Unit #4: Case-based, Cognitive Flexibility & Learning Objects

Similarities & differences…

The two learning models presented, case-based and cognitive flexibility, share a number of characteristics including:  context-based learning; a focus on problem-solving and decision-making skills; and incorporation of cases.  Both models seem more suited to instruction of advanced subjects.  There was frequent reference in the readings to use of these models in learning for professions (e.g.,  medicine, law, teaching, business, etc.). 

There are some distinct differences between the two approaches.  The case-based methods have a distinct, linear process (e.g., defining the problem, collecting and analyzing the data or facts, identifying potential solutions, analyzing the pros and cons, and deciding upon a final solution.) 

The cognitive flexibility model also analyzes complex problems, but does not do so in a linear approach like case-based learning.  The non-linear, less structured approach of the cognitive flexibility model is intentional, with the thinking that the student will be better able to adapt and transfer the new knowledge learned to new situations.  Jacobson & Spiro identify the following problems with learning complex material by traditional methods that they feel are overcome with the cognitive flexibility approach:  oversimplification and compartmentalization of knowledge, and a focus on memorization.  Jacobson & Spiro reported how cognitive flexibility approaches led to improved ability to transfer knowledge.

Learning Objects are different from both of the above, because they are flexible content resources for the online environment, but they are not instructional models.

Initial reactions…

Personally, I have had a lot of exposure to case-based learning and feel one of the most useful aspects is that the student learns a methodical process by which to approach and solve complicated problems.  The process really sticks with you and is very adaptable to complex problems in many different fields. 

The cognitive flexibility approach seems to be similar to learning that occurs while doing original research for a masters or PhD.  There is a lot of relatively unstructured searching and learning in order to develop a framework within which the problem can be solved.

Barriers…

One main downside mentioned for case-based reasoning is that if inappropriate cases are chosen as the basis for solving a new case, this will result in an inappropriate solution.  This is where instructor intervention is needed.

I am not a huge fan of the ill-structured environment of the cognitive flexibility approach, but I do see its value in terms of learning that is readily adaptable to new and different situations.  However, I do wonder if this method could actually lead to over-complication and confusion in some situations.  It seems like instructor guidance is particularly crucial for this method to be effective.

Learning objects seem to have great potential as flexible, reusable content resources for online learning.  Some of their barriers include the question of quality, the need for financial resources to create, and possible loss of contextual learning.

Use of the models…

I frequently use case-based learning methods.  The students learn to solve a realistic problem and usually they are intrigued by the stories presented in the cases.  I think this is a very engaging way to present content.  In teaching pharmacy technology students, I will often use case studies that involve a medication error and they will analyze the problem, why it occurred and how it could have been prevented.  Studying a case where a patient dies as a result of a pharmacy error, really hits home.

High quality learning objects are a resource that I would definitely use and I hope will increase in availability. 

Web-based tools or resources…

The case-based and cognitive flexibility models are very adaptable to online environments and probably more effective when presented this way.  Links to many resources (websites, graphic, video, etc.) can be provided from a single point, simplifying presentation.  Cases can be presented more realistically using video instead of text.

Learning objects can utilize text, embedded video, voice, animation and online assessment (Pro-profs, etc.)

Unit #3 - Context Models

Similarities

The Unit 3 models use different means to reach the same end which is learning in context.  The models assume that learning is more effective when it occurs in an environment that is closer to a realistic situation.  The models have iterative characteristics, building in repetition and revision until the desired end-point is achieved.  There is also a focus on higher-order skills including analytical, problem-solving, decision making and critical thinking.  All seem to be suitable for use by groups or individual learners.

Differences

There are distinctive differences amongst the models, providing a unique angle for each.  For example, a main focus of goal-based scenarios is identifying a meaningful motivator for learning.  Anchored instruction incorporates the aspect of appreciating the value of a discipline through the learning process.  STAR legacy uses a cyclical, step-wise approach, providing a visual map as a navigational tool.  An early step in STAR legacy includes recall of prior information as a way of inventorying knowledge and improving self-assessment of learning needs.  MOST emphasizes use of multimedia and especially visual media for enhancing literacy development.

Initial Reactions

My initial reaction was that these are all relatively complex models compared to the earlier models we learned about.  In thinking about why the models are so complex and about how involved the natural learning process is, I began to better appreciate the models.  The contextual models more closely align the instruction to the biological process of learning.  For example, Schank et al. discusses how information often cannot be retrieved from memory when learned out of context (e.g., for a test).  His goal-based scenario model promotes “learning by doing”.

When I was in graduate school, we had to pass a comprehensive oral/written qualifying examination in order to progress and do our independent research.  Every year there were several students who failed the exam.  I can remember the professors lamenting about how these students knew lots of information, but had absolutely no concept of “the big picture” or the relevance of the information they knew.  I think such gaps are much less likely when learning occurs in context.

Barriers

Designing instruction based on pathways essential to learning makes sense.  However, the downside to these models in my mind relates to whether they can be effectively implemented for wide scale use in practice, due to their complexity.  For example, Schwartz et al. (1999) describe how STAR legacy, a flexibly adaptive model, evolved from difficulties experienced with practical implementation of the Jasper anchored instruction program.

The model that I would be most likely to try in practice would be STAR legacy.  The stepwise cycle provides a logical progression to follow and a relatively flexible approach. 

Web tools/resources

The web seems particularly useful in presenting the context models.  Video, images and text capabilities enhance the ability to simulate a real-life challenge.  The ability to link to various web sites or pages simplifies the presentation.  Multimedia is explicitly incorporated in MOST, with the assertion that a combination of video/audio/print/oral media (e.g., tools could include voice thread, YouTube, etc.) accelerates literacy development.

Unit 2 - Collaborative & Group Models

First impressions…

      I was impressed by the research substantiating improvements in learning when the various cooperative and group models are employed (Johnson et al. 1998; Haller et al. 2000; Hung et al. 2003).  Plus, learning clearly goes well beyond content, with documented improvements in critical thinking, problem-solving skills, communication, teamwork, attitudes toward learning, etc. 

      While the outcomes research is very supportive of cooperative and group instructional models, I also got the impression that success requires a very skilled instructor willing to expend significant effort, due to many difficult-to-control factors (e.g., student reluctance, personality issues, group dynamics, etc.)  The instructor’s job is more challenging because in addition to content expertise, there is a need to take on a facilitator role, teach skills for group interactions, etc.

      Finally, I was struck by how aptly Johnson et al. (1998) tied the individualistic focus of our American culture to the observation that cooperative approaches have taken a backseat to individualist/competitive approaches to learning in the US.

Similarities & Differences…

      The theories covered in Unit II have many commonalities, including group interactions, active learning, a student-directed component, and the ability to engage higher-order thinking and problem-solving skills.  Guided design and PBL share the common approach of using a problem that forms the basis for learning.  All of the methods seem particularly well suited for preparing students for future success in the workplace.  The cognitive apprenticeship seems least similar to the other models in that group interactions are not an integral part and learning occurs along side a teacher. 

Use in Practice…

      In my teaching, I do try to make use of these theories.  In general, the cooperative/group theories are great, but my impression is that they work best with mature learners, possibly because of their prior experience with group interactions.  In the community college setting, we have a heterogeneous mix of students.  The less experienced, and usually younger students can be quite negative about engaging in group work, which can be a barrier.  But, this is not to say that they will not benefit from the experience.

      I use a modified jigsaw (cooperative) approach (where the students teach each other) for anti-depressant drugs.  In my experience with jigsaw, I have noticed the potential for “less than adequate instruction” noted by Blocher (1980).  To hopefully counteract this negative, I post content outlines at the conclusion of the session. 

      While teaching pharmacy calculations, I use a cognitive apprenticeship approach. Nearly all of the content is presented in terms of “real-life” pharmacy math.  I often see the problem described by Collins et al. (1989) where students struggle because they rely on recognition of standard patterns of problems, rather than problem-solving skills.  Interestingly, this “mindset” is a major stumbling block and the course has a relatively high failure rate.  The readings made me think about using more modeling and scaffolding to foster development of necessary cognitive skills.

      The PBL approach, using case studies, is a great approach for learning topics such as ethics, disease treatment, drug interactions, etc.  Guided design is a great method too, however, it seems better suited for particular fields such as business and engineering, with complex problems having multiple potential solutions. 

Web adaptations…

      Tools that can assist in adaptation of these models to online environments include group workspaces (e.g., Google groups, etc.) and communication tools (e.g., chat, e-mail, Elluminate, discussion forums, etc.).  Koschmann (1995) described specific tools for PBL in medical education, such as video for case presentation, data banks for information queries and data sharing groupware.  Recently, at Clemson Univ, I saw a demonstration of a computer-simulated patient that can be verbally interviewed by nursing students.  The “patient” responds verbally with appropriate answers.  Development of this technology could significantly enhance web-based PBL for medical education. 

ECI 517: Unit 1 Blog

My initial reaction to the PSI and A-T models is that while they are relatively older instructional models (vintage 1960s), both have experienced continual interest and implementation, albeit with some modifications.  For a theory or model to persist in the field for nearly 50 years suggests that it has validity and usefulness.  Another initial reaction I had, particularly after reading Kulik et al. (1979) and Kozma et al. (1978), is how difficult it is to compare and assess the effectiveness of various instructional models (e.g., A-T vs. lecture-based instruction).  The Kulik et al. (1979) article points out how numerous studies comparing the same two instructional methods can have vastly different results, emphasizing the need to critically evaluate the literature before jumping to any conclusions about a particular theory or model.  This is striking because my experience is mostly in the biological/physical sciences where studies have a limited number of variables and can be very tightly controlled and hence such high variability is not as common.  Comparing two models of instruction is apparently very challenging due to the number of variables, etc.

The PSI and A-T models share many common characteristics including an individualized approach, self-paced instruction, clear learning objectives, breakdown of content into smaller units and upfront time and financial investments to develop materials.  Both models tend to be effective at conveying material but are not particularly suited for developing higher order analytical or critical thinking skills.  Both models seem very adaptable to a web-based system of instruction, particularly since instruction is divided into small units.

A significant barrier for both of these models would be student populations with lack of maturity, motivation and persistence.  For example, students who procrastinate may not be successful with an individualized / self-paced model of instruction.  Obviously, since the A-T model focuses on the audio modality for delivery of material, this model would not be suitable for hearing impaired students or students who have a strong preference for non-auditory styles of learning.

As for my own teaching, I do like the PSI model.  This model seems well suited for health science content such as pharmacology, because there is usually a large volume of material that can be readily divided into small segments.  The practice of breaking content into small chunks tends to improve comprehension and gives the perception that the subject is manageable as opposed to completely overwhelming.  Often pharmacology content requires solid comprehension of introductory concepts, and so the PSI principle of regular assessment to demonstrate mastery before progressing is particularly desirable.

I would not choose to use the A-T method, mainly because of my own personal bias (probably not a good reason).  Generally, I prefer non-auditory styles of learning.  Plus, my introductory biology course at Ohio State in the 1970s was taught by the A-T method (it was interesting to learn that the A-T method was initially developed for college biology instruction at neighbor Purdue Univ).  Although the course was very well done, my impression was that it was sort of a “discount” brand, providing a cheaper and more efficient way to teach the thousands of students needing Bio 101!  However, I do believe that elements of the A-T approach can be successfully incorporated into modern web-based instructional approaches.