Conference PaperPDF Available

Implicit and Explicit Memory in Learning from Social Software: A Dual-Process Account


Abstract and Figures

Inspired by a recent surge to understand social cognitive processes in collaborative knowledge building, we have devised an experiment in which students learned from contents of a wiki. One of the informative results we observed was a dissociation between implicit and explicit memory measures that we used to track student's learning: an association test, and the drawing of concept maps, respectively. We put these initial results in the context of experimental research in cognitive psychology and show how the co-evolution model (Cress and Kimmerle, 2008) could account for them. With several network measures, we also suggest some ways of how to measure assimilation and accommodation, both in internal and external knowledge representations.
Content may be subject to copyright.
Implicit and Explicit Memory in Learning from Social
Software: A dual-process account
Tobias Ley 1,2, Stefan Schweiger2, Paul Seitlinger3
1 Center for Educational Technology, Tallinn University, Narva Mnt 25,
10120 Tallinn, Estonia,
2 Cognitive Science Section, University of Graz, Universitätsplatz 2, 8010 Graz, Austria,
3 Knowledge Management Institute, Graz University of Technology, Inffeldgasse 21a,
8010 Graz, Austria,
Abstract. Inspired by a recent surge to understand social cognitive processes in
collaborative knowledge building, we have devised an experiment in which
students learned from contents of a wiki. One of the informative results we
observed was a dissociation between implicit and explicit memory measures
that we used to track student’s learning: an association test, and the drawing of
concept maps, respectively. We put these initial results in the context of
experimental research in cognitive psychology and show how the co-evolution
model (Cress and Kimmerle, 2008) could account for them. With several
network measures, we also suggest some ways of how to measure assimilation
and accommodation, both in internal and external knowledge representations.
Key words: Co-evolution model, implicit knowledge, network analysis
1 Collaborative Knowledge Building in the Use of Social Software
With the recent rise of social software and its use for educational purposes in schools,
universities and companies, there has been a renewed interest in collaborative
knowledge building. Cress and Kimmerle [3] have recently suggested a co-evolution
model of collaborative knowledge building in wikis and other social software systems
[6]. Knowledge building is understood as a co-evolution of internally and externally
represented knowledge by means of internalization and externalization processes.
Drawing on Piaget’s theory of cognitive development, it is assumed that in the face
of making sense out of encountered information with prior knowledge happens
through internal processes of assimilation and accommodation. In assimilation, an
existing schema is activated to encode information encountered. In accommodating,
on the other hand, an existing schema is restructured or a new one is generated
because the encountered information does not fit any existing schema. Similarly,
Cress and colleagues propose external mechanisms of collaborative knowledge
building in the wiki which they also differentiate into (external) assimilation and
accommodation where the former can be characterized by mainly quantitative
changes to the wiki, and the latter by qualitative restructuring.
Inspired by this model, we devised an experimental study in which students
learned from a wiki on the subject of narratology. Our particular focus in this study
was on measuring internal assimilation and accommodation: we were interested in
what students would learn from the content, how their internal knowledge
representation would change as a result of interacting with the wiki, and how this
could be measured. Encouraged by our previous research in memory processes in the
use of social software [8], we used two tests from two distinct classes of memory tests
to measure knowledge of our subjects. One test was assumed to be more sensitive
towards explicit knowledge that students had gained (drawings of concept maps), and
the other assumed to be more sensitive towards implicit knowledge. For the latter, we
employed an association tests in which subjects are asked to freely associate words
given a certain stimulus word [8]. The number of associations, then, is a measure of
the strength of implicit representation in memory for the stimulus word.
One of the somewhat surprising results of this study was a dissociation between the
implicit and the explicit memory measure. In this paper, we will discuss these
dissociations because they can advance theoretical understanding of collaborative
knowledge building and the underlying cognitive mechanisms. This has also practical
implications. Particularly, if assimilation and accommodation could be
(automatically) measured, it would be possible to devise feedback mechanisms that
could guide group work or formative assessment by a teacher. In the next section, we
will present the results of the experiment as they pertain to the dissociation. We then
relate these results to research into implicit and explicit memory processes in
cognitive psychology. From this, we suggest how the co-evolution model could
account for these results. We close with discussing practical implications.
2 An Experiment on Learning from Wikis
In our experimental study, students learned from a wiki about Narratology, the theory
and study of narratives and narrative structure. They had to learn about a model which
introduced several different narrative modes based on the role of the narrator and the
style of narrative. The model can be depicted graphically as a circle in which different
narrative modes can be placed at different sections around the circle depending on
where they fall on dimensions such as narrator vs. reflector, outside vs. inside
perspective. For example, one narrative mode is that of the first-person narrator which
is characterized by a narrator, who has an identity in the story and who takes an inside
perspective, i.e. having access to inner feelings of that person.
After learning a part of the model, students had to classify short passages of prose
according to its narrative mode. While one group of students received texts that were
in accordance with the model they had learned, others received passages that included
narrative modes not yet learned. We assumed the first group could assimilate the
examples into the existing schema they had constructed, while the second group
would have to accommodate their internal schema. Hence, assimilation vs.
accommodation instilled in our subjects was our main experimental manipulation
which we will call learning condition (ACC vs. ASS). 16 subjects were tested in each
condition. As dependent measures of their learning, students drew concept maps and
were asked to freely associate terms given a certain stimulus term (association test).
The stimulus terms were taken from the study material. Both tests were applied prior
(pretest), as well as after the manipulation at the end of doing all exercises (posttest).
Just looking at the number of concepts introduced in the concept maps and the
number of associations reveals an interesting pattern: The number of concepts
introduced in the concept maps increased from pretest to posttest for both groups, a
clear sign of explicit leaning (ACC group: Mpre=12.3 vs. Mpost=14.8; t15=-2.36, p<.05;
ASS group, Mpre=10.8 vs. Mpost=14.6; t14=-6.76, p<.001). When looking at the number
of associations, on the other hand, these only increased for the ASS group (Mpre=3.8
vs. Mpost=4.4; t15=-2.92, p<.01), but not for the ACC group (Mpre=4.3 vs. Mpost=4.4;
t14=-0.08, ns). This points to a dissociation of the two memory measures: given the
experimental manipulation (i.e. inducing assimilative vs. accommodative processes),
an effect on explicit memory was observed, but not uniformly on associative strength.
Because we were especially interested in the structural changes of students’
knowledge, we employed several network measures as they have previously been
found to detect changes in structural aspects of knowledge [2] [4]. These measures
can be employed both to concept maps, as well as to association networks which are
formed by the stimulus terms from the association test and a co-occurrence relation.
We measured the density of the network (proportion of edges) and its overall
betweenness centrality (which measures the average shortest path between all nodes).
Fig. 1 shows the results for density of the two networks. For the Concept Maps
(left), the main effects were not significant, but the interaction effect time x learning
condition was significant (F1,29 = 4.84, p < .05, ω2 = .15): while there was no change
in density in the ACC group over time (t15 = -.44, ns, r = .17), there was a substantial
drop of density in the ASS group over time (t14 = 4.56, p < .001, r = .50). For the
association network (right), there was also no main effect, but the interaction
approached statistical significance and the effect size indicated a moderate effect
(F1,29 = 2.32, ns, ω2 = .10), suggesting that the sample size might have been too small
to show statistical significance.
Fig. 1. Density of Concept Maps (left) and Association networks (right)
What is noteworthy about these results is that the two interaction effects for the
two measures (one significant, one approaching significance) are in opposing
directions. While in the assimilation condition, the drawn Concept Maps were sparser
in the posttest than before, there was a tendency that density increased in the
association networks. For the accommodation condition, these effects were reversed.
A higher density in the concept maps points to the fact that more relations were drawn
between concepts, while a higher density in the association networks may indicate
less distinct internal categories (more similar associations given two stimuli).
Fig. 2 shows the same results for the measure of betweenness centrality.
Hierarchical networks have higher degrees of betweenness centrality, while for
networks that are more evenly connected, betweenness centrality is lower. For the
Concept Maps (left), the time x learning condition interaction showed that the change
in the ASS group was different to the change in the ACC group (F1,29 = 14.69, p <
0.001, ω2 = .18). While the drop of centrality in the ACC group was strong (t15 = 9.44,
p < .001, r = .62), the ASS group did not show a significant change over time (t14 =
1,24, ns, r = .29). For the Association Networks (right), no effects were observed.
Fig. 2. Betweenness Centrality of Concept Maps (left) and Association networks (right)
Taken together, the results support the notion that structural changes in students’
knowledge were different for the two groups. While for the assimilation group,
concept maps became sparser and more hierarchical, the concept maps of the
accommodation group were more evenly connected while having a comparable
degree of density. This is generally in line with the co-evolution model.
Accommodation (and hence restructuring or newly established schemata) would lead
to more network-like structures in the concept map with a lower degree of
betweenness centrality (see also [4]). In contrast, assimilation would lead to a simple
adding of nodes to an existing hierarchical structure, and as a consequence
betweenness centrality would be higher.
The study gives other evidence of dissociation of the two measures, explicit and
implicit: we came up with several measures to determine whether students had
learned more (e.g. number of errors in the concept map and number of correct
associations in the association test). Correlational evidence suggest that for better
students betweenness of the concept map was higher while betweenness of the
network formed from co-occurring associations was lower. Again, distinctness of
internal concepts would likely increase centrality of the network which was observed
for the implicit measure. The explicit measure, on the other hand, showed a decrease
in centrality which would point to the fact that better students had discovered more
explicit connections between concepts.
3 Implicit and Explicit Memory in Knowledge Building
The results pertaining to the dissociation are informative form a theoretical
perspective because they shed light on the underlying (non-observable) memory
processes. While our results certainly are only initial findings, they may point to the
important and differential role that implicit and explicit memory processes play in
collaborative knowledge building. In experimental psychology, the dissociation
between implicit and explicit processing in memory tasks has long been used to
discover underlying memory processes of encoding and retrieval [1] [9]. While
explicit memory preserves the context from the study episode, implicit memory does
not support the conscious retrieval of the study context. Retrieving from explicit
memory involves an experience of conscious and deliberate recollection, while
retrieving from implicit memory is relatively automatic and effortless and is
associated with an experience of familiarity or “just knowing” [5]. While explicit
memory is mainly responsible for the conscious recollection in typical memory tests
(e.g. recognition or recall tests of studied material), implicit memory facilitates
performance on certain tasks (e.g. quickly associating words as in our association test)
without conscious intention of recalling the study episode.
Why then would implicit and explicit memory processes play a role in
collaborative knowledge building? If assimilation and accommodation can be
understood as a process of schema activation and restructuring [3], we may deduce
that implicit and explicit memory processes are differentially impacted: when making
sense of encountered information by applying an existing internalized schema, the
information is processed in a schema congruent manner. We assume that this is
predominantly based on implicit processing. Activation of the schema in semantic
memory should happen to a large degree unintentionally, especially in a situation of
experienced performance. The application of the schema, then, should mainly be
based on semantic or procedural knowledge both again to a larger degree implicit.
This is in line with research showing that under certain conditions the activation of a
schema remains unconscious and schema-congruent information is only shallowly
processed. In our experiment, this may have been the reason why in the assimilation
group, the implicit memory measure (associative strength of the concepts learned)
showed a stronger improvement than in the accommodation group.
When, on the other hand, one is confronted with schema-incongruent information
and the activated schema fails to provide a suitable way to represent the external
information, the information is processed more elaborately, both semantically and
perceptually. This is in line with [7], who found higher levels of explicit recollection
for incongruent items, while implicit familiarity was greater for congruent items.
Information processing in this situation should therefore predominantly impact
explicit memory. The co-evolution model assumes that as a result of this conflict, an
existing schema is changed, or a new schema is generated. This should be largely
dependent on explicit processing, as high levels of semantic elaboration are involved
to perform this constructive task. Again, in our experimental results, there is clear
indication of this restructuring in the accommodation group for the explicit measure
(increased network density and decreased centrality of the concept maps). The
implicit measure, on the other hand, does not show these same effects.
4 Conclusions and Outlook
We have presented some initial results of an experiment designed to test the co-
evolution model of collaborative knowledge building. We have pointed out results
that show a possible dissociation between implicit and explicit memory processes
which are theoretically instructive. Two limitations of this research have to be
mentioned. First and foremost, the results are exploratory in so far as we did not have
any a-priori hypotheses about the dissociations. Hence, the results have provided the
basis for hypothesis generation in section 3 that will inform future studies. Secondly,
it has to be said that tests for explicit and implicit memory have shown different levels
of reliability. This has to be kept in mind and accounted for in future research.
As a next step, we will be transferring the research to a situation where students
collaboratively generate artefacts. The graph-based measures we have suggested in
this paper are appropriate also for this purpose. For example, the link structure of
generated wiki articles could be analyzed as a measure of explicit knowledge. Similar
to the association networks, a term network could be constructed using term co-
occurrence on a sentence level obtained from the text students produce in the wiki.
This could be used as a measure for implicit knowledge in external representations.
These measures would constitute theory guided and validated measures, and they
should mirror the internal measures introduced in the present study.
1. Brainerd, C.J., Wright, R., & Reyna, V.F.: Dual-Retrieval Processes in Free and
Associative Recall. J. of Memory and Language, 46, 120-152 (2002)
2. Coronges, K.A., Valente, T.W., Stacy, A.W., Coronges, K.: Structural comparison of
cognitive associative networks in two Populations. J. of Appl. Soc. Psych., 37, 2097-2129
3. Cress, U., Kimmerle, J.: A systemic and cognitive view on collaborative knowledge
building with wikis.', J of Computer-Supported Collab Learning 3 (2) , 105-122 (2008)
4. Hay, D.B., Kehoe, C., Miquel, M.E., Hatzipanagos, S., Kinchin, I.M., Keevil, S.F.:
Measuring the quality of e-learning. British J. of Educ. Technol., 1037-1056 (2008)
5. Kelley, C.M., Jacoby, L.L.: Recollection and Familiarity: Process-Dissociation. In E.
Tulving & F. Craik (Eds.) The Oxford handbook of memory, pp. 215-228, Oxford
University Press (2000)
6. Kimmerle, J.; Cress, U., Held, C.: The interplay between individual and collective
knowledge: technologies for organisational learning and knowledge building, Knowledge
Management Research & Practice 8(1) , 33-44 (2010)
7. Lampinen, J.M., Copeland, S.M., Neuschatz, J.S.: Recollections of things schematic: Room
schemas revisited. J. Exp. Psych.: Learning, Memory and Cognition, 27, 1211-1222 (2001)
8. Ley, T., Seitlinger, P.: A Cognitive Perspective on Emergent Semantics in Collaborative
Tagging: The Basic Level Effect. In International Workshop on Adaptation in Social and
Semantic Web, pp. 13-18, CEUR Workshop Proceedings (2010)
9. Nelson, D.L., et al.: How implicitly activated and explicitly acquired knowledge contribute
to the effectiveness of retrieval cues. Memory & Cognition, 35, 1892-1904 (2007)
... This should enable a learner to receive meaningful, relevant and individualized support for his learning needs in the context of his work, and take better advantage of the multitude of learning opportunities that are available around him. As much research around Communities of Practice (Johnson, 2001;Lave & Wenger, 1991;Wenger, 1998), Intersubjective Meaning Making (Stahl, Koschmann & Suthers, 2006;Suthers, 2005b;Suthers, 2006) and collaborative knowledge building (Cress & Kimmerle, 2008;Kump, Moskaliuk, Dennerlein & Ley, 2013;Ley, Schweiger & Seitlinger, 2011;Stahl, 2000) has shown (see also Section 2 for more details), a critical factor is in how far the use of technologies are based on shared meaning. ...
... Cress and Kimmerle's Co-Evolution Model (Cress & Kimmerle, 2008;Ley et al., 2011;Kump et al., 2013) is pulled upon to specify the interaction of internal and external knowledge representations and their development over time. It describes their interaction as a form of co-evolution between the cognitive (user) and the social system (social media) and assumes that both systems influence each other based on the two processes of internalization and externalization (see Figure 3): First, person three (cognitive system = CS3) internalizes some information (light triangle) of the wiki (social system) into her own knowledge and learns something (transfer from state CS3 to CS3'). ...
... (Harrer et al., 2008) 6 Thereby, the model draws upon schema theory and the ideas of Piaget (Piaget, 1970;Block, 1982;Miller, 2010) to precisely analyze the underlying learning mechanism in both systems and their influence on knowledge representations. Methods for a sophisticated (structural) analysis of corresponding internal knowledge representations via concept maps and association tests have been suggested by Ley et al. (2011) and Kump et al. (2013). ...
... In these laboratory experiments, the authors found that incongruence would trigger different modes of externalization (accommodation and assimilation), and that different modes of externalization would result in different modes of internalization. This finding is also backed by a study of Ley et al. (2011), who found preliminary evidence that incongruence of the information in the artifact and the knowledge of a participant leads to internal accommodation and assimilation, and that these processes can be detected by different knowledge measures. In their study, cognitive conflict was triggered by different exercises that matched or did not match what participants had encountered in the wiki, thereby triggering participants to assimilate or accommodate. ...
... Specifically, to distinguish internal accommodation and assimilation, measures of internalization should also allow us to differentiate modifications in knowledge structures that resulted from the two modes of internalization. To measure the extent of internalization, and to distinguish modifications in knowledge structures that are due to accommodation and assimilation, in their experiment in the laboratory, Ley et al. (2011) used concept maps (e.g., Novak, 1972, 1998; Ruiz-Primo, 2000; Ruiz-Primo and Shavelson, 1996), and word association tests (e.g., Meyer, 2007, Jonassen, Beissner & Yacci, 1993). These measurements are basically also applicable in settings where the prior knowledge of the participants as well as the information which they process during the experiment cannot be controlled. ...
... Besides measuring the extent of internalization, studies suggest that concept maps also measure structural knowledge that is not accessible by conventional test (Ifenthaler, Masduki & Seel, 2009; Markham, Mintzes & Jones, 1994; Wallace & Mintzes, 1990). In line with preliminary findings by Ley et al. (2011), we thus assume that concept maps allow a distinction between internal accommodation and assimilation. ...
The co-evolution model of collaborative knowledge building by Cress & Kimmerle (2008) assumes that cognitive and social processes interact when users build knowledge with shared digital artifacts. While these assumptions have been tested in various lab experiments, a test under natural field conditions in educational settings has not been conducted. Here, we present a field experiment where we triggered knowledge co-evolution in an accommodation and an assimilation condition, and measured effects on student knowledge building outside the laboratory in the context of two university courses. Therefore, 48 students received different kinds of prompts that triggered external accommodation and assimilation while writing a wiki text. Knowledge building was measured with a content analysis of the students‟ texts and comments (externalization), and with concept maps and association tests (internalization). The findings reveal that (a) different modes of externalization (accommodation and assimilation) could be triggered with prompts, (b) across both conditions, this externalization co-occurred with internalization (student learning), and (c) there is some evidence that external assimilation and accommodation had differential effects on internal assimilation and accommodation. Thus, the field experiment supports the assumptions of the co-evolution model in a realistic course setting. On a more general note, the study provides an example of how wikis can be used successfully for collaborative knowledge building within educational contexts.
... Evidence from the previous research [18] indicates towards the direction that the Uptake model offers an interesting methodological framework, especially thanks to its ability to achieve media independence, which has been deemed problematic for the studies in the past (e.g. [19]). ...
Full-text available
A growing interest in research focuses on teachers’ large-scale socio-technical networks. Social learning approaches such as social constructivist theory is well established, however, the current challenges lie in creating reliable methods to gather evidence of how and under which conditions social learning takes place in such socio-technical networks and how does it support teachers’ lifelong learning goals. The field of Learning Analytics (LA) addresses the issue of individual learners, whereas Social Learning Analytics (SLA) addresses that of groups’ processes in knowledge construction. The eTwinning action is used as a case study for applying the concepts of Social Learning Analytics. Our interest is on teachers’ co-operation behaviour and patterns within a socio-technical network and how that can support teachers’ continuous professional development. The eTwinning platform currently hosts more than 160000 European teachers. We first introduce the underlying pedagogical and lifelong learning related assumptions regarding teachers’ co-operation. To better understand the type of activities that teachers undertake in eTwinning, they are classified according to the OECD‘s indices for teachers’ co-operation. This creates the core of the eTwinning Analytics framework, which operationalises activities and allows them to be better measured and monitored. A snapshot of data from the platform is used as a proof of concept.
Full-text available
Shelley’s Heart is a locative learning tool geo-linked to the churchyard where Mary Shelley is buried along with the heart of her husband, poet Percy Shelley. To commemorate the 200th anniversary of the publication of Frankenstein, this interactive multi-media project debuted on Halloween night, 2018. Featuring biographical facts, quotes and fictional narratives related to the Shelleys and their friends, John Keats and Lord Byron, it provides an active learning experience that is freely available to the general public. Locative projects often feature either factual information (museum audio guides), or fictional stories (ambient literature), but Shelley’s Heart combines fact and fiction in order to promote critical thinking. Refining this approach required extensive trial and error experimentation, collaborative brainstorming, user testing, and several phases of production. Insights garnered from this process led to the development of strategies that simultaneously promote implicit and explicit cognition. Cognitive science refers to this interplay as, the ‘dual process model’ [Kahneman, D. 2011. Thinking, Fast and Slow. New York: Farrar, Straus and Giroux]. This paper explores how the creation of Shelley’s Heart led to the development of three key dual process design strategies: ‘Match,’ ‘Affect’ and ‘Prime’, collectively known as ‘MAP’.
Tag clouds generated in social tagging systems can capture the collective knowledge of communities. Using as a basis spreading activation theories, information foraging theory, and the co-evolution model of cognitive and social systems, we present here a model for an extended information scent, which proposes that both collective and individual knowledge have a significant influence on link selection, incidental learning, and information processing. Two experimental studies tested the applicability of the model to a situation in which individual knowledge and collective knowledge were contradictory to each other. The results of the first experiment showed that a higher individual strength of association between a target in demand and a tag led to a higher probability of selecting corresponding links, combined with less thorough information processing for non-corresponding links. But users also adapted their navigation behavior to the collective knowledge (strength of associations of tags) of the community and showed incidental learning during navigation, which resulted in a change of their individual strength of associations. The second experiment confirmed these results and showed, in addition, that the effects also occurred for indirect associations. Altogether, the results show that the extended information scent is an appropriate and fertile model for describing the interplay of individual knowledge and the collective knowledge of social tags.
Tagging systems represent the conceptual knowledge of a community. We experimentally tested whether people harness this collective knowledge when navigating through the Web. As a within-factor we manipulated people's prior knowledge (no knowledge vs. prior knowledge that was congruent/incongruent to the collective knowledge inherent in the tags). As between-factor we manipulated whether people had tag clouds available or not. In line with the Information Foraging Theory and with the Co-Evolution Model of individual learning and collective knowledge building, we found that people's prior knowledge and tag clouds influenced their navigation. Tags which were congruent with people's prior knowledge had a higher probability of being selected. A knowledge test showed that participants adapted their individual conceptual knowledge to the collective knowledge. This incidental learning shows that people harness collective knowledge just by navigation with tag clouds.
Full-text available
Researching the emergence of semantics in social systems needs to take into account how users process information in their cognitive system. We report results of an experimental study in which we examined the interaction between individual expertise and the basic level advantage in collaborative tagging. The basic level advantage describes availability in memory of certain preferred levels of taxonomic abstraction when categorizing objects and has been shown to vary with level of expertise. In the study, groups of students tagged internet resources for a 10-week period. We measured the availability of tags in memory with an association test and a relevance rating and found a basic level advantage for tags from more general as opposed to specific levels of the taxonomy. An interaction with expertise also emerged. Contrary to our expectations, groups that spent less time to develop a shared understanding shifted to more specific levels as compared to groups that spent more time on a topic. We attribute this to impaired collaboration in the groups. We discuss implications for personalized tag and resource recommendations.
Full-text available
This paper shows how concept mapping can be used to measure the quality of e-learning. Six volunteers (all of them 3rd-year medical students) took part in a programme of e-learning designed to teach the principles of magnetic resonance imaging (MRI). Their understanding of MRI was measured before and after the course by the use of concept mapping. The quality of change in individuals' maps was assessed using criteria developed to distinguish between meaningful and rote-learning outcomes. Student maps were also scored for evidence of conceptual richness and understanding. Finally, each map was compared directly with the content of the electronic teaching material. The results show that many of the student misconceptions were put right in the course of their learning but that many of the key concepts introduced in the teaching were ignored (or sometimes learnt by rote) by the students. This was because the teaching material locked these new ideas in structures and terminology that precluded meaning-making among non-experts. Our data suggest that students' prior knowledge is a key determinant of meaningful learning. We suggest that this must be acknowledged if the design and use of electronic teaching material is also to be meaningful. Ultimately, measures of student learning are the only authentic indicators of the quality of teaching through technology.
Full-text available
Wikis provide new opportunities for learning and for collaborative knowledge building as well as for understanding these processes. This article presents a theoretical framework for describing how learning and collaborative knowledge building take place. In order to understand these processes, three aspects need to be considered: the social processes facilitated by a wiki, the cognitive processes of the users, and how both processes influence each other mutually. For this purpose, the model presented in this article borrows from the systemic approach of Luhmann as well as from Piaget's theory of equilibration and combines these approaches. The model analyzes processes which take place in the social system of a wiki as well as in the cognitive systems of the users. The model also describes learning activities as processes of externalization and internalization. Individual learning happens through internal processes of assimilation and accommodation, whereas changes in a wiki are due to activities of external assimilation and accommodation which in turn lead to collaborative knowledge building. This article provides empirical examples for these equilibration activities by analyzing Wikipedia articles. Equilibration activities are described as being caused by subjectively perceived incongruities between an individuals' knowledge and the information provided by a wiki. Incongruities of medium level cause cognitive conflicts which in turn activate the described processes of equilibration and facilitate individual learning and collaborative knowledge building.
Full-text available
This article presents a framework model that defines knowledge building as a co-evolution of cognitive and social systems. Our model brings together Nonaka's knowledge-creating theory and Luhmann's systems theory. It is demonstrated how collaborative knowledge building may occur – in an ideal situation – within an organisation, when people interact with each other using shared digital artefacts. For this purpose, three different technologies are introduced as examples: social-tagging systems, pattern-based task-management systems, and wikis. These examples have been chosen to demonstrate that knowledge building can occur with respect to both declarative and procedural knowledge. The differences and similarities between these technologies, as far as their potential for organisational knowledge building is concerned, are discussed in the light of the framework model.
Full-text available
In 2 experiments, the authors examined the effects of schemas on the subjective experience of remembering. Participants entered a room that was set up to look like a graduate student's office under intentional or incidental learning conditions. They later took a recognition memory test that included making remember-know judgments. In Experiment 1, they were tested during the same session; in Experiment 2 they were tested either during the same session or after a 48-hr delay. Consistent with the authors' predictions, memory for atypical objects was especially likely to be experienced in the remember sense. In addition, false remember judgments rose dramatically after the 48-hr delay, especially for participants in the incidental learning condition. Results are discussed in terms of schema theory, fuzzy-trace theory, and the distinctiveness heuristic.
In 2 experiments, the authors examined the effects of schemas on the subjective experience of remembering. Participants entered a room that was set up to look like a graduate student's office under intentional or incidental learning conditions. They later took a recognition memory test that included making remember-know judgments. In Experiment 1, they were tested during the same session; in Experiment 2 they were tested either during the same session or after a 48-hr delay. Consistent with the authors' predictions, memory for atypical objects was especially likely to be experienced in the remember sense. In addition, false remember judgments rose dramatically after the 48-hr delay, especially for participants in the incidental learning condition. Results are discussed in terms of schema theory, fuzzy-trace theory, and the distinctiveness heuristic.
Recent dual-retrieval accounts of free recall postulate that a memory target can be recalled either by directly accessing its verbatim trace or by reconstructing it from semantic or other relational information. We introduce a simple paradigm, derived from the classic Estes RTT procedure, that separates direct access from reconstruction and that separates reconstruction from a metacognitive judgment process that authorizes reconstructed targets for output. Results are reported from four experiments, two that applied the paradigm to free recall and two that extended it to associative recall. The principal findings were that (a) direct access was enhanced by manipulations that made targets' surface forms easier to process or that focused recall on individual targets, (b) reconstruction was enhanced by manipulations that made targets' meaning content easier to process or that focused recall on groups of targets, and (c) such manipulations produced single dissociations, double dissociations, and reversed associations between direct access, reconstruction, and metacognitive judgment. We discuss how this paradigm might be exploited to unify dual-retrieval conceptions of recall and recognition.
The cognitive associative structure of 2 populations was studied using network analysis of free-word associations. Structural differences in the associative networks were compared using measures of network centralization, size, density, clustering, and path length. These measures are closely aligned with cognitive theories describing the organization of knowledge and retrieval of concepts from memory. Size and centralization of semantic structures were larger for college students than for 7th graders, while density, clustering, and mean path length were similar. Findings presented reveal that subpopulations might have very different cognitive associative networks. This study suggests that graph theory and network analysis methods are useful in mapping differences in associative structures across groups.