Learning theory (education) - Cognitive Processes and Transfer

Cognitive Theories of Memory and Learning Introduction Understanding how people learn and remember information is fundamental to educational psychology and instructional design. Over the past several decades, researchers have proposed several influential theories that explain the cognitive processes underlying memory and learning. This chapter explores the major theories that have shaped how educators design instruction—from early models of how information moves through memory, to modern theories explaining how learners integrate multiple types of information, to principles about how knowledge transfers across different contexts. The Atkinson–Shiffrin Model: The Multi-Store Model of Memory In 1968, Richard C. Atkinson and Richard M. Shiffrin introduced the multi-store model of memory, one of the most influential frameworks for understanding how information is stored and processed. This model proposes that memory consists of three distinct stores that information passes through sequentially. Sensory Memory is the first store, where information from our senses arrives briefly. This store can hold a large amount of information, but only for a very short duration—less than a second for most sensory information. For example, when you read a line of text, all the visual information enters sensory memory initially, but it quickly fades unless you pay attention to it. Short-Term (Working) Memory is the second store, where information you're currently thinking about is held. Information can remain here for about 15-30 seconds, and it's where active processing occurs. However, this store has a limited capacity—typically around 7 ± 2 items (you might remember a 7-digit phone number temporarily, but 10 digits becomes much harder). When you solve a math problem in your head, you're using your working memory to hold numbers and intermediate results. Long-Term Memory is the third store, where information can be stored for extended periods—from hours to a lifetime. This store appears to have essentially unlimited capacity. Once information is successfully transferred here, it can be retrieved later when needed. The key insight of this model is that information must progress through these stages: sensory memory → short-term memory → long-term memory. Information can be lost at any stage if it's not adequately processed. This model explains why repeating information (rehearsal) helps move it into long-term memory, and why distractions cause us to lose information before it can be transferred to long-term storage. The Baddeley–Hitch Working Memory Model: Beyond Simple Storage While the Atkinson–Shiffrin model was groundbreaking, it treated working memory as a simple, unitary storage system. In 1974, Alan D. Baddeley and Graham J. L. Hitch proposed a more nuanced view called the working memory model, which describes working memory as an active system with multiple components, each specialized for different types of information. The working memory model includes: The Phonological Loop handles verbal and acoustic information. This component stores sounds and spoken words, allowing you to temporarily hold and manipulate language. When you repeat a phone number to yourself to remember it momentarily, you're using your phonological loop. The Visuo-Spatial Sketchpad handles visual and spatial information. This component stores images and spatial relationships, allowing you to mentally visualize objects or navigate space. When you mentally rotate an object or visualize directions to a familiar location, you're using your visuo-spatial sketchpad. The Central Executive is an attentional control system that directs attention, coordinates the other components, and performs higher-level processing. This is the "boss" of working memory—it decides what information to focus on and how to manipulate it. Episodic Buffer (added later) provides additional capacity and integrates information across different types, binding together visual, spatial, and verbal information. The critical difference from the Atkinson–Shiffrin model is that Baddeley and Hitch showed that working memory is not a single, unified space. You can simultaneously hold verbal information in the phonological loop and visual information in the visuo-spatial sketchpad without them interfering as much as you might expect. However, tasks that both require the phonological loop (like solving a verbal math problem) do interfere with each other. This model has profound implications for education: it suggests that presenting information in multiple modalities (words and pictures together, for example) can be more effective than presenting the same information in only one modality, because different types of information use different parts of working memory. Cognitive Load Theory Building on our understanding of working memory, Cognitive Load Theory (developed by Thomas deJong and others) explains why cognitive resources matter so much for learning. The theory starts from a simple but powerful premise: working memory has limited capacity, and the amount of mental effort required to process information—the "cognitive load"—directly affects how much students can learn. Cognitive Load Theory distinguishes between three types of load: Intrinsic Cognitive Load is the inherent difficulty of the task or material. Learning calculus has higher intrinsic load than learning basic arithmetic because the concepts are more complex. Extraneous Cognitive Load is the mental effort required by poorly designed instruction or irrelevant information. For example, if a textbook describes a complex biological process while simultaneously showing flashy animations that aren't related to the explanation, it increases extraneous load without improving learning. Germane Cognitive Load is the mental effort that directly supports learning. This is the productive, useful cognitive work—organizing information, connecting it to prior knowledge, and building mental models. The central principle is that total cognitive load = intrinsic load + extraneous load + germane load. Since working memory capacity is limited, if extraneous load is high, there's less capacity available for germane load, and learning suffers. The implication is clear: effective instruction minimizes extraneous load and optimizes germane load, allowing students to focus mental effort on genuinely productive learning. Cognitive overload occurs when the total load exceeds working memory capacity. When this happens, students cannot adequately process information, new knowledge is not properly encoded into long-term memory, and learning fails. This is why a textbook crammed with dense paragraphs, multiple side-notes, and competing visuals may actually impede learning compared to a clearer, simpler presentation. Multimedia Learning The previous theories all suggest that how information is presented matters enormously. Multimedia learning—the use of visual and auditory channels simultaneously—provides a powerful way to enhance learning, but only if designed correctly. The key insight is based on cognitive capacity: humans have separate processing channels for visual/spatial information and auditory/verbal information. When a learner sees an animation with accompanying narration (rather than on-screen text), the visual system processes the animation while the auditory system processes the narration. This distributes cognitive load across two channels, reducing overload. If that same information were presented as an animation with written text overlaid, it would force the visual system to process both simultaneously, likely causing cognitive overload. Richard Mayer's Multimedia Learning research (2009) identified several key principles for effective multimedia instruction: The Modality Principle states that presenting information through both visual and auditory channels is more effective than presenting it through only one. Accompanying visuals with narration (not text) often works better than visuals with text. The Redundancy Principle warns against presenting the same information in multiple forms simultaneously. If you show an animation and provide identical narration and include the same text, you're creating extraneous cognitive load by forcing learners to process redundant information. The Coherence Principle suggests that including interesting but irrelevant material (decorative graphics, interesting but off-topic information) increases extraneous load and decreases learning. Keep multimedia focused on the learning objective. These principles directly apply Cognitive Load Theory: they're designed to reduce extraneous load while preserving germane load. <extrainfo> Multiple Intelligences: While this outline mentions "Multiple Intelligences and Multimedia Learning" as a heading, multiple intelligences theory (Gardner's theory that people have different types of intelligence—linguistic, logical-mathematical, spatial, musical, bodily-kinesthetic, interpersonal, intrapersonal, and naturalistic) is not covered in sufficient detail in this outline to warrant explanation. Multiple intelligences theory is often discussed in educational contexts but is less directly integrated into memory and cognitive load models. </extrainfo> Transfer of Learning All the theories so far help explain how information is processed and stored, but educators care about a deeper question: how do students use what they've learned in new situations? Transfer of learning is the application of knowledge or skills learned in one context to a different, often new context. Transfer is harder than it sounds. Historically, Edward Lee Thorndike discovered that transfer is rare: students who master a skill in one context often fail to recognize that the same skill applies to a different problem. This finding challenged the assumption that learning automatically transfers. Understanding Transfer: Surface Structure vs. Deep Structure The reason transfer is difficult relates to how students perceive problems. Every problem has two aspects: Surface structure refers to the specific content and context—what objects are described, what numbers are used, what the problem looks like on the page. A problem about apples and oranges and one about boys and girls have different surface structures, even if the mathematical operations are identical. Deep structure refers to the underlying principles and solution procedures—the conceptual framework and steps needed to solve the problem. The "what to do" beneath the surface details. Research shows that students tend to focus on surface structure when learning. When they encounter a new problem with different surface features, they don't recognize it as the same deep structure, so they don't transfer their knowledge. For example, a student who masters solving kinematics problems might fail to recognize that the same mathematical principles apply to economics problems—because the surface structure is completely different. Factors That Promote Transfer Certain instructional features reliably promote transfer: Task features matter. Simulations allow students to apply knowledge in varied, realistic contexts. Problem-based learning forces students to use knowledge flexibly rather than just practicing isolated skills. Exposure to varied examples with different surface structures but the same deep structure helps students recognize the underlying principles. Learner features also influence transfer. Reflection on past experiences helps students connect current learning to prior knowledge. Participation in discussions exposes students to multiple perspectives and solution approaches. Students who develop awareness of their own learning processes transfer better. Bridging is a specific technique to enhance transfer. In bridging, students analyze relationships between a familiar task and new problems, with an emphasis on abstract principles rather than surface details. For example, asking students to reflect on what their past test performance reveals about effective study strategies helps them bridge to new courses. They move from thinking "I did this in chemistry" to thinking "this is a principle about how learning works," making transfer to other subjects more likely. Benefits of Transfer When transfer occurs, it has profound benefits. Students can rapidly acquire new tasks without starting from scratch—knowing how to learn one language facilitates learning others. Transfer supports language processing because grammatical principles transfer across contexts. Perhaps most importantly, transfer fosters higher-order cognitive thinking because it requires students to think abstractly about principles rather than just execute procedures. <extrainfo> Additional Research Context Contemporary research on transfer of learning has continued to develop these themes. Sandra M. Cormier and Jeremy D. Hagman's edited volume Transfer of Learning: Contemporary Research and Applications (2014) and Alison McKeough's Teaching for Transfer: Fostering Generalization in Learning (2013) provide comprehensive overviews of current thinking in this field. These works build on the foundational principles discussed above while exploring recent findings and applications. </extrainfo>