The field of developmental psychology is fraught with some very popularized but misunderstood dichotomies.  Nature versus is nurture is probably the most well known, but another important distinction is that between competence and performance.  Jeff mentioned a little bit about this distinction a few months ago in his post about desirable difficulties in the classroom but it is worth bringing up again because it’s a dichotomy that is pervasive throughout psychology.
The basic distinction is this: whenever we want to measure whether someone knows something (a concept, a skill, a procedure, etc.), we create a task for them that should require that target knowledge for successful completion of the task, and then we measure the subject’s performance on the task in order to infer whether they have that knowledge or not.  For example, if we want to know whether a student knows how to do algebra, we give him an algebra test, and then measure how well he does.  So far we’re only talking about performance.

Competence comes into the picture when we start thinking about how good our tasks are: is it possible that a student who does poorly on an algebra test really does know algebra and the test is just a bad one?  If we acknowledge that our tests might be bad, then we might observe poor performance even when participants are actually competent.  This distinction between observable performance and underlying competence is one that drives the field to continually seek better tasks and methods to assess competence.

However, this distinction can be quite problematic.  An unfortunate problem is that competence, by definition, cannot be directly observed – we must always use task performance as an indicator of competence.  BUT there is no way to know whether a failed performance is due to a bad task or a lack of competence.  A tendency, especially in research on infants, has been to seek new tasks rather than assess why old ones failed (Sophian, 1997).  If a new task produces successful performance, it in some ways “trumps” the results from unsuccessful tasks rather than contributing to a larger picture of what infants can and cannot do.  It’s kind of like saying, “See! They could do it all along, we were just using the wrong test” rather than acknowledging the significance of the fact that infants are successful at some tasks and in some situations but not others.

This notion is further complicated by the fact that most knowledge, skills, and procedures are not acquired in one fell swoop: as we learn we are unsuccessful many times before the first time we get it, and after that first time we will be unsuccessful many more times before we achieve consistent success.  So is it really fair to say, for example, that a toddler “knows how to count” the first time she counts to 10 without a mistake? Or is it better to wait until she can count with some level of consistency before we say that she “knows” how?  And even if she can recite the numbers from 1 to 10 without mistakes, does she know what she’s doing or what it means?  Can she count toys just as well as fingers or people or sounds?  Does a successful performance necessarily imply competence?  These questions are not easily answerable, but the truth is definitely more complicated than simply “knowing” or “not knowing.”

Some psychologists would even go so far as to say that the distinction between competence and performance is unnecessary.  Their argument is that competence doesn’t have any practical significance because everything we do in the world is performance.  In this view, it would be more appropriate to investigate what kinds of tasks produce successful performance and whether those tasks reflect real-life challenges with practical significance.  In the case of algebra students, for example, perhaps it doesn’t matter whether they have an underlying algebraic competence; what really matters is whether (and in what situations) they can use algebra to successfully complete real-world tasks like taking tests.  In that case, we don’t care about assessing competence, we care about improving performance.

My personal opinion is that the idea of competence does make sense on some level.  I have seen enough students struggle through algebra and physics to know that consistent successful performance requires an underlying conceptual basis that I don’t feel is performative by nature.  I believe that is important for cognitive development research to acknowledge and consider how competence (or conceptual understanding) changes over time and experience.  We are far from knowing exactly how conceptual change happens, but I think that pursuing the issue will give us further insight into how to improve students’ performance in those fields.  As long as the competence/performance distinction continues to produce good research questions, it will remain a cornerstone of the cognitive development literature.

For further reading about the competence/performance distinction in cognitive development (especially mathematics), I recommend Catherine Sophian’s work:

• Sophian, C. (1997). Beyond competence: The significance of performance for cognitive development. Cognitive Development, 12: 281-303.
• Sophian, C. (2008). Rethinking the starting point for mathematics learning. In: Saracho, O.N., and Spodek, B., eds. Contemporary perspectives on mathematics in early childhood education, 21-44. © 2008 Information Age Publishing, Inc.