Search

Computer terminology and modeling are employed in an attempt to precisely describe human algorithmic behavior. Following the lead of many contemporary cognitive psychologists (e.g. Anderson & Bower; Quillian; Simon), the components and the structural relations teaching and learning algorithms are cast into a systematic model. Within the model, the concepts of an algorithm's depth and width, and the components' processing order as parallel or serial are defined and examples are presented. The model not only maps the logical order of events, but also allows the efficiency of various configurations to be calculated. Next the model's control activities are described, with special attention given to the level of synchronization of processing units or branches as well as the corresponding gate-keeping functions. Finally, the results of two experiments are compared with the model to ascertain its heuristic value.

The literature on performance objectives is reviewed under the following headings: definitions of performance objectives; function of performance objectives; pros and cons of performance objectives; and form of performance objectives. The review indicates a general agreement, unsupported by research, that performance objectives have value for learners, instructors, and designers; that there is no operational definition of the term performance objective; and that, although there is general agreement regarding the three formal characteristics of a performance objective, examples of performance objectives still differ greatly from one another in form. The results of six original studies are reported.

An algorithmic procedure was examined under several instructional approaches to yield design principles pertinent to military training. Learning factors were selected from information-processing theories to test their applicability with instruction directed by learning algorithms. A version of a logical, or familiar, and a computational algorithm was generated, further divided into verbal and symbolic (mathematical) forms, and presented to sixty undergraduates. Multivariate analyses were employed on two dependent measures, instructional effectiveness and efficiency. Posttests were administered both immediately following instruction and one week later. Attention was also given to the criterion-referenced effectiveness of the various algorithmic forms. Learners achieved high scores following very brief instruction, with verbal and logical groups outperforming their comparison treatments. Retention losses after the time delay were found to closely match the typical needs of a training environment. Tentative principles for using algorithms in instructional design include emphasis on concrete, familiar content, short, compact structures, and complete mastery of each preceding operator or discriminator. The efficiency of the algorithm approach is highlighted as having unusual utility in meeting the future demands of schooling and training. (Author)

This study investigated the effects of (a) student self-evaluation and (b) instructor feedback on the self-evaluations on the ability of students to evaluate their own materials and on their performance. Fifty-six education students were rated with respect to materials which they produced in an instructional media course. Students in the first experimental group used a check list to evaluate their own products prior to submitting them for grading by the instructor. Students in the second group did the same, but received comments from the instructor regarding their self-evaluation; students in the control group did not engage in any form of self-evaluation. The relationship between student score predictions and instructor scores on the final class product was reasonably high for the control group. However, the combination of self-evaluation and instructor feedback on self-evaluation raised this figure to .81. Criteria specific feedback on self-evaluation seems to be an effective way to develop student skills in estimating the quality of their performance. It was anticipated that self-evaluation itself without any instructor feedback on the accuracy of the evaluation would raise the correlation between student and instructor scores. Analysis revealed that self-evaluation itself had no beneficial effects, and in fact, seemed to have a negative effect. It is our contention that self-evaluation (i.e., the effective use of objectives) is contingent upon an ongoing process of critical comparison.

The opening section of this paper discusses these issues and attempts to define what algorithmic prescriptions have to offer for the instructor. Next, the question of effectiveness and efficiency is examined in two experiments. The first study provides preliminary data regarding the form of algorithm and the presentation modes to be used. The second study asks 77 undergraduate volunteers to master and utilize one of three forms of an algorithm: a prose discourse, a flowchart procedure, or a mastery-inducing flowchart technique. Also examined in a factorial design was the effect of the presence or absence of the algorithm in subsequent problem solving and the learners' ability to retain the algorithmic content over a period of time. The results provided very strong data for the overall effectiveness of algorithms, showing from 52% to over 80% accuracy following an average of only 15 minutes of instruction and practice. Surprisingly, almost no decrement (only 1%) in performance was observed across groups after a one-week delay. Although scores were high after the algorithm representation was withheld, subjects did better when allowed to continue using it. In general, the prose and flowchart representations were equally effective, with the mastery-flowchart group performing significantly worse. The flowchart group alone was much more efficient than the other groups.

The three studies, described in this report, concerned the assessment of procedures which influence instrument sensitivity in military flight instruction. The ability to determine the relative location of an in-flight vehicle from a set of dial indicators was influenced more by students' expectations than by context. Sensitivity to heading and attitude information was increased when learners were supplied with overt attention control. The teaching of instrument sensitivity appears best facilitated by employing publically observable responses during training, and seems to be a function of the degree to which the learner can be directed to use a spacial processing mechanism during initial encoding. (Author)

The primary goal of this study was the identification of variables which influence transfer from cognitive pretraining to perceptual-motor skill acquisition. The results clearly support the central hypothesis that the direction of transfer is dependent on the type of Verbral Prescriptive Rules (VPRs) or instructional cues which were learned during cognitive pretraining. Systematically developed rules led to more precise perceptual motor behavior than currently operational rules, which appear to inhibit rather than facilitate performance. The second objective of the study was the discovery and validation of prescriptive principles for the design of perceptual-motor instruction.