Keller's Personalized System of Instruction: Was it a fleeting fancy or is there a revival on the horizon?

Abstract

Keller's Personalized System of Instruction dominated the literature in the teaching of psychology and behavior analysis in the 1970s and 1980s. After this brief flourish of interest, PSI research trickled off to a nearly imperceptible stream in the 1990s. However, with the increasing availability and ease of use of computers and the internet, along with demand created by an ever-growing need for distance education, PSI is beginning to curry favor among a new generation of faculty members. The question of whether this resurgence of interest will be permanent revival or fleeting fancy will depend on how researchers deal with the PSI paradoxes that stymied the researchers of the 1970s and 1980s. This article will review the current state of PSI research including the riddles left to be untangled, illustrate how computers have affected the PSI movement, and discuss the role PSI might play in distance education. Keywords: Personalized System of Instruction, PSI, Mastery-Based Learning, Distance Education, Fred Keller, Keller Method, Keller Plan, Mastery. ********** Keller's Personalized System of Instruction Keller created the Personalized System of Instruction (PSI) in the late 1960s in order to help students in Brazil be able to learn course material without an instructor standing by their side. Soon after, he brought his PSI program back to the United States. Due to its heavy reliance on behavioral principles, it was quickly adopted by many psychology professors and by individuals outside of psychology. Keller (1968) outlined five basic components that he deemed to be essential for a PSI class: (1) mastery of course material, (2) the use of proctors, (3) self-pacing, (4) stress upon the written word, and (5) use of lectures and demonstrations primarily for motivational purposes. In a standard PSI course, the course material is broken down into small units of study (e.g., one textbook chapter). The unit mastery component requires that students learn this small quantity of information and pass a test over this information by reaching some mastery criterion (e.g., 80%, 85%, 90%, or 95% correct). If students do not reach the mastery criterion then they restudy the information and retake the unit test as many times as it takes for them to demonstrate mastery of the material. Course credit is awarded when the unit has been mastered and there is no penalty imposed for not passing a unit test on a given attempt. The intent behind this is to reinforce test-taking attempts and mastering those tests while not punishing incorrect responses or failed attempts at mastery. Another element in PSI is the use of proctors. Proctors, alternatively called mentors, peer-reviewers, or tutors, are students who have previously mastered the material. Students can either be ones who have previously taken the course and are hired or given course credit for serving as proctors (called external proctors by Sherman, 1992), or they can be students enrolled in the course who have previously mastered a given unit of study (called internal proctors; Sherman, 1977). The proctors provide individualized feedback to PSI students about their unit test performance and often provide individualized tutoring in areas where the student is weak. The self-pacing feature of PSI courses allows students to move through the course material at their own pace. Thus, they can spend less time on material they understand and more time on areas they find difficult. In the initial PSI courses developed by Keller, students were not constrained by the traditional semester barriers. Rather, they could continue to work on a given course until they passed all of the unit tests. Finally, within Keller's system the instructor is seen as the facilitator of learning rather than the person who imparts knowledge. For PSI students classroom meetings are typically used to help clarify material and motivate students to be engaged learners. …

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 67 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.