CHAPTER 2 The Learn Unit: A Natural Fracture of Teaching

TERMS AND CONSTRUCTS TO MASTER

• Operants
• Vocal learn units
• Learn unit
• Frame of programmed instruction
• Respondents
• Textual learn units
• Sequencing learn units
• Programmed Instruction
• Antecedent
• Essay learn units
• Successive approximation
• Teaching machine
• Response
• Three-term contingency
• Learn unit prerequisites
• Reader repertoire
• Consequence (postcedent)
• Writer repertoire
• Generalized reinforcers
• Scripted curriculum
• Setting events
• Number correct/incorrect
• Prosthetic reinforcers
• A natural fracture
• Setting stimuli
• Engaged academic time and
• Natural conditioned reinforcers
• opportunity to respond
• Consequence
• Individualized instruction
• Establishing operation
• Basic unit of pedagogy
• Reader, writer, writer as own reader
• Criterion referenced objectives
• repertoires
• Cost-benefits analysis
• Operant chamber
• Nucleus operant
• Intraresponse time
• Instructional history
• Programming
• Phylogenetic contributions
• Role of the target Sd
• Interlocking operants
• Correction operations
• Conversational unit
• Instructional control
• The student’s three-term contingency
• Speaker as own listener
• Insertion of learn units
• Guided notes for presentations

A MEASURE FOR TEACHING

Education requires a measure that contacts the natural fractures of instruction—one that includes both the behaviors of the teacher or teaching device and the behaviors of the student. The measure must provide immediate feedback to the teacher, be a valid prediction of long-term outcomes, and be an integral part of instruction. The identification and acceptance of such a measure will serve to establish strong practices and effective practitioners. We believe that the research literature has converged on just such a natural fracture of pedagogy.

A natural fracture is a unit of a compound that separates naturally from other components as a result of lawful conditions. It is an absolute unit (Johnston & Pennypacker, 1992). For example, in geology the break in geological strata on the face of a cliff is one instance of a natural fracture. DNA material, zero centigrade, and atoms are others. The identification of, and the quest for, natural fractures are more common in the natural sciences than in the social sciences. One of the characteristics of behavior analysis that aligns it with the natural sciences is a preference for the identification of natural fractures. Once identified, natural fractures of behavior (e.g., the operant) become direct measures of absolute units of behavior as opposed to artifacts or inferences about behavior (Johnston & Pennypacker, 1992; Sidman, 1960). Measures of constructs such as IQ, “locus of control,” personality, or engaged time are examples of measures that are not natural fractures. The operant and the respondent, on the other hand, are natural fractures of behavior.

The identification of measures that contact the natural dimensions or fractures of the world, including behavior, typically functions to produce rapid progress in the sciences (Mach, 1960; Skinner, 1938, 1953). B. F. Skinner (1938) identified the operant early on in the history of the science of the behavior of the individual. Instead of a natural fracture, Skinner might have measured the by-products of the operant such as the number of time intervals an organism was “engaged” in an operant chamber or the time that an organism was engaged in reaching the end of the maze (actually a prevalent measure at the time). It is difficult to imagine how our current science of behavior could have evolved based on “on task” measures of an organism. It was the identification of the operant as a natural fracture rather than an artifact of behavior/environment interaction that led to rapid progress.

At present, much of education relies on measures of artifacts of behavior or inferences about behavior such as ratings of students or teachers’ “perceptions,” engaged academic time, and the presence or absence of services (e.g., contact hours of teaching, the number of periods taught) (Brophy & Good, 1986; Hamburg, 1992; Stallings, 1985). Other common measures are scores on standardized tests, but these are not absolute units. They do not relate to the specific instruction received by the student, nor do they show the moment-to-moment behavior changes of the student. However, we shall show how some of this research on artifacts of behavior together with research in behavior analysis support convergence on a natural fracture of pedagogy—the learn unit.

THE BASIC UNIT OF PEDAGOGY

The learn unit consists of the least divisible component of instruction that incorporates both student and teacher interaction and it predicts new stimulus control for the student (Greer, 1994a). It is present when student learning occurs in teaching interactions, and when it is absent student learning does not occur. It is a countable unit of teacher and student interaction that has the potential to change the behavior of each party individually or jointly. That is, the goal for both parties is for the student to come under the control of the target Sd (stimulus discriminative) and the consequence of behavior. The three-term contingency or operant (antecedent, behavior, and consequence) is what the student is to learn. The teacher organizes the student’s environment such that the student’s operant emerges. The teacher also has a series of operant behaviors that must interlock with the student’s three-term contingency in order for the operant to emerge. The learn unit also needs to occur within the motivational conditions in which the student’s operant will be needed (i.e., the operant is to become part of the student’s repertoire).1 In order for this to happen, the teacher must respond in certain ways to the presentation of the student Sd and to the resulting behavior or its absence from the student. In effect, the teacher “learns” from the response of the student—that is, the teacher learns what to do next from the student’s performance.

A basic scenario for the interlocking three-term contingencies of the teacher and student are as follows. The Sd for the teacher to present an Sd to the student is the student’s attentive behavior. Presentation of an Sd by the teacher to the student is the teacher’s response and the Sd for the student’s response. The behavior or behavioral product of the student to the teacher presentation of the student’s Sd serves as a consequence for the teacher’s behavior and an Sd for the teacher’s next response to the student (e.g., reinforcement of the student behavior or correction of an incorrect response), while the teacher’s response is the consequence for the student.2 The completion of this single learn unit functions as a consequence for the teacher’s last response. The learn unit is a measure of the symbiotic relationship between the behavior of a teacher and a student.

There are many possible interactions involved in a single learn unit. For example, if the student is not attentive, another set of three-term contingencies is emitted by the teacher to bring the student’s behavior under instructional control; thus the final presentation to the student would involve more teacher interactions than that described in the basic unit. Several examples are presented in Table 1. They all presume that the student has previously mastered the prior skills necessary to begin learning the new operants.

THE RESEARCH BASE

Learn units must have all student components [establishing operations, discriminative stimuli (Sds), response opportunities, and consequences that reinforce or correct). The teacher must emit the necessary behaviors that set the occasion for the contingency for the student. The teacher’s behaviors are themselves two or more three-term contingencies. According to the existing research a learn unit must include the following.