Motivation
Before CAI can be discussed, one must first familiarize himself/herself with the definition of motivation as well as several theories on motivation. McClelland (1987) wrote that motivation is a broad and loosely defined field, and "covers everything from detailed investigations of the physiological mechanisms involved in animal drives to elaborate analyses of the unconscious motives behind abnormal or symptomatic acts in a person to factor analyses of the motives of people assign to themselves to explain their behavior" (p.1). Ford (1992) defined the concept of motivation as "the organized patterning of an individual's personal goals, emotions, and personal agency beliefs (p.78).
Ideally, a student should be intrinsically motivated, meaning that the task itself is motivating and the student feels satisfaction from participating in the activity. However, when students are not intrinsically motivated (indicating low levels of motivation toward the particular task), external reinforcers may help increase the student's interest and motivation; e.g., grades, money, etc. In the case of CAI, extrinsic motivators (external reinforcers) can be built into the instruction to reinforce learners when they answer correctly, demonstrate effort through persistence, etc. Reinforcement following a particular desired behavior improves the likelihood that such behavior will reoccur. A reinforcer is a type of extrinsic motivator that increases the frequency of the event it ensues.
Computer-Aided Instruction (CAI)
CAI's roots go back to the 1950's when the first computer programs were developed. CAI was primarily used as a means of delivering instruction in place of the regular teacher, or as a drill and practice-type of supplement to regular instruction. CAI has since developed into much more than it was in those early years, and has been scrutinized by educators and researchers as to its practicality and viability as an educational tool. Research on CAI has repeatedly shown that by using CAI in addition to regular classroom instruction, students show significantly higher gains in academic achievement across all content areas ( Kenzie, Sullivan, & Berdel, 1992).
According to Rasmussen and Davidson (1996), one of the most powerful features of CAI is its capacity to individualize instruction to meet the specific needs of the learner. Self-paced instruction, the ability to present content in a variety of ways (i.e.: text, audio, video, and graphics), and features such as hypertext, make CAI an effective learning medium. The use of CAI in classrooms has increased greatly over the years. As schools face continually growing problems with class sizes and heavier workloads, educators are looking at CAI as a means of enhancing instruction. However, as with any other form of instruction, CAI is not the end all instructional tool. While many students are able to benefit from its attributes, it is important for educators to remember that some students require other strategies to meet their educational needs.

CAI & Motivation
Motivation has a dual role in CAI. First, the students' levels of motivation prior to using CAI, greatly influences the overall success of their experience. Second, CAI can be a motivator to encourage students who have low levels of intrinsic motivation for learning a particular subject. This paper will discuss both roles of motivation in CAI.

Models of Motivation
One well-known motivation theory is Keller's ARCS model (1987). Keller's model suggests strategies for stimulating the motivation to learn. ARCS is an acronym for the four points in his model: Attention, Relevance, Confidence and Satisfaction. Attention involves the arousal of interest in learners, the stimulation of an attitude of inquiry and the maintenance of attention. Relevance refers to tying instruction to make it relevant to the students' personal interests or goals. Confidence refers to the students' expectations for success, and Satisfaction refers to the process or results of the learning experience. Keller (1999) has suggested applying his ARCS model to CAI as well as to traditional learning environments.
Another important factor in motivation and CAI is self-determination. Deci and Ryan (1985) address the importance of self-determination (or choice) in intrinsic motivation. More recent research suggests that choice is an important factor in CAI as well (Kinzie, Sullivan, & Berdel, 1992; Yang & Chin, 1996).
 
 

Self-Efficacy
With regards to motivation and achievement, how a student feels about his/her ability to learn is directly related to what is referred to as self-efficacy. According to Bandura (1997), self-efficacy is an individual's perception and belief of his or her capability to organize and execute the courses of action required to produce given attainments. People's beliefs in the efficacy have a wide-variety of effects on what courses of action they choose to take, how much effort the choose to exert and how long they will persevere when confronted with obstacles.
Computer technology is changing the basic structure of education by providing an instant interface for self-directed learning. To individuals who have a low sense of self-efficacy, technology is intimidating. However, to those students who are efficacious in their abilities, the use of technology appears simple. Often it simply comes down to the previous amount of experience that an individual has with computers and whether or not those experiences were positive or negative. For example, one's experience with technology, plays a significant role in the efficacy one has towards using it in subsequent applications.
Teacher efficacy has a direct impact on the efficacy of their students. Teacher attitudes and beliefs of their own abilities to use computers and technology for both personal and instructional purposes, are instrumental in how students perceive their own abilities. How confident a teacher is in his or her own abilities will determine whether an to what degree technology will be integrated into the curriculum. A teacher who is more confident in his or her abilities will: a) be more likely to use technology in the classroom, and b) will convey a greater sense of efficacy and 'can do' to their students (Abu-Jaber & Qutami, 1998). Teacher efficacy is also greatly influenced by the amount and availability of support received from co-workers and administrators.
 
 

Preferences for Learning
Many learners prefer processing information primarily through sight, and they can become frustrated with teachers who mainly use the traditional auditory (lecture) type of instruction. Likewise, others have strong preferences for more auditory or tactile (hands-on) types of instruction. When used properly, computer-assisted instruction (CAI) can enhance the learning process for learners of all types, regardless of learning preferences. 
CAI helps to increase motivation through the use of a wide variety of software programs, which can stimulate learners' natural curiosity with the use of video (graphics), audio, and interactive applications. By allowing learners to work at a more individualized level and pace as is inherent in CAI, learners experience less frustration at being 'held back' when ready to move ahead or, as in the case of a slower learner, frantically trying to keep up. CAI also can release teachers from the burden of instructional delivery and grading, thereby allowing them more one-on-one time with individuals who need their assistance.
In a group setting, CAI has proven to assist in the levels of interaction between students and their teachers. An example of this is in a study done by Beauvois (1995) in which a group of French foreign language students participated in a lab-networked 'E-Talk' forum (or class chat room). The students, while not required to speak only French within the forum, were increasingly motivated to use only French and to interact with each other through the use of this forum. It was found that student motivation and self-efficacy were significantly increased through this online dialogue with their peers and that much of the stress and anxiety previously experienced by the students were greatly reduced through participating in this dialogue/forum.
Conversely, in a different study done by Hayward (1994), it was determined that students were less likely to interact with each other during in class discussions in an environment where computers are present but not used as a part of the class. It was also determined that many students were intimidated by the presence of computers and thus were less likely to interact as a result of this discomfort.

Achievement Level
Hativa (1989) conducted a study to determine differences in attitudes according to students' aptitude, gender, grade and S.E.S. level, found that the attitudes of very high achieving students differed from the general population in that they liked working in a CAI environment due primarily to the diversity of activities that were made available to them through the CAI program. In contrast, the attitudes of very low -achieving students varied from the general trend in that they enjoyed CAI work much less. No significant differences were determined between attitudes of high versus low-achieving students on the positive feedback provided by CAI.

Socio-Economic Status
According to studies done by Hativa (1989), and one by Attewell and Battle (1997), students from different SES backgrounds varied on their preference from CAI work. It may be surprising to some that those students from lower SES families appeared to enjoy CAI significantly more than other students. This phenomenon was attributed to the positive feedback they received and the novelty of typing versus hand-writing a paper. Students from higher SES families and who, very likely, have access to computers at home, are not as motivated by the use of computers so the 'novelty effect' that inspired the other students has no effect on them as it has long since worn off.

The Role of Gender
Research in gender bias on motivation, as witnessed through attitudes towards CAI, has produced mixed results. On one hand, many studies suggest that female students display more anxiety, apprehension, and avoidance of computers than male students. Other studies imply that females are not avoiding computers, but simply have less access to computers, both in school and at home (Kimbrough, 1999). Many other studies suggest that there are no gender differences in attitudes or access to computers, or that gender differences in attitudes towards computers are more likely due to differing levels of experience (Nelson & Watson, 1991). Teh and Fraser (1995) reported that past studies indicated that boys generally felt more comfortable using computers than girls, and that boys monopolized computers in CAI settings. Katz, Maitland, Hannah, Burgaf and King (1998) confirmed Teh et al's (1995) finding, but added that there was a tendency for women to view the computer as more useful than men. Other research by Kimbrough (1999) revealed that higher percentages of female students participated in on-line tutorials and with greater frequency than did male students. This paper focuses on four main influences on the role of gender within CAI: math, social interactions (families and schools), gender biases in software, and the influence of age on gender differences.
Research conducted in the 1970's and 80's has linked gender differences in motivation and CAI to math anxiety because of the types of courses taught with computers, as well as the location of computers within schools. In the 1970's, computer programming and math were the only computer courses available, and the computers were placed within the math and engineering departments. Both departments have traditionally been male dominated (Nelson & Watson, 1991; Teh & Fraser, 1995). 
Nelson and Watson (1991) found that social interactions among children, their families, and schools can affect attitudes and motivation of children towards technology in different ways, according to gender. Nelson and Watson (1991) noted that children's attitudes are shaped by interactions with their parents and others. Beginning in preschool, boys have more exposure and encouragement by their family members (primarily fathers) than girls to interact with computers. Furthermore, more computer games are purchased for male children, encouraging a positive attitude and increased motivation for boys to become involved with computers. This pattern continues and the gender gap increases with age, as boys are more frequently sent to computer summer camps, where they gain more experience with computers. Other research (Nelson & Watson, 1991; Katz et al, 1999; Teh & Fraser, 1995) indicated that females are socialized to view computers in a "soft mastery" view, and become more involved with artistic communication skills through the computer, whereas boys are socialized to conquer the computer and to view it as an objective mechanism. These differences in socialization have influenced students to pursue different activities, such as males entering into mathematical and scientific aspects of computing and females into artistic ones. Archer's study, as cited by Nelson & Watson (1991), revealed that parental encouragement can overcome negative school experiences and their expectations can be the deciding factor, which can positively motivate girls to engage in computer activities.
Although the family plays a major role in the developing attitudes and future motivation of children towards computer activities, teachers also influence students' attitudes and motivation in this area. By 1988, the NEA found that only 15% of American teachers were using CAI in their curriculum (Abu-Jaber and Quatami, 1998). This resistance to incorporate computers into the classroom can be explained by a lack of in-service training made available to teachers. Nelson and Watson (1991) suggested that teacher's negative attitudes towards using computers in instruction, is passed onto their females students.
Another contributing factor to the differences in motivation and attitudes towards CAI is the gender bias in computer software. Historically, males have dominated characters and story lines in textbooks and fairy tales, and traditionally, male characters assumed the more active, adventurous, and heroic roles. This pattern has continued in computer games and educational software (Nelson & Watson, 1991). Furthermore, most educational programs are focused on topics that have traditionally been male oriented, such as science, adventure, war, and technology. Nelson and Watson (1991) cited Biraimah's study, which found that sixty-three percent of the characters examined in her evaluation of software programs were male. Furthermore, males represented both a greater variety of professions as well as more active roles than did the female characters. These gender biases in educational software have resulted in decreased motivation for females to participate in CAI.
Gender inequities generally become more pronounced with age. In kindergarten age, gender gaps are not as clearly defined, and both sexes are eager to participate in CAI (Bergin, Ford & Hess, 1993). In addition, research indicates that at this age, teachers treat both genders equally with regard to emphasis and encouragement to participate in computer-aided instruction. By middle school, these differences in computer interest are more clear, as evidenced through disparities in computer self efficacy, actual computer use, and willingness to consider computer related careers (Bergin et al., 1993).
Further gender differences within CAI have been observed. Hativa (1989) noted that boys like CAI work somewhat more than girls, and that boys also like receiving more difficult exercises from the CAI work than girls. In addition, boys are less bothered by the time limits for solutions. In the same study, Hativa (1989) found that there was a significant difference in preference for competition, indicating that girls were much less likely to choose CAI, which involved competition.
 
 

General Guidelines and Culture
Design interface can have a large influence on how motivating a software program is. Aside from general guidelines and the incorporation of motivational models such as Keller's (1987) ARCS or Deci and Ryan's (1985) suggestions for self-determination, there are cultural considerations. Some general guidelines for motivation in interactive multimedia instruction include suggestions for typography, graphical images, color, animation/audio, integration, and motivation (Heum Lee & Boling, 1999). Some of the guidelines for typography include consistency in addressing textual cues and signals to the learners, using upper and lower case letters, high contrast between letters and background, etc. Other suggestions include using simple, clear images and graphics for instructional or attention focusing effects rather than simply for the sake of having a graphic, being conservative with color, keeping color consistent, using animations sparingly, and using animation that is consistent with the learning objective. Bradshaw (2000) stated, "Good design is not context neutral-what works well for one content area and audience does not work well for all other content areas and audiences" (p.1). For example, while it may be considered appropriate to use animals with human like qualities in animation or story lines in the United States, it is not appropriate in other countries, such as Mexico. Few CAI programs are designed to accommodate cultural views. It was suggested by Williams-Green, Holmes, and Sherman (1998) that to do so would enhance the meaningfulness of the learning environment. For example, in Western countries, the orientation is toward higher levels of individuality; weak uncertainty avoidance; small power difference; and femininity. The goal should be to identify cultural values that will impact instructional design decisions and apply the cultural values into the instruction (Williams-Green et al, 1998).

Self Determination
Deci and Ryan (1985) defined self-determination as the experience of choice, or in other words, a perceived internal locus of causality. When self determined, one acts out of choice rather than obligation or coercion. Similarly, there is an issue of learner and program control within the design of software used in CAI. In general, there have been mixed findings regarding learner and program control in CAI. Kinzie et al. (1992) found that learner control shows higher levels of continuing motivation among males and females. Furthermore, Kinzie et al (1992) found that when learners missed a question, they preferred to have personal control over whether or not to review relevant material before going on to another review question. Program control resulted in higher performance in males, but learner control shows non-significant advantages in performance (Kinzie et al., 1991). Contrary to Kinzie et al.'s (1992) findings, Yang and Chin (1996) found that groups under program control performed better on post test, but there was not a significant difference in motivation between the two groups. Hativa (1989) found that in CAI, students primarily dislike: a) time limits; b) inability to edit work; c) detection of mistakes by computers, d) lack of challenge; and e) competition with class mates that the CAI work enforces. Similarly, Deci and Ryan (1985) found that in regular classroom environments, deadlines significantly diminished subjects' intrinsic motivation.

Adaptivity
Astleitner and Keller (1995) attempted to show how theories and empirical findings of research on motivation could be integrated in a formal model in order to both describe and predict motivation within the framework of motivationally adaptive CAI. The authors mentioned how most of the design considerations for motivating learners within CAI have been based upon changing levels of difficulty only according to one task performance or to only two dimensions (easy to difficult), and individual learning times were not considered. In addition, animation has been used as a motivator, but it has been used so much that it can become a de-motivator, repeating the same kind of feedback over and over again. Other problems have been that learning objectives are not clearly stated, which leads to a sense of helplessness among the learners. Shortcomings within CAI were traced back to problems in implementing motivational strategies and failures in distinguishing different levels of adaptivity (Astleitner & Keller, 1995). The authors continued by suggesting that in the past CAI included gimmicks as motivators in the past, rather than specific motivational strategies. Furthermore, when motivational strategies and tactics were considered, they were not theoretically sound, and if they were, they were not implemented well because of both hardware and software limitations. Another problem identified by Astleitner & Keller (1995) was that motivational strategies within CAI have not been adaptively implemented. In other words, they do not adapt to the learners' needs at any particular point over an extended period of time. Astleitner and Keller (1995) wrote:
"To develop a model of the learners' motivational states via computer, a general theoretical framework must first be used to enable the calculation of dynamic and interactional motivational components in a general manner. Second, this global framework has to be specified with differentiated motivational theories and empirical results. Finally, the complete model must be implemented in a computer simulation for predicting motivational states of a learner."

References

Abu-Jaber, M., Qutami, N. (1998). Students' self-efficacy of computer through the use of cognitive thinking style. International journal or instructional media, Summer 1998, 25(3), 273-284.

Astleitner, H., & Keller, J.M. (1995) A model for motivationally adaptive computer-assisted instruction. Journal of Research on Computing in Education, 27, 3, 270-280.

Attewell, P. & Battle, J. (1997). Home computers and school performance. NY: The information society, 15, 1-10.

Bandura, A. (1997). Self-efficacy, The exercise of control. NY: W.H. Freeman and Company.

Beauvois, M.H. (1994). E-Talk: Attitudes and motivation in computer-assisted classroom discussion. Computers and the Humanities 28: 177-190.

Bradshaw, A.C. (2000). Evaluating visuals for instruction. In Griffin, R.E., Gibbs, W. and V. Williams (Eds) Natural Vistas: Visual Literacy and the World Around Us. Loretto, PA: International Visual Literacy Association.

Csikszentmihalyi, M.(1990). Flow: The Psychology of Optimal Experience. New York, NY: Harper Perennial.

Dalton, D.W., Hannafin, M.J., & Hooper, S. (1989). Effects of individual and cooperative computer-assisted instruction on student performance and attitude. Educational technology research and development, 37(2), 15-24.

Deci, E.L. and Ryan, R.M. Intrinsic Motivation and Self Determination in Human Behavior. New York: Plenum Press, 1985.

Ford, M.E (1992). Motivating Humans: Goals, Emotions, and Personal Agency Beliefs. Newbury Park, CA: Sage Publications, Inc.

Hativa, N. (1989). Student conceptions of, and attitudes toward specific features of a CAI system. Journal of computer-based instruction, 16 (3), 81-89.

Hayward, P.A. (1994). When novelty isn't enough: A case study of students' reactions to technology in the classroom. College student journal, 28 (3), 320-325.

Katz, L., Maitland, M., Hannah, R., Burggraf, K., & King, S. (1999). The effects of gender and academic program on learning styles and attitudes of undergraduate students using multimedia, web-based anatomy labs. Selected papers from the 10th International conference on College Teaching and Learning, 81-87.

Kinzie, M.B., Sullivan, H.J., & Berdel, R.L. (1992). Motivational and achievement effects of learner control over content review within CAI. Journal of Educational Computing Research, 8(1), 101-114.

Lee, S.H. & Boling, E. (1999). Screen design guidelines for motivation in interactive multimedia instruction: A survey and framework for designers. Educational Technology, May-June, 19-26.

McClelland, D.C.. Human Motivation. Cambridge, NY: Cambridge University Press, 1987.

Nelson, C.S., & Watson, J.A. (1990-91). The computer gender gap: Children's attitudes, performance and socialization. Journal of Educational Technology Systems, 19 (4), 345-353.

Ross, J. & Shulz, R. (1999). Can computer-aided instruction accommodate all learners equally? British journal of educational technology, 30 (1), 5-24.

Swenson, R. P., Anderson, C. (1982). The role of motivation in computer-assisted instruction. Creative computing, 134-139, October 1982.

Teh, G.P.L. & Fraser, B.J. (1995). Gender differences in achievement and attitudes among students using computer-assisted instruction. International Journal of Instructional Media, 22(2), 111-120.

Wang, Y. (2000). Training teachers using computers. THE Journal, May 2000, 27 (10) 66-72

Williams-Green, J., Holmes, G., & Sherman, T.M. (1997-98). Culture as a decision variable for designing computer software. Journal of Educational Technology Systems, 26 (1) 3-18.

Yang, Yong-Chil & Chin, Wee-Kyo (1996-97). Motivational analysis on the effects of type of instructional control on learning from computer-based instruction. Journal of Educational Technology Systems, 25 (1), 25-35.