Marsha Lovett

Cognitive Psychology; acquisition and transfer of problem solving skills

What are the findings or theories from your area of expertise that we could apply to higher education?

There are two empirical studies that I point to whose results can be used to improve instruction. In particular, these exemplify some of the simple things that can be done to enhance transfer and generalization of knowledge. 

The first is by Cummins (1992: JEP:LMC) in the domain of algebra.  She found that when subjects were given simple instructions to compare and contrast analogous problems, they were much more likely to induce a generalized problem schema (& transfer the problem solving skills).  This finding helps to illustrate that some of the things we can do are very simple: participants benefited from the comparison task without even solving.  It is often just a matter of asking questions in a way that prompts students to provide their own self-explanation or will allow some other processing that will allow students to generalize their results (see recent work by Chi that extends her work on the self-explanation effect). 

The second study is by Catrambone (1998: JEP:G; 1996: JEP:LMC; 1994: Mem&Cog) regarding procedural skills and schemas.  He was able to demonstrate that if you highlight the structure of problem solving steps in examples presented to subjects, the subjects are more likely to generalize the procedure to other problems.  He found that this was even effective when subgoals were highlighted by being formatted into groups with arbitrary labels.  His explanation for this was that highlighting the steps to students helps to lead to self-explanation.

Another important contribution to this is Singly and Anderson’s theory of transfer of skill (Singley & Anderson, 1989).

What are the (most important) unsolved problems?

The first one involves the research on epistemology and students’ beliefs about learning.  There has been a considerable amount of work classifying students’ beliefs about knowledge or describing the impact of metacognitive beliefs on learning.  Much less work has been done on understanding how students change their views on learning.  Often this change has been thought to occur naturally as a result of maturation, but an unanswer question is: can we create interventions that will help to effect this change.

For example, students’ typical belief that learning is quick and problem solving is immediate is an obstacle to their learning: they think that if they can’t do something right away it can’t be done. The belief that authority on knowledge comes from “the teacher” or is “in the book” can also impede deeper learning.  Again, the unsolved problem is: How can we help to change these beliefs?

The second issue involves the link between motivation and cognition.  People tend to study either cognitive issues or motivational issues; an important problem is understanding (and dealin with) the interactive link between the two.

A final issue is how students should be scaffolded in their instruction (both computer-aided and other forms of instruction) and how and when this scaffolding should be removed. There is empirical work on this topic but not much theoretical development in this area. I believe that framing the empirical results in a theoretical structure (and extending them) would help to guide instructors in how to apply these results in new situations.

What should be included in an agenda for research?

In developmental psychology, we have gone from emphasizing the description of developmental states to actually describing the changes between states and the mechanisms for accomplishing such changes.  In a similar way we could have people trying to understand the changes that occur in students’ epistemology (see first unsolved problem above), not just describing a student’s state of mind but describing the change process.  Additionally, it is important to get people collaborating on different pieces of learning from a variety of perspectives, ranging from the cognitive to the motivational. 

What prototypes can you point us toward where principles from the science of learning are already being applied (e.g., activities, courses, fields of study, degree programs, or entire systems)?

We are in the process of refining instructional design in an introductory statistics class at Carnegie Mellon. At each step of the redesign, we try to bring more of the principles of cognitive psychology and learning research into practice (cf. Lovett & Genovese, 1999 from The American Statistician; Lovett, in press: A collaborative convergence on studying reasoning processes: A case study in statistics. To appear in S. Carver & D. Klahr Cognition and Instruction: 25 Years of Progress. Mahway, NJ: Erlbaum).  We put theory into practice when we select the materials for the course and design activities that engage students in learning.  The last step of the process will be to incorporate a cognitive tutor to help students solve statistical problems based on their own level of understanding.  This work is based on cognitive tutors that have been very successful at the high school level in algebra and geometry classes.  The work of Ann Brown (Communities of Learning, I believe a relevant ref is from 1992 Journal of the Learning Sciences) is also a good example of interdisciplinary and collaborative work.

 What are the major problems with or barriers to redesigning higher education?

 Everyone is inventing their own wheel for solving instructional problems.  People are not building enough on the work that has already been done and they fail to fully explore what is already out there, both in terms of what works and what does not work.  One part of this problem is that the information is not readily available or easily accessible.  The other piece of the problem is more of a cultural impediment in that people get invested in “name brand work.” This produces resistance for people to build on other people’s work.

 Do you have any ideas for overcoming them?

 The lack of information problem could be addressed with searchable databases or archives that could provide a place to reference work that is being done in applying cognitive and learning principles to higher education, and  to document what works in different areas and what does not.

 What do we need to do so that one outcome of the retreat is to effect change (in ways that we want)?

After looking at some of the different perspectives on the most important problems, it would be helpful to find the commonalities in these problems and then to set out to have people attack these problems using shared resources.  One success would be to get people communicating and looking for ways to get people together for longer-term collaborative projects.  Herb Simon gave a speech at Carnegie Mellon on instruction and learning with a title something like, “Why isn’t teaching a team sport?”   This title gets at the cultural issue of the solo aspect to instruction and instructional design.  The statistics project we are working on is a very collaborative effort, with cognitive psychologists, instructional designers, statisticians, and software designers all working together to tackle this large problem.  This requires extra effort (from developing a common language for communicating to figuring out how to meet with 5 other busy researchers for whom this is not a major project), but it would be great to get people to work on and maintain long-term collaborations.