Category Archives: cognitive apprenticeship

Knowledge, Mind, and Assessment

What is Knowledge?
Knowledge is socially negotiated, situated, and distributed.  It is what Sir Karl Popper would term a “World 3” artifact, it’s own entity, taking on a form of its own (what Scardamalia and Bereiter would call a “conceptual artifact”) within a knowledge community.

Knowledge is situated in the sense that concepts are situated within the context of the surrounding environment, and these concepts are progressively developed.  It is a product of the activity, context and culture in which it is learned and used.  Conceptual knowledge as a tool is only fully understood through its use.  Furthermore, its use changes the user’s world view, causing the user to adopt the belief system of the culture in which the conceptual knowledge tool is used (Brown, Collins, and Duguid; 1989).  Brown, et al. (1989) propose a “cognitive apprenticeship” model of education, which promotes learning through activity, tool, and culture.  Cognitive apprenticeship is supported by collaborative social interaction and social construction of knowledge.  It honours the situated nature of knowledge, embeds learning in activity, and makes deliberate use of the social and physical context.

Bereiter (1997) argues for the value of overcoming situatedness of cognitionby creating World 3 knowledge objects that are transferrable across different situations.  He points out that the problem with situatedness of cognition is that as one learns more about a phenomenon, one knows increasingly more about the phenomenon as framed within that situation, making this knowledge less likely to be generalizable to other situational contexts.  Furthermore, treating all knowledge as situated would make knowledge objects invisible.  Bereiter (1997) calls us to treat ideas as “objects of inquiry” (p. 18), requiring “Disciplined movement back and forth between (Popper’s) World 1 (the physical world) and World 3 (immaterial world of knowledge and abstract objects)” (p. 18), yielding the “hypothetico-deductive method” and endless possibility for theory development, problem solving, and design.  Incidentally, Sir Karl Popper’s philosophy of knowledge is based on his proposition of 3 worlds:

  1. World 1 – the physical world
  2. World 2 – the world of the mind (mental states, ideas, perceptions)
  3. World 3 – the immaterial world that is the product of the human mind (e.g. the concepts and ideas represented or expressed by books, papers, paintings, symphonies; but not physical books, papers, or paintings themselves)
What is Mind?  Where Does Knowledge Reside?
Sebastian Seung of MIT discusses his theory of mind:  “I Am My Connectome” at a TED Talk (July, 2010):

Nueroscientists since the 19th century have hypothesized that one’s memories, personality, and/or intellect also reside in the connections between one’s neurons.  This would imply that knowledge resides in our neural connections.  These connections form what Seung calls one’s “connectome“.  According to Seung, neural activity encodes our thoughts, feelings, and mental perceptions; but neural activity can also cause our own neural connections to change (hence changing our connectome).  Conversely, our experiences can change our connectome.  Thinking can change one’s connectome.  Neural activity constantly changes in our brain, but it is the brain’s neural connections (which make up the brain’s neural network) that determine the pathways along which this neural activity flows.  These neural pathways as a whole is the connectome.  The Seung Lab at MIT seeks to read memories from connectomes, and Seung is leading the new field of connectomics to test these hypotheses.

As I consider both Bereiter and Seung’s theories of mind, I see the one commonality is that they both see knowledge as residing in connections and composed of networks of connections.  In Seung’s case, these connections are of a physiological nature residing in the human brain.  In Bereiter’s case, these connections are conceptual in nature, residing in World 2 and World 3; with collaborative knowledge building as the vehicle by which individual learner’s conceptual connections combine and build upon those of other learners in the creation of a World 3 knowledge object.

The Nature of Knowledge:  Implications on Assessment
Bereiter (1997) points out that situativity theorists have not advanced any educational ideas due in part to their failure to define the purpose of education and to their confusion between process and product.  The collaborative knowledge building that I have witnessed in my experience as a graduate student, classroom teacher, and instructional leader aligns with Bereiter’s assertion that “knowledge implicit in the process” (p. 23) should be distinguished from “knowledge that is the product of the process” (p. 23); and yet, this distinction is not recognized in K-12 classroom practice, curriculum policy, or assessment and evaluation policy.  Our education system is still misguidedly preoccupied with learning goals and the embodiment of knowledge in the form of reports and presentations, failing to see knowledge taking a form that can be worked with or even packaged and sold (Bereiter, 1997).  This is decidedly antiquated thinking, considering our knowledge society’s wealth will come from knowledge work – no longer from manufacturing as in the industrial society of old.  “What distinguishes knowledge work is not using knowledge by creating or adding value to it” (Bereiter, 1997: p. 23).  This is not to say that we should do away with learning goals, merely that the focus should shift from these to learning for deeper understanding and the construction of knowledge through collaborative means supported by technology.

Hutchins and his colleagues developed the distributed cognition approach in the 1980s as a way of understanding collaborative work practices, or “cognitive systems” – complex interactions between multiple people, artifacts to perform an activity, and technological systems (Rogers, 2006).  There are two levels of analysis:  (1) analysis of ‘the propagation of representational state across media’ and (2) analysis of human interactions (i.e. the problems, the breakdowns, distributed problem-solving, role of non/verbal behaviour, coordinating mechanisms, communication during collaboration, knowledge sharing/accessibility).

While I can see the value of having a distributed cognition framework with which to study the dynamics of collaborative work, I find that this approach emphasizes the interactions enacted to complete the task, and does little to examine closely how ideas/memes/theories evolve within and among a community.  Furthermore, the distributed cognition approach does not seem recognize knowledge as its own entity, as a ‘product’.  Rather, it acknowledges knowledge as a process but focuses on the knowledge community’s development of that process, not on the conceptual evolutionary process of knowledge in its own right.

This is especially problematic when one considers Carl Bereiter’s (2002) contentions that that the mind is not a container, knowledge and conceptual artifacts do not reside in learners’ minds,  and that the mind does not reside in the head of a person, nor does it necessarily reside within one person.  For Bereiter, knowledge objects and conceptual artifacts belong to World 3.

If distributed cognition analysis of a knowledge community can be used in such a way as to reveal how the community’s mind/cognition is distributed among the community, and hence, trace how the conceptual development of knowledge as a conceptual artifact is gradually constructed within the community; then distributed cognition could be a powerful approach to prove Bereiter’s theory of mind, and aid in the classroom assessment of collaborative knowledge building in asynchronous online environments.

In this regard, Activity Theory (Hew & Cheung, 2003), developed by Alexei N. Leont’ev and Sergei Rubinshtien, seems more promising.  Based on the premise that tools mediate between subject and object processes, rules mediate between subject and community processes, and roles mediate between community and object processes; one can use this framework to analyze several combinations of any three processes as illustrated by any three edges of the following triangles:

If the basis of our knowledge society’s economy is going to be knowledge work, then it stands to reason that schools must teach learners to be proficient knowledge workers.  Assessment is vital to this learning process, for teachers and learners alike.  Applying Gunawardena et al’s (1997) 5 phases of active knowledge construction and Henri’s (1992) 5 types of critical thinking (Hew & Cheung, 2003) to assess individual and community knowledge construction would be very useful in this regard.  In addition to being good knowledge workers, today’s learners will need to become excellent collaborators as well.  With this in mind, it would be helpful to apply Rourke et al’s (1999) 3 constructs of social presence and affective responses (Hew & Cheung, 2003) to assess social presence, and to apply Kirkley et al’s (1998) 7 means of moderator assistance (Hew & Cheung, 2003) to assess facilitation skills – or what I like to call “cognitive leadership” skills.  Of course, I do not mean to suggest that we implement the above mentioned frameworks in their pure form, but rather to reform our current assessment and evaluation policies and practices based upon a combination of these frameworks.

Bereiter, C. (2002). Education and Mind in the Knowledge Age. Mahwah, NJ: Lawrence Erlbaum Associates. Online:

Bereiter, C. (1997). Situated cognition and how to overcome it. In D. Kirshner & J. A. Whitson (Eds.), Situated cognition: Social, semiotic, and psychological perspectives (pp. 281-300). Hillsdale, NJ: Erlbaum.  Available:

Brown, A. L., & Campione, J. C. (1996). Communities of Learning and Thinking, or a Context By Any Other Name. Contemporary Issues in Teaching and Learning, 120-126.

Brown, J. S., Collins, A. & Duguid, P. Situated Cognition and the Culture of Learning (1989). Educational Researcher, Vol 18(1), 32-42  Available at:

Cole, M. & Wertsch, J. V. (1996).  Beyond the Individual-Social Antimony in Discussions of Piaget and Vygotsky.  Online:

Hew, K. F. & Cheung, W. S. (2003). Models to evaluate online learning communities of asynchronous discussion forums. Australian Journal of Educational Technology 2003, 19(2), 241-259. Online:

Hung, D. & Chen, D. T. (2001). Situated Cognition, Vygotskian Thought, and Learning from the Communities of Practice Perspective: Implications for the design of Web-based E-Learning.  Educational Media Internaitonal, 38(1), 3-12. Online:

Lave, J., & Wenger, E. (1990). Chapter 1 in Situated Learning: Legitimate Peripheral Participation. Cambridge, UK: Cambridge University Press. Online:

Vygotsky, L. (1934) 2. Piaget’s Theory Child Language and Thought.  Online:

Rogers, Y. (2006) Distributed Cognition and Communication. In The Encyclopedia of Language and Linguistics 2nd Edition. Edited by Keith Brown Elsevier: Oxford. 181-202. Online:

Russell, R. (?). Experience based learning theories.  The Informal Learning Review. Online:

Scardamalia, M. (2002). Collective Cognitive Responsibility for the Advancement of Knowledge. Liberal Education in a Knowledge Society, 67-98.

Scardamalia, M. (2000). Social and Technological Innovations for a Knowledge Society. In S. S-C. Young, J. Greer, H. Maurer, & Y.S. Chee (Eds.). Proceedings of the ICCE/ICCAI 2000: Volume 1. Learning Societies in the New Millennium: Creativity, Caring & Commitments. (pp. 22-27). National Tsing Hua University, Taiwan: Taipei. Online:

Scardamalia, M., & Bereiter, C. (1996). Student Communities for the Advancement of Knowledge. Communications of the ACM, 39(4), 36-37.

Smith, M.K. (2002). ‘Jerome S. Bruner and the process of education’, the Encyclopedia of Informal Education. Online:

Piaget, Vygotsky, Situated Cognition Vs. Mitra’s SOLE & Granny Clouds

I’ve been reading about social and cultural influences on learning, as well as Situated Learning:

Piaget, Vygotsky, Situated Cognition
Where Piaget asserted that children developed from the individual to the socialized (i.e. nonverbal autistic thought –> egocentric thought & speech –> logical thinking & socialized speech), Vygotsky argued that children developed from the socialized to the individual (i.e. social –> egocentric –> inner speech).

In situated cognition theory, activity and situations are integral to cognition and learning.

Vygotsky’s developmental theory and Situated Cognition theory can indeed work in tandem!  These theories imply that:

  1. Learners need to be situated within a collaborative learning context so they may co-investigate socially, eventually leading to internalized learning for each individual.
  2. These co-investigation activities must be authentic tasks, situated within the context of the subject domain of study, such that learners see themselves as practitioners of that domain.

Both #1 & #2 can be done face-to-face (F2F) or online, but I always think that a blended model (e.g., F2F + online) works best.

Mitra’s SOLE and Granny Clouds
Having read about Piaget and Vygotsky’s somewhat opposite theories of child development, it has made me give more thought to Sugata Mitra’s TED talk about The Child-Driven Education:

The idea that children will collaboratively learn with digital tools they’ve never used, heard of, or seen before; inspite of purposeful lack of instruction and scaffolding – is evidence to me that collaborative learning is natural to the human condition.  This is incredibly profound!  Furthermore, Sugata Mitra’s work seems to support Vygotsky’s developmental theory, rather than Piaget’s; as it seems that the children in his studies developed their skills and knowledge socially before the new knowledge became internalized within each individual child.

In his TED talk, Mitra mentions 2 quotes from Sir Arthur Charles Clarke, that I find very profound:
“A teacher that can be replaced by a machine, should be.”
“If children have interest, then education happens.”

The second half of this TED talk is particularly intriguing.  Sugata Mitra’s work on Self Organizing Learning Environments (SOLE) and the Granny Cloud.  He informally organized children into small groups and gave each group a set of questions, as well as the freedom to access the internet with 1 laptop,  discuss amongst their group, drop in on other groups, and no other scaffolding.  He’s found that such informal groupings and vague scaffolding sets the stage for children to self-organize as a learning community, producing deep learning.  His Granny Cloud work involved 200 volunteer English grandmothers who regularly taught English to children in India via Skype.

Mitra Vs. Situated Cognition
Mitra’s SOLE work is in sharp contrast to my classmates’ (Scott J.’s KEC note #1286 and Chad L.’s KEC note #1310) and my own thinking (KEC note #1404), that collaboration skills need to be explicitly taught.  In Mitra’s SOLE work, this is clearly not the case!  Is it possible, that by explicitly teaching collaboration, we stifle the degree to which children can collaborate effectively and learn deeply?  It also interesting to note the absence of any situatedness in Mitra’s SOLE work.  In fact, all of the learning tasks that he presented to various populations of children in various countries (India, Italy, England) were intentionally out of context and in some cases, not even in a language that the children knew!  Perhaps then, situated cognition is not a necessary component of effective/deep learning, merely an enhancing component?

Mitra’s speculation:  “Education is a self organising system, where learning is an emergent phenomenon…”.  He estimates that it will take 5 years and under $1 million to prove this experimentally, and that’s what he intends to do.  Mitra’s theory is a direct contradiction to Brown, Collins, and Duguid’s “cognitive apprenticeship” model (1989), where learning is embedded in activity and deliberate use is made of the social and physical context to immerse learners in the culture of the knowledge domain under study.  The one idea that Mitra’s work and the Cognitive Apprenticeship model have in common is that learning is supported by collaborative social interaction and social construction of knowledge.  As previously noted, Mitra’s work also contradicts the notion of “situated cognition“, in which activity and situations are integral to cognition and learning.

So who’s right?  There seems to be supporting evidence for all arguments!  Perhaps it’s a question of efficiency.  If we leave learners to collaboratively self-organize, they will learn deeply if given enough time.  If we situate (via cognitive apprenticeship) and scaffold learners’ collaborative learning, they will learn deeply, perhaps in less time?

Finally, Mitra emphasizes that the future of educational change is “A question of attitude, not technology”, and breaks down “The Arithmetic of Change” for us:
1 billion children
100 million mediators
10 million SOLEs
$180 billion
10 years

This indeed is an exciting and optimistic outlook for education!  Perhaps there is hope yet of rocketing the education dinosaur into the 21st century…

Note:  Sugata Mitra’s papers regarding the studies he mentioned in his TED talk can be found here (UTORid needed).