INTRODUCTION

"Give me a fish, and I will eat for a day; teach me to fish, and I will eat for the rest of my life." So goes the oft-quoted philosophy of Lao Tzu, Father of Taoism and supplier of quotations for educators around the globe. The philosophy is a sound, inspiring one, valuing the teacher as changer of lives. It is a philosophy that leads many an individual, myself included, to become an instructor of adults. How many ways, though, are there to teach another person? To fish? To read? To speak in front of others? Methods are as numerous as instructors; indeed, there may be enough ways for every individual to have his/her own. But perhaps no method better exemplifies Tzu's philosophy than behavior modeling. This method of instruction embodies the component of learning by doing.

BEHAVIOR MODELING - LITERATURE REVIEW

Behavior modeling has its roots in Psychology, and behavior modeling training is based on much of the work of A. Bandura. Social learning theory, which provides the foundation for behavior modeling, asserts that most behaviors are learned by observation and modeling. It seems clear that human learning occurs in the absence of direct experience. One does not necessarily need to be electrocuted to know that it is foolhardy to stick a fork in an electric socket. Further, social learning theorists assert that if a model experiences positive reinforcement as a result of a behavior, then the observer is more likely to attend to the model's actions. However, the attention a learner pays to a model's performance is not necessarily all that is required for learning to take place. Bandura asserts that learning by observation, in fact, takes place via four processes: attention, retention, motor reproduction, and motivation (cited in Decker and Nathan, 1985). The learner must pay attention to the model as the behavior is performed. Then, the learner must process the information internally, to somehow remember the behavior the model performed. The behavior must then be performed by the learner him/herself. Finally, the learner must be inspired, or have some motivation to then use the behavior he/she has learned. Decker and Nathan suggest that the four processes as outlined by Bandura are included within the five components of behavior modeling training. Modeling is the direct presentation of the behavior by a model. Second, retention processes refer to the internal memory devices the learner uses to assimilate the observed behavior and remember the process. Third, behavioral rehearsal takes place when the learner attempts to reproduce the behavior previously modeled. Feedback, as provided by the trainer and other learners, is necessary for the learner to check for learning and to fine tune the behavioral process. Finally, transfer of training refers to the efforts of the model and trainer to make the desired behavior rehearsal resemble the actual work place or job as much as possible (Decker and Nathan, 1985).

Behavior modeling as described previously has been used successfully in organizations for years. However, as nothing is static, businesses and workplaces are changing. Training must then change to meet the new needs or today's organizations. In an era of more workers taking on more responsibilities with no increase in time, the need to train workers in more skills that can be internalized and applied in shorter time frames is critical. New training designs must take into consideration broader topics and include a variety of situations in which learners can apply skills. Also changing is the way in which learners want to learn. More interaction, variety and options are the demands of today's learner. Advances in technology, too, influence training opportunities and ways in which behavior modeling can be utilized as an effective training tool. Today's behavior modeling training reflects all of these influences. Rather than focusing on a complete skill and hoping that the learners walk away with that skill and can apply it, today's behavior modeling training begins with a basic overview of content, followed by immediate application. Many positive models demonstrate various situations in which the skill can be utilized. More content is added and applied, piece by piece, as difficulty and skill level increases. The variation in the training holds the learner's attention, involves the learner immediately, and provides more opportunity for retention. Simulation is incorporated more and more in today's behavior modeling training sessions. Self-discovery becomes an important feature of learning as it allows learners to feel the content.

Though the presentation of behavior modeling training has changed a bit, the basic processes remain the same. Still critical is the skill practice portion, though today we must develop more options for carrying out this portion of the training to meet the growing needs of today's learner. Sound preparation, live interaction and feedback by a trained observer remain integral to the success of behavior modeling training (Pescuric and Byham, 1996).

There are several factors that may influence the success of behavior modeling training. These include characteristics of the model, characteristics of the observer, characteristics of the task or behavior being modeled, and methods of presentation of the model.

In a study by Ayres, Ayres, Grudzinskas, Hopf, Kelly, and Wilcox, three components of performance visualization including modeling were analyzed for necessity in helping people enhance speech performance and reduce public speaking apprehension. The study found that modeling performance enhanced a subject's positive thought processes and helped reduce communication apprehension scores. In the study, the subjects viewed a videotape of a speaker giving a successful speech. The subject was to then make a mental movie of the speech and imagine him/herself as the speaker. The speaker in the videotaped model was Barbara Jordan, apparently a strong famous speaker. The writers asserted that the subjects might have had a difficult time imagining themselves to be as effective as was Ms. Jordan (Ayres and others, 1995). Researchers have found conflicting evidence regarding the influence of the skill level of the model on the learner (cited in Ferrari, 1996). It would seem obvious that the model should possess a high competence or expertise in the subject matter in order to provide a sound model of the desired behavior. However, high level models may intimidate the learner, as they may possess a skill that the learner cannot hope to imitate. Novice models, on the other hand, may be more similar to the learner and it may then be easier for the learner to envision him/herself actually performing the behavior being modeled. Also, novice models may be prone to actually make mistakes while modeling the behavior or task, providing an opportunity for the observer to learn not only from the modeling, but also from the model's errors and corrections. The influence of the level of the model may depend upon the type of task being modeled (Ferrari, 1996).

In addition to the level of the model, there are other characteristics of the model that have proven to affect the success of behavior modeling training. When the model is similar in age, sex and race to the learner, modeling can be facilitated. When the model controls resources desired by the learner, or appears friendly and helpful, learning may take place. Finally, and perhaps most importantly, as mentioned earlier, when the model receives a reward as a consequence of the behavior, the effects of modeling are improved (Decker and Nathan, 1985).

Characteristics of the observer also have an impact as to whether modeling, and as a result, learning, will occur. When the observer has been directed to model or be prepared to perform the behavior ahead of time, modeling is enhanced. If the observer senses a similarity in background and attitude to the model or is attracted to or "likes" the model, there is a greater likelihood that modeling will take place. Finally, if the observer receives a reward for performing the desired behavior, the process is more predisposed to success (Decker and Nathan, 1985).

Another factor affecting an observer's ability to successfully model the person to be imitated is "…the extent to which learners can observe and regulate their own actions during observational learning" (Ferrari, 1996, p. 210). Self-regulation allows the observer to fine-tune his/her behavior after modeling has occurred and take control of his/her learning, an aspect of training necessary in today's settings. Another characteristic of the observer, critical to the success of behavior modeling training influences this self-regulation: self-efficacy. Self-efficacy refers to the belief of the individual of his/her ability to effectively confront difficult situations and execute an appropriate course of action. Interestingly, while self-efficacy enables the observer to apply the behavior being modeled to a relevant situation unique to him/her, self-efficacy itself is a behavior that can be modeled (Fields, 1993).

Generally, modeling can be effective for both simple and complex behaviors or tasks depending on the way in which the behavior/task is presented. Obviously, the task or behavior should first and foremost be observable. If the behavior or task is presented in a vivid and detailed manner, an impression is immediately made on the observer. In complex tasks especially, the least difficult behaviors/steps should be presented first and built on once a foundation is established. The task/behavior should be modeled a number of times so that the learner can observe and assimilate during the repetition. Details should be minimized. Finally, modeling should be displayed positively, with a model depicting what should be done, rather than emphasizing what should not be done (Decker and Nathan, 1985).

There are many ways in which the task or behavior to be modeled may be presented to the observer. Common methods include live presentation, videotaped presentation, audiotaped presentation, written presentations and combinations of all. In an experiment by Donald Ivey (1985), this variable was manipulated to examine the effects of model presentation. In this experiment, Ivey varied the use of presenting learning points around the display of a videotaped model, as well as adding subtitles to the video. Ivey found that the variation in time of presentation of learning points had no influence, while the use of subtitles did increase learning (Ivey, 1985). In another experiment, researcher Mary Hyte (1985) examined the effects of a variety of modes of behavior modeling on learning. Specifically, she presented a videotape, an audiotape, a script and a slide/audiotape modeling mode. She found that there was very little difference between the types of modes presented (Hyte, 1985). Effectiveness of modeling mode may depend on the characteristics of the observer and how that observer learns most effectively.

A demonstration of the effectiveness of live modeling can be witnessed in a study by Tim Hacker (1996) regarding the effect of teacher conferences on peer reviews in an English Composition class. Hacker found significant learning took place among students in his class who met with him in a teacher's conference prior to providing peer responses in class (Hacker, 1996).

METHOD ANALYSIS

Behavior modeling training has been utilized for years. But is it effective? Why is it a suitable method of instruction for adult learners? It seems clear that children utilize modeling as a learning tool throughout their development. Is it a method that translates to adults? Examining the characteristics of adult learners, it becomes clear that behavior modeling is indeed an effective method of instruction. Adults are self-directed learners. Through feedback and self-regulation, the adult learner has a great deal of autonomy in the modeling process. Adults are goal-oriented. Well-planned behavior modeling training emphasizes the key points to be addressed in a modeling experience and notifies the observer of the elements to attend. Adults are practical and need a reason to learn (Cantor, 1992). One of the key characteristics of behavior modeling is the emphasis on transfer of training from the modeling session to the work setting in which it is to be performed (Decker and Nathan, 1985). Finally, adults bring many life experiences and lessons with them to any training environment (Cantor, 1992). Much of the success of behavior modeling depends on the learner's recognition of how the modeled behavior will fit into the experiences and situations the learner has already encountered. Without an existing context - the learner's knowledge - the behavior would have no meaning.

PERSONAL USE OF METHOD

As a trainer of people with disabilities and disadvantages, I have found behavior modeling to be a valuable instructional tool. The focus of my employment experience has been on assisting these individuals in finding and retaining gainful employment. Job search success for my trainees is dependent upon their packaging and responsiveness. Job retention for them, in my experience, depends on several variables: successful performance of job duties, a demonstrated ability to get along with others, and a positive, motivated attitude about working.

It is critical in the job search process that trainees look their best at all times. I model appropriate professional attire every day for my trainees. When they do not respond by dressing appropriately, they receive immediate and specific feedback regarding their choice.

In interviewing, appropriate responses to difficult questions are more easily modeled than explained. Because responding to interview questions should be natural and unrehearsed, writing or drilling specific answers is an ineffective way to train learners in this skill. Modeling appropriate interview behavior and responsiveness, however, is quite effective. In this process, I select a trainee who has demonstrated a strong proficiency in appropriate interviewing skills. This individual then models a complete interview with me as the employer for the class to observe. Those who are strong in interviewing skills observe me; those who are weak, observe the trainee interviewee. We then divide into pairs in which the "strong" trainees interview the "weak" ones. Each attempts to replicate the behavior of the model he/she observed. It is a most effective method.

In many cases, I have worked as a job coach for individuals with disabilities entering employment. In this setting, behavior modeling is an essential instructional strategy. As a job coach, I first listen, with my trainee, to verbal instructions from a manager or trainer from the work site. We then, together, observe a modeled demonstration by that manager/trainer of the job task to be performed. At that point, I perform the work behavior and listen for feedback from the manager/trainer. Upon successful demonstration of the task, I then become the model; my trainee, the observer. We repeat the steps in our new roles. The trainee has had the opportunity to be the observer twice, and has two distinct modeling demonstrations. The trainee eventually performs the behavior and receives feedback from me. Eventually the trainee is able to perform the task or behavior independently and appropriately in the work setting.

In my experience in working with people with disabilities and disadvantages, I have found that one of the most common reasons my trainees lose their jobs is because of an inability to handle conflict. Conflict resolution is a skill I teach in class utilizing behavior modeling training. I act as the model, or show a videotape of a model. Trainees observe, then rehearse, receive feedback, and continue to practice. Transfer of training to the workplace is my goal as an instructor. During this training, I am always amazed by the personal experiences and situations my trainees have encountered. We are able to build on these experiences by modeling each situation and helping the trainee learn to handle conflict appropriately.

Very often, I encounter trainees who have had negative experiences in the workplace and who have a fear of success. Motivating these individuals and helping them to acquire a positive attitude is a challenge. The most successful method I have found to instill these intangible qualities in my trainees is to model them. Every day I focus on remaining positive, on being excited and motivated in every class, on demonstrating an appropriate work attitude. There is, in my opinion, no better way to motivate another person. This technique is "catching." If I demonstrate excited and motivated behavior about a lesson or task, more often than not, my trainees respond by matching my behavior. I, in return, provide immediate feedback by reinforcing their behavior with praise and intermittent rewards. The hope is, as always, that transfer of training will take place and the attitude will carry over to the workplace setting.

PERSONAL PHILOSOPHY

I have discovered through the course of this class, that I teach from the behaviorist perspective. That is, I operate and instruct based on the principle that manipulation of environment and external rewards facilitates learning. Behavior modeling, obviously, is a training method that is dependent on manipulating environment and rewards. It is a fitting choice for my perspective. As I study the other perspectives, however, I am learning that I am a behaviorist because of my teaching method. In other words, it is not the perspective within which I believe all learning is most effective. I do, however, believe that it is an effective perspective within which to conduct training for people with disabilities. The majority of my trainees have developmental disabilities. That is, they have mild to moderate mental retardation or specific learning disabilities. Higher cognitive processes are not always possible or present. Many of my trainees learn only from concrete example and are unable to understand or retain cold, written, or regurgitated facts. Behavior modeling is an effective and appropriate training tool. But the realization that I identify more with the Developmental perspective and, even more surprising, the Social Reform perspective has been both shocking and pleasantly enlightening.

CONCLUSION

The use of behavior modeling in training is frequent and effective. Adults learn by doing, which is the defining characteristic of this method of instruction. It is a familiar and enjoyable way for learners to assimilate and apply tasks and behaviors in relevant settings. The method is easy and inexpensive to implement. Behavior modeling is flexible and appropriate for a variety of learners and learning situations. It is, in essence, the way in which we enable learners to "eat for the rest of [their lives]."

REFERENCES

�

Ayres, J., Ayres, D. M., Grudzinskas, G., Hopf, T., Kelly, E., Wilcox, K. (1995). A component analysis of performance visualization. Communication Reports, 8, 185-191.

Cantor, J. A. (1992). Delivering instruction to adult learners. Middletown, OH: Wall and Emerson, Inc.

Decker, P. & Nathan, B. (1985), Behavior modeling training. New York: Praeger.

Ferrari, M. (1996). Observing the observer: self-regulation in the observational learning of motor skills. Developmental Review, 16, 203-240.

Fields, J. C. (1993). Unlocking the paralysis of will. School Administrator, 50, 9-13.

Gay, G. (1995). Modeling and mentoring in urban teacher preparation. Education and Urban Society, 28, 103-118.

Hacker, T. (1996). The effect of teacher conferences on peer response discourse. Teaching English in the Two-Year College, 23, 112-126.

Hyte, M. L. (1985). A comparison of modeling modes in the behavior modeling process. Raleigh, NC: North Carolina State University.

Ivey, D. (1985). An attempt at refinement of the behavior modeling technique through manipulation of some attentional/retentional variable. Raleigh, NC: North Carolina State University.

Pescuric, A. & Byham, W. (1996). The new look of behavior modeling. Training and Development, 50, 25-30.