Ernst Z. Rothkopf is Cleveland E. Dodge Professor of Education, Emeritus at Teachers College, Columbia University, where he taught from 1985 until his recent retirement. He earned his Ph.D. in experimental psychology from the University of Connecticut and he has served as President Division 15 of the APA, Editor of Review of Research in Education, and has served on the Editorial Boards of Cognition and Instruction, Educational Psychology Review and the Journal of Educational Psychology. He has edited four books, published over 120 articles and has been a major influential figure in Educational Psychology for the last five decades.
NAJP: What are you currently working on, researching and writing?
ER: One of the areas in which I have been working, (on and off), for the last twenty years is the integration between background and foreground during learning. Incidental and optional elements of teaching and studying situations such as location, setting, emotional states, or the surface characteristics of examples or word problems, become automatically linked to targeted instructional information, and can later play a role in retrieval.
For example, students recall more when they are tested in the place that they have learned, or are listening to the music they heard during studying, and they are more successful in applying abstract principles or procedural rules to concrete domains that have been used as examples or exercises during teaching. General solution rules which apply to large problem spaces are applied more quickly and accurately to combinations of specific instances to which they have been applied before. I use the term particularization to refer to these adventitious associative linkages between the particulars of individual's experience and general knowledge. The notion of particularization helps us understand how one becomes an expert.
NAJP: What do you mean by "particularization" and what is its role in educational psychology?
ER: By particularization I mean the associative linkages between surface elements of problem situations and deeper knowledge elements..
Three findings from this work are particularly interesting:
1. Problems that submit to general rule-governed solutions can be solved either a) by applying a rule-governed algorithm, or b) by directly recalling an instance like the present problem to which the algorithm has been applied in the past. When both processes are initiated, (b) is faster and wins the race.
2. Particularization offers an (at least partial) account of expertise. Becoming an expert means increased sampling of the specific instances in the problem space. One reason that experts become faster is that more of their solutions come from direct memorial access to previous instances and fewer require slow algorithmic solutions.
3. The linkage of specific situational factors with instructionally foregrounded knowledge provides redundant retrieval paths and addresses for the targeted knowledge elements. The use of technological information portals such as computers and video monitors tend to leach situational context from instructional episodes (e.g. in computer-based distance education) and this has adverse effects on performance. Such leaching can be overcome by diversification of instructional locations and other means--and that is one important reason for context enrichment.
NAJP: Many years ago, you wrote about "mathemagenic activities" What is the current "state of the art" of these activities or have they fallen by the wayside? How do we control these activities?
ER: Historically, this concept emerged for me in the early sixties. The operant conditioning people were (rather unthinkingly) equating reinforcement with knowledge of results in acquiring substantive knowledge through programmed instruction. That was obviously incorrect for two reasons. First, because the cessation of feedback (KoR) did not result in anything like extinction of verbal knowledge, and second, because KoR was not a necessary condition for such learning. …