Posted by: Michael Atkisson | October 15, 2010

Improving Online Learning by Getting Rid of Direct Instruction


The purpose of this post is to examine how online learning may be improved by omitting ineffective learning and instructional practices.  This is not a new line of inquiry, though it usually takes the form of examining the extent to which learning methodologies can enable the same type of learning in online environments as the can in residential ones.  Others asked this of correspondence and distance education as they came on the scene.  As  Larreamendy-Joerns and Leinhardt explain in their review of online education, “Most of the concerns about distance education have focused on the limitations inherent in different delivery technologies (e.g., correspondence, radio, television, Internet) as they seek to replicate critical features of class room instruction: social interaction, prompt feedback, engaging activities, instructional flexibility, the dynamism of a knowledgeable scholar, and adaptation to individual needs” (2006, pp. 579-580).  Though the technology that mediates learning is an important design consideration in online learning environments, as Larreamendy-Joerns and Leinhardt (2006) note, good instructional design can trump the limitations of the delivery technology.  Nevertheless, Larreamendy-Joerns and Leinhardt (2006) only focus on direct instruction and the attempts to replicate its dynamics online. With regards to other learning methodologies, Larreamendy-Joerns and Leinhardt merely recognize that there are significant challenges in facilitating other types learning online such as problem-based learning or Communities of Practice rather than delineate the specific ways in which these environments may or may not be realized online.  This paper attempts to go further by identifying a specific learning methodology, cognitive apprenticeship, and how it would facilitate effective learning in online environments.

In the US, online learning is nearly ubiquitous.  My employer, Allen Communication Learning Services, operates in a 20 billion dollar per year market for corporate elearning.  Most universities in the US have some sort of online component to the courses that they offer.  And online universities, especially for-profit institutions such as University of Phoenix, have seen enrollment grow from 365,000 to approximately 1.8 million students according to a GAO report (Ashburn, 2010).

Having worked in the corporate elearning industry for the past four years as instructional design consultant for leading companies across several industries, and having many family members, associates, and friends who have taken online courses from for-profit higher-education institutions, I have seen that many of the online learning environments are ineffective in terms of helping learners achieve robust learning outcomes.  Many would say that the second-rate elearning is due to reduced fidelity in comparison to residential learning environments (Larreamendy-Joerns & Leinhardt, 2006).  I believe, however, that if the elearning takes the form of direct instruction on line, it is not the technology medium that limits the potential of the learning, but the direct instruction.  The lecture is hardly the most authentic and effective instructional method available.  It seems that the social environment surrounding direct instruction in residential education provides workarounds that enable learning to occur in spite of the direct instruction.

When the social component of learning is not adequately considered in the design of learning environments or instruction, whether online or residential, the quality of the learning will be impacted.  Wang and Schwen (2003) call the co-mediation of technology and socialization in online learning environments the sociotechnical system of elearning, which takes into consideration elearning technology, the organizational context, the learning prerequisites, and the learning task as the fundamental unit of analysis. I will apply this model in order to describe the various elearning environments in this paper.

The Information Processing and Gangé’s Nine Events of Instruction

Information processing (IP) is the principle metaphor of cognitive science (Hunt & Ellis, 1999).  It assumes that the world is a large data set ready to be inputted into the mind through the senses.  Information is filtered, compared with long-term memory and either is discarded or enters short-term memory and or long-term memory.  Learning, in the IP metaphor, essentially is the storing and retrieving of information in long-term memory.  Many learning and instructional paradigms embody the IP assumptions, particularly in the field of instructional technology, which has largely responsible for the generation of the online learning world.  Hence, much of online learning focuses on the idea that information needs to be stored and recalled.

Gangé’s nine events of instruction (1985) is a central instructional theory that assumes the IP metaphor and underpins much of elearning.  The nine events are: gaining attention, informing learners of the objective, stimulate recall of prior learning, presenting stimulus, providing “learning guidance,” eliciting performance, providing feedback, assessing performance, and enhancing retention and transfer.  These events are seen almost as a conduit between the instructor and the individual mind of the learner and lend themselves well to the idea that the conduit could be the physical instantiation of online learning through the Internet.

In terms of the sociotechnical system of elearning, an online course that was designed with the nine events of instruction in mind would clearly account for three of the four areas that Wang and Schwen (2003) call for: elearning technology, learning prerequisites, and learning task.  Elearning technology is addressed to the extent that instructor/student interactions are facilitated so that information can be transferred to the learner’s mind.  This would include a space for delivering and receiving direct instruction and instructional materials.  Learning prerequisites are addressed in Gangé’s (1985) model because it calls for the stimulation of prior learning.  Many online courses contain a page at the beginning of the course that describes the skills and knowledge that a learner must have in order to take the course.  The learning task is also accounted for in the following steps: presenting stimulus, eliciting performance, and assessing performance.  Most elearning courses provide new information, provide an application activity and have some sort of multiple-choice assessment, whether in knowledge checks through out or in a test at the end.

The organizational context is the area of Wang and Schewn’s (2003) sociotechnical model in which it becomes questionable whether online learning based on the IP metaphor adequately facilitates intended learning.  In the nine events, organizational context is covered by the following steps: gaining attention, informing learners of the objective, providing “learning guidance,” providing feedback, and enhancing retention and transfer.  These steps embody the one-to-many communication relationship inherent in a traditional, residential direct instruction environment.  In much of online learning, these steps represent the one to one communication relationship between the instructor and the student who is being actively engaged at any given time, whether live or asynchronously.  Though it is possible to conceive of implementing Gangé’s (1985) nine events of instruction online in a way that encourages social interaction similar to residential direct instruction, it is not an explicit part of the model.  Students may be aware of chats that the instructor has had with other students, or students may chat regarding the content of the instruction as they feel so inclined, but most students online, as in residential learning will not socially engage in significant ways when the model is direct instruction.  The learning is assumed to take place in the conduit between the instructor and the learner’s mind.  This assumption is not a result of the technological affordances of online course delivery, but of the information processing metaphor upon which the learning environment is constructed.  Complaints are voiced about the lack of social fidelity in online learning environments, but it seems that this is an artifact of the instructional method, not the environment.  Residential learners subjected to direct instruction are not socially participating in learning authentic to the context in which the learning is used either.

Cognitive Apprenticeship

Other forms of learning theory, methodologies, and instruction more adequately account for the social aspect of learning than direct instruction.  These social components of learning activity can take place in either residential or in online learning environments.  Cognitive apprenticeship (CP), for example, is a type of classroom instruction that integrates social participation and the negotiation of meaning as a key aspect of learning.  CP is similar to traditional apprenticeship learning in that,

  1. apprentices model expert work,
  2. the expert scaffolds the apprentices’ learning by providing visual artifacts and models that helps them generate a conceptual framework to reference,
  3. the expert coaches the apprentices by giving them cues along the way that helps guide their participation, and
  4. the expert fades from leading the participation by granting responsibility to the apprentices to lead the learning and participation commensurate with their skills and knowledgeably (Collins, Brown, & Holum, 1991).

CP is different from traditional apprenticeship, however, in to principle ways: its purpose and its method.  The purpose of CP is to develop expertise in learners, instead of having them become journeymen.  CP differs in method by focusing on how to bring the thinking and problem solving of the expert to the surface so that they may be observed by the learner (Collins et al., 1991).  CP uses modeling, scaffolding, coaching, and fading, but it also includes articulation, reflection, and exploration, elements that are necessary for learners to be able to observe the instructor’s expertise, make sense of it and develop it for themselves.

A good example of CA is the math instructor in Collins, Brown, and Holum’s (2001) seminal article.  This math instructor took time to solve math problems in front of his class that he knew that he may or may not be able to solve.  He used a think-aloud protocol to explain to his students the domain knowledge, heuristics, and decision making he was using in order to work through the problems.  His students were able to model, not just the rote completion of solvable problems, but also the strategies that he used as a mathematician to deduce his way through a new situation.

Another example in the same article is of reciprocal teaching of reading comprehension.  The instructor was working with students who had no trouble with decoding words, just understanding what was written.  The instructor would ask herself key questions about sentences, paragraphs, and sections of the reading, she would summarize points about the reading, and she would make predictions about what she thought would come next in the reading.  The instructor would have the individuals in the reading group take turns practicing doing what she was doing (modeling); she would coach them along the way and fade as the students were more increasingly able to perform the task with the desired competence.

Cognitive Apprenticeship in Online Learning

CA is an instructional method that integrates the social aspect of learning as a key ingredient to the desired outcomes.  The question then becomes for this paper, can the social interaction required for learning in CA be facilitated in an online learning environment?  One way of finding out is by returning to Wang and Schwen’s (2003) sociotechnical model of elearning.  CA in an online environment would require certain technologies in order to facilitate the necessary social interaction.  Whether live video conferencing on Adobe Breeze, Microblogging with Twitter, or message boards and instant messaging in Blackboard, the technology is available to provide considerable collaboration and social interaction in online environments.  The learning prerequisites would include the abilities to interface with the technology mentioned, as well as the desired level of prior knowledge for the task or project.  If the learning task were, for example, to enumerate and list the key philosophical assumptions in the preamble of the US constitution, reciprocal teaching would be an excellent format to use online, teacher and students take turn asking and answering key questions, making summaries, and creating predictions.  The organizational context would be mediated by the technology, but the social participation required in CA could be done asynchronously or synchronously.  It could be done asynchronously, through blogs, micro blogs, message boards, or video posing boards.  It could be done live through online conference calls, video conferences, group chat or micro blogging.  Through the technology provided, modeling, scaffolding, coaching, fading, articulation, reflection, and exploration all may occur.  These types of learning interactions do not occur in an online environment just because the technology is there, but because the social interactions are set by the expectations laid out by the learning metaphor.


The IP metaphor provides the fundamental learning assumptions for many of the instructional paradigms through which much of online learning courses are created today.  Gangé’s (1985) nine events of instruction is a prominent cognitive science-based instructional prescription that exemplifies the instruction designed by most instructional technologists.  The complaints about online learning, I feel, are in large part a derivative of inadequate instructional theory, not the technological medium. IP metaphor-based learning methodologies are not good at explicitly calling for the integration of the social aspects of learning in instruction and the creation of learning environments, which often results in direct instruction that is isolated from how knowledge and skills are socially developed and negotiated, whether in residential or online learning.  Using cognitive apprenticeship as an example of learning theories and instructional methods that integrate social practice into the learning process, it can be demonstrated that online learning environments are fully capable of enabling rich learning, not just of domain knowledge of facts and skills, but of expertise in the use of that knowledge and skills, such as the use of heuristics and control strategies that experts in that field of practice would use.  The rich learning is not dependent as much on the technological medium as the design of the learning method.


Ashburn, E. (2010, August 4). U.S. senator lashes out at for-profit education. Reuters. Washington. Retrieved from

Collins, A., Brown, J. S., & Holum, A. (1991). COGNITIVE APPRENTICESHIP: MAKING THINKING VISIBLE. American Educator, 15(3), 6-36.

Gagné, R. M. (1985). Chapter 12: A Theory of Instruction. In The conditions of learning and theory of instruction. Holt, Rinehart and Winston.

Hunt, R. R., & Ellis, H. C. (1999). Chapter 1: Introduction to Cognitive Psychology. In Fundamentals of cognitive psychology. McGraw-Hill.

Larreamendy-Joerns, J., & Leinhardt, G. (2006). Going the Distance with Online Education. Review of Educational Research, 76(4), 567-605.

Wang, F., & Schwen, T. (2003). Organizational Factors that Influence E-learning Development and Implementation in the Corporate Context. Performance Improvement Quarterly, 16(3), 7-29.


  1. […] theories, and technologies into a new area. Ayes went on to describe the data behind the success of direct instruction, a teaching method built around scripted educational lessons that focus on call response […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s


%d bloggers like this: