Benassi, V. A., Overson, C. E., & Hakala, C. M. (Eds.) (2014). Applying science of learning in education: Infusing psychological science into the curriculum. Retrieved from the Society for the Teaching of Psychology website: http://teachpsych.org/ebooks/asle2014/index.php
Reviewed by: Tim Gladson, Open Educational Resources Curator (email@example.com)
This anthology, informally published by the Society for the Teaching of Psychology (Division 2 of the American Psychological Association), provides an excellent review of research studies and underlying theories in the cognitive science of memory and learning—especially in the context of college students. The book explains the research behind many major learning principles and suggests practical ways that instructors at all levels can incorporate these principles into their teaching. The principles are well-established, although the editors acknowledge in the introduction that some of the principles still need to be researched in more depth to determine the circumstances under which they are most relevant and effective.
The first 13 chapters in Part 1 (all except the final chapter) each succinctly explain a well-documented learning principle, such as active learning or metacognition. After a brief summary, the authors synthesize numerous specific research studies, providing a sufficient literature review for instructors to understand the origin of these principles and their nuances. The authors often describe effect sizes, exceptions, and boundary conditions. Each chapter concludes with practical advice for instructors who want to implement these principles. Part 2 contains 4 chapters of additional advice for faculty seeking to incorporate learning science into their instruction. Topics include assessment techniques for instructional interventions, tips for professional development, and a summary of major student obstacles. Part 3 provides 6 case studies and research studies that apply principles of learning. Since all chapters in the book stand alone, readers may choose to read selectively based on their personal interests.
The authors intentionally use accessible language, without sacrificing substance. Even so, general readers may find it helpful to skim the details of various research studies, in order to focus on the general conclusions. Since the book is a labor of love rather than a commercial publication, it can be downloaded for free on the STP website. Unfortunately, there are a handful of minor typos and grammatical errors throughout the book due to a lack of professional copyediting. All in all, the book is a high-quality resource that will benefit instructors in any discipline. Learning how students learn can transform your teaching and assessment, leading to greater student engagement and success.
Key Take-Aways from Chapters 1-13
Learning and Cognition
Chapters 1, 3, 5, 9, and 11.
Learning is a multi-step process. We begin by encoding new information. We then organize information into a coherent mental model (schema) and integrate it with existing knowledge. Lastly, we retrieve information. The act of retrieval develops neural connections or pathways. Each time we retrieve and process information, that information will be retained longer and be more readily accessible. Likewise, skills such as problem-solving will become more automated. Among other factors, the success of learning can be defined by how long we retain information and how well we transfer knowledge to novel concepts or applications.
Both encoding and retrieval are much more effective when information is broken into manageable chunks, spaced out over time, and interleaved with other topics. Spacing and interleaving apply to all levels of learning, such as individual study sessions, individual classes, and entire programs of study. By contrast, massed instruction (essentially cramming) is quickly forgotten. Interleaving new information with the review of older information is usually more effective than when curriculum is strictly blocked into discrete units; this is caused by an inherent spacing effect, the comparison of discreet topics and examples, and the diminishing returns of massed repetition (as students essentially go into auto-pilot).
Many learning principles are based on cognitive load theory, which suggests that our brains have limited processing capacity. Information is processed in both sensory memory (raw data from our senses) and working memory; we are limited in both the amount of information that can be processed simultaneously, and how long information can be held in temporary storage. By contrast, long-term memory appears to have more storage capacity than we can use up during our life-times. If an instructor provides a learner with more information than they can process at once, the learner’s capacity for cognitive load is maxed out. Therefore, instructors should try to eliminate any extrinsic load (unnecessary distractions) and divide intrinsic load (necessary load) into manageable portions. Extrinsic load results in extraneous processing, whereas intrinsic load results in essential processing (selecting and organizing) and generative processing (integrating).
Learning can be enhanced or hindered by the prior knowledge that learners bring to a subject, in both accuracy and quantity. Correct prior knowledge gives students a starting point to build upon, while incorrect prior knowledge can be difficult to supplant. In terms of quantity, the efficacy of several instructional methods often depends on how much prior knowledge a learner has. Due to the expertise reversal effect, methods that assist novice learners are ineffective or even hinder advanced learners. Novice learners are assisted by prompts and extra explanatory information, but these tactics distract advanced learners. For complex subjects consisting of several components, novice learners learn best when the components are woven together, whereas advanced learners learn better when they can focus on discrete components and apportion their time to the topics that need more review (split-attention effect). Similarly, advanced learners benefit from complex tasks, such as studying errors (examples of incorrect reasoning or problem-solving) or self-generated questions and explanations about the material. Novice learners lack the necessary foundational knowledge for these advanced tasks, leading to cognitive overload.
The structure building framework models how our brains organize information while reading. As we progress through a narrative, we build mental schemas of the material and how various components relate to each other. As our understanding of the narrative deepens, we re-arrange and re-combine schematic structures into more integrated and complex structures. High-structure builders naturally construct more efficient, complex structures as they read, and they are more adept at distinguishing key information from superfluous details. Low-structure builders tend to build a larger quantity of smaller structures, containing more non-essential information; these readers must bear more cognitive load as they spend more effort re-arranging their schematic structures while reading. Instructors can assist low-structure builders by clearly identifying key topics beforehand, giving readers an initial blueprint to guide comprehension.
In order to minimize extrinsic load, instructors should omit any non-essential information (coherence principle) and clearly identify main points and how a lesson is organized (signaling principle). To avoid excessive intrinsic load, break up lessons into manageable chunks (segmenting principle) and begin each lesson with a brief overview (pre-training principle). Instructors may also be able to reduce intrinsic load by providing an engaging “social presence” (personalization, voice, embodiment, and image principles). These social principles suggest that instructors use everyday language and provide a pleasant persona in order to mimic normal human interaction [the authors do not discuss how social presence may decrease distractions and hence extrinsic load].
Language and images are processed by different parts of our brain (dual channels). Consequently, we learn better when we receive information simultaneously using both words and images (multimedia principle). The multimedia principle applies to any modality, from text and images in a physical book to voice and images in a lecture or video. However, multimedia learning can be hindered if words are received in multiple ways simultaneously (the redundancy principle, such as voice and written text) or if the multimedia elements are not synced and co-located properly (temporal and spatial contiguity principles). Similarly, if words are spoken rather than written (hearing vs. vision), the learner has more visual capacity available to process images (modality principle).
Chapters 2 and 6-10.
Learning is enhanced by exerting effort during both encoding and retrieval. The generation effect shows that students learn more when they actively create information during learning (such as taking lecture notes or summarizing a passage after reading it); by contrast, students learn very little through passive methods alone (such as reading text or listening to a lecture without taking notes). Students frequently ask instructors to solve problems that they could do themselves, or to provide extensive lecture notes. Instead, instructors should provide guidance and encouragement to solve problems independently and provide brief notes identifying key topics. When studying, methods that require retrieval of information (such as test questions) are much more effective than passive methods that try to re-encode information (such as re-reading a textbook or lecture notes).
Students generally receive most content by passively reading textbooks, followed by listening to lectures. Unfortunately, most students have a fairly limited capacity for reading (or listening) and comprehension. Likewise, most students take very poor reading and lecture notes, by both focusing on non-essential information and omitting key concepts. Although re-reading text usually doesn’t contribute to learning, advanced reading strategies such as read-recite-review and concept mapping can produce more efficient structure building. Unfortunately, advanced reading strategies take considerable training and practice to learn and may require more time than busy students are willing to invest.
Building on the concept of active learning, there are many forms of desirable difficulties that foster student exertion and deeper learning, as long as the difficulties are not excessively challenging. For example, giving students problems that are challenging or require novel applications of known principles can lead to robust learning. But if the problems exceed the learner’s capacity, they will probably only experience frustration.
There are several forms of active learning, which often overlap. Testing forces students to retrieve information and provides guidance on which concepts they need to study more. Research on the testing effect is distinct from assessment, as its purpose is to generate learning rather than to measure a learner’s competency. Testing can be integrated into reading assignments through embedded questions. Students can also test themselves with self-generated questions, provided they are trained in how to create meaningful questions. Questions that require analysis and synthesis are more valuable than ones that simply require recall of facts. Instructors may need to offer a small number of points to encourage participation in testing for learning.
Similarly, self-explanation involves a student explaining course material as they understand it, in their own words. Self-explanations generate recall and synthesis; they also promote metacognition about what the student has and hasn’t mastered. Worked examples allow instructors to demonstrate principles while modeling self-explanations. Worked examples are not limited to problem-solving applications, although they are especially popular in math and science instruction. Worked examples are especially effective when studying multiple variations of the same problem (to practice different strategies and to identify underlying principles) as well as related problems with the same cover story (to help students distinguish the nuances between similar principles). Complex problems should be broken down into meaningful building blocks, and various components of diagrams or equations should be color-coded for easy mapping between the components and the lesson principles. Worked examples demonstrating common errors help advanced students understand and avoid incorrect reasoning (although studying errors can be confusing for novice learners).
Unfortunately, students prefer study methods that require less effort, and they usually don’t generate quality questions or self-explanations on their own. Students need to be taught why these techniques are worth the effort, followed by training and modeling by instructors.
Chapters 1, 4, 12, and 13.
Learning is assisted by metacognition, the learner’s ability to accurately gauge how much they currently understand and how they need to improve. Metacognition can be seen as the awareness to effectively plan, assess, and modify learning strategies in an ongoing, iterative process. Although it is related to self-awareness, self-regulation, and motivation, researchers lack consensus on how broadly to define metacognition and how to define the relationships among these concepts [as of 2014].
Some researchers see metacognition as including a student’s concept of how learning works, as well as their self-perception of how proficient they are at learning. Many students falsely believe that their ability to learn is fixed (entity theory of intelligence); when faced with learning challenges, an entity attitude may result in learned helplessness and giving up. Teaching students about the brain’s ability to change over time (plasticity) and instilling a positive growth mindset leads to greater persistence and improved learning (incremental theory of intelligence).
Most students tend to overestimate their preparation for tests and their likely scores. Students with entity attitudes tend to engage in fewer metacognitive practices and have higher levels of overconfidence (or unwarranted lack of confidence) when they do assess their learning. These students also tend to engage in ineffective study strategies, to study less than they should, and to focus on the wrong topics. Since cramming strategies often provide very short-term boosts to recall of facts, these students may overestimate their actual learning progress (neglecting long-term retention and not distinguishing factual recall from deeper, conceptual learning). On the other hand, students with incremental attitudes tend to examine their learning more often and more accurately; they also tend to use more effective study strategies (such as self-explanations) and invest their study time more strategically. Likewise, students with incremental attitudes tend to be motivated by mastery of the content and skills, while students with entity attitudes tend to be motivated by measures of their performance; the pursuit of mastery is correlated with metacognition skills, but the pursuit of performance goals is not.
Unfortunately, metacognition skills are difficult to teach. The most effective method appears to be modeling best practices within the context of regular courses (such as thinking out loud while performing a worked example), along with encouragement and coaching. Group work and discussions can also be structured to inherently engage students in metacognition (although the assignment must be well-designed, and students must actively participate).
Emotional and Social Factors
Chapters 2 and 4.
Effective learning also involves emotional intelligence and social encouragement. For example, many students are embarrassed when they produce errors (such as incorrectly solving an algebra equation), as though this indicates an intellectual deficiency. Ironically, experiencing and correcting errors is actually one of the most effective pathways to learning. Instructors should encourage students to embrace errors as opportunities.
Students also learn more when they are motivated to learn. This includes having a clear picture of the learning journey they are on, having learning goals that complement their instructor’s teaching goals, and receiving regular feedback on how well they are progressing towards their learning goals (both formal and informal assessment, which promotes both motivation and metacognition). Instructors consistently overestimate the quantity and quality of feedback they provide to students (empathy gap). Due to limited resources, instructors also prefer giving group-level feedback rather than individual feedback; unfortunately, group feedback usually has little impact on students.
However, generic attempts to foster self-esteem and excessive, unmerited praise are usually ineffective methods of encouraging learners. Although well-intentioned, these methods can actually cause harm by eroding the instructor’s credibility or instilling a fear of failure. Although difficult to implement, personalized, one-on-one feedback is best.
If you don’t have time to read Applying Science of Learning in Education, I recommend the following short articles from the Encyclopedia of Cognitive Science (available electronically through the NU Library):