves an intention to understand and impose meaning” (p. 4). A consistent aspect reported with deep-level thinking is the relationship students seek between concepts, creating more of, “an intrinsic interest in learning and understanding” (p. 4). Dinsmore and Alexander (2012) have reported on prevailing assumptions surrounding the notions of deep and surface processing. It is commonly believed that deep processing, “promotes better learning outcomes, while surface processing promotes weaker learning outcomes” (p. 500). The most easily traced outcomes include limited recall and reproduction of content information, new concepts described anecdotally, and memorization or non- prioritized listings of facts. This style of processing places greater emphasis on learning for the sake of information regurgitation. Deep processing strategies are defined largely by a learner’s comprehension of materials, “together with the process of relating and structuring ideas, looking for underlying principles, weighing relevant evidence, and critically evaluating knowledge” (Dolmans, et al., 2016, p. 1089).

Contrary to the judgments typically attached to the comparisons of surface-level vs. deep-level processing, a number of theorists have convincingly argued against a value-laden dichotomy, and that both forms of processing are equally important and necessary for learning, each offering unique advantages. An example of the favoring of deep-level over surface-level processing is found in Jay, Caldwell-Harris, and King (2008), who state that, “a shallow level of processing is one in which only superficial or physical aspects are encoded. A deeper level of processing takes more time and effort to activate the semantic meaning of stimulus” (p. 85). Jay et al. (2008) report also that surface learning often, “results in poor recall,” and that concepts that receive deeper levels of processing, “persist longer” (p. 85). However, it is important to recognize that by better understanding the relationship of surface and deep processing, teaching practice can be more positively informed and learning outcomes may be improved. This can be accomplished by studying how surface or deep processing assists or inhibits learning in a variety of contexts.

As Dinsmore and Alexander (2016) report, “the most advantageous level of processing for two individuals may not be or perhaps should not be the same but should depend on their stage of development [of concepts] and their performance goals” (p. 214). This challenges the assumptions that deep processing is always a stronger approach for outcomes and that the development of surface processing is invariably weaker. Alexander and Judy (1988, p. 391) have demonstrated that, “knowledge and strategic processing were inextricably intertwined [and that]... processing cannot be addressed in isolation” (Dinsmore & Alexander, 2016, p. 214). The elected processing approach, whether surface-level or deep-level, is utilized based upon a learner’s goals and nature of the learning activity (Biggs, Kember, & Leung, 2001; Entwistle & McCune 2004; Lonka & Lindblom-Ylånne 1996; Loyens, Gijbels, Coertjens, & Côté, 2013). However, evidence is inconsistent, and definitions and methods of studying the processes are sometimes ambiguous, and attempts to codify processes for reporting are, to date, incongruent (Block, 2009; Dinsmore, Alexander & Loughlin, 2008; Heijne-Penninga, Kuks, Hofman & Cohen-Schotanus, 2008; Murphy & Alexander, 2002).

Aha! moments, which can also be understood as moments of insight, could arguably indicate the point at which surface processing begins to transform into deep processing. A number of studies have identified insight as its own unique form of learning, and have argued that insight-learning has an effect on the higher executive functioning of the mind (Kounios & Beeman, 2015). It has been shown that when students are engaged in insight learning experiences and also succeed in solving problems in this way that there is a noticeable increase in long term memory (Cranford & Moss, 2012), and the production of a positive emotional state (Ekman, 2006; Fredrickson & Branigan, 2005). Further, it has also been shown that strong, positive learning habits are reinforced when students experience Aha! moments (Kounios & Beeman, 2014; Mednick, 1962).

Modern research on insight and how sudden changes in cognition influences learning dates back to the early twentieth century (Wallas, 1926). Dr. Karl Bühler (1907) coined the phrase “Aha- Erlebnis,” or translated: Aha-Experience. Bühler used this phrase to describe a moment in learning when “suddenly, the lights come on!” (p. 341). Bühler was extremely curious about the cognitive connections that were made, and how learning changed, as the result of Aha-Erlebnis. Bühler states,

it is clear that our AHA-experience leads us always to the deeper sense/meaning; you could say that this deeper understanding/comprehension may always be preceded by a “shallow” or superficial comprehension – just understanding the meaning of the words – and that constitutes the preliminary whole... but we would [now] have the whole thing... where we can prove it, at the deeper understanding, made comprehensible. (p. 16)

Check back soon for Part 2!

“The information given ”...were alone, completely insufficient to answer the question.””

— M. Franklin, Secondary School Science Teacher