The following aspects of learning are relevant to course design and operation.

Deep learning

The literature differentiates between shallow and deep learning. Shallow learning means that students memorize a lot of information for an exam. In students’ brains, knowledge elements are loosely connected among themselves, and with existing elements. This more-or-less guarantees that the information will be lost after the exam. This article explains.

Deep learning is typified by dense connections between knowledge elements. Connected knowledge resists decay. Deep learning takes more time than shallow learning, since it requires more and stronger connections. Fewer concepts can be included in a course that encourages deep learning than one that encourages shallow learning.

Which one a student uses depends on both the student and the course. Some students may want to learn a topic deeply, but be forced into shallow learning by a course’s structure. If their grade depends on learning lots of information for a test, students will use shallow learning.

Many courses discourage deep learning, no matter what students’ preferences.

Rich experiences encourage deep learning. Let’s talk about ways to create rich experiences.

Active learning

The first is, not surprisingly, hands-on use. This is one dimension of active learning. Active learning builds dense connections between knowledge elements.

Emotion and social interaction

Working with other humans can deepen understanding. One reason has to do with the fact that knowledge has affect tags. A knowledge element with strong affect will be recalled more easily than a “bland” knowledge element.

Social interaction is inherently emotional. A positive social experience can tag the knowledge involved, perhaps leading students to be more positively disposed to the information systems field. Of course, a negative experience can do the reverse.

Tasks, not tech

Exercises that have a relevant context encourage deeper learning. The canonical example is from math: solving abstract equations, versus story problems. The former is about the technology of math. Students have trouble applying abstract formulas to real tasks. Story problems help with that.

Exercises with a business context:

• Help link new knowledge with existing information, making knowledge construction (see below) easier.
• Motivate exercises. That is, they help students understand why one choice is better than another.

An exception to tasks-in-context comes from computer science education research. Students at the beginning of the course are coming to grips with tech basics. Adding context as well can overwhelm their limited cognitive resources (see below). Initially, it may be better to focus on learning tech basics, without the overhead of context.

This period should be short. Once students understand the tech’s basic structure, add context. However, when learning more advanced tech knowledge, it may help to switch back briefly to context-free learning.

Formative feedback

Formative feedback is more likely to lead to deep learning than summative feedback. Formative feedback gives students a list of things they did right and wrong for each exercise. If they have an opportunity to improve, they are more likely to use the feedback. Summative feedback gives a grade, without other information.

Summary

• Focus on core knowledge.
• Active learning.
• Task focus. Show concepts in the context of tasks, not in abstract terms.
• Formative feedback.

Constructivism

Knowledge is actively constructed in each student’s brain. Experiences builds on existing knowledge. If there are gaps in existing knowledge, the student may not be able to benefit from a new experience.

Humans have powerful visual systems. Presenting information in images as well as text (dual encoding) makes construction more likely, and deepens connections.

Learning is contextual. Knowledge that students construct in one context (e.g., an algebra course), will not automatically be applied to another (e.g., accounting). Students often have to be prompted to make the transfer.

Mindful abstraction is key to transfer. Patterns can help. See this article for more.

Implications:

• Sequencing is important. Think about prerequisite knowledge during course design.
• Present information visually as well as in text.
• Patterns can help with transfer.

Cognitive limits

Humans have limited ability to process information. They are easily overwhelmed.

Some of that capacity is used when decoding information. That capacity is not available for learning. The lower the decoding overhead, the better.

As mentioned, humans are good at image processing. Images can reduce cognitive load, compared to text.

Chunking reduces overload, by combining multiple pieces of information into one piece. For example, 2-4-8 is three pieces of information in Florida. In Oakland County, Michigan, it is one piece: the local area code. Human brains move into and out of chunks, shifting their level of abstraction as they think.

Structuring content may aid chunking. Keep lessons short, and explain how lessons are connected into larger conceptual units.

Patterns (see above) can help with chunking. To an experienced programmer, “use a loop to read records from a database query into an array” (a common programming pattern) is one chunk. When writing the code, the programmer descends into the chunk, focusing attention on the chunk and its components. Once the code is finished, the programmer moves up a level of abstraction, and again treats the code as one chunk.

Scaffolding can reduce lesson complexity. Scaffolds do some work for students, e.g., giving them part of the code for a program. Scaffolds should be removed over time.

Implications:

• Make content easy to process. Eschew obfuscation.
• Use images.
• Keep lessons short.
• Add structure to encourage chunking.
• Give students common patterns.
• Scaffold difficult tasks initially. Reduce scaffolding over time.

Metacognition

(I use the term generally, to refer to any beliefs and feelings about learning.)

Students have fixed or growth mindsets. A fixed mindset suggests, for example, that people are either naturally good at math, or not. If you’re not, there’s no use trying to learn. The effort is wasted. A growth mindset suggests that people can learn math, though it takes effort.

The BIS course should encourage a growth mindset. Simply explaining the difference between fixed and growth mindsets can have an effect.

Early wins help students believe that they can learn the material. Early defeats can have the opposite effect.

Emotions matter. Computer work is frustrating. Students need to be able to continue, regardless. The course should prepare students for frustration, and offer advice for dealing with it.

Implications:

• Encourage a growth mindset.
• Give students early wins.
• Prepare students to be frustrated and angry.