Collaborative learning research has revealed highly inconclusive findings regarding learning outcomes (Kester & Paas, 2005). Even if groups are effectively formed and cognitive learning processes are successfully supported, beneficial effects on learning are not always found. It has become clear that placing learners in a group and assigning them a task does not guarantee that they will work together, engage in effective collaborative learning processes, or guarantee positive learning outcomes (Kester & Paas, 2005). We believe that these results have, among other things, been caused by a lack of attention for the structures that constitute human cognitive architecture when deciding whether to use an individual or collaborative learning environment and when designing collaborative learning environments. Using cognitive load theory (CLT: Paas, Renkl, & Sweller, 2003, 2004), this study considers the human cognitive architecture (Sweller, 2004), specifically the limitations of the working memory capacity at the individual level, as an important reason to assign complex learning tasks to groups rather than to individuals. The processing capacity of working memory (WM) is limited to only 4 plus or minus 1 information element (Cowan, 2001) when these elements are interrelated because they have to be combined, contrasted or worked on. CLT is concerned with the learning of complex cognitive tasks, where learners are often overwhelmed by the number of interactive information elements (i.e., the intrinsic cognitive load is too high) that need to be processed simultaneously before meaningful learning can commence. This study considers groups as information processing systems consisting of multiple working memories. It can be argued that groups have effectively more processing capacity available than an individual information processing system with one WM. In a group, the cognitive load can be shared among group members enabling them to deal with more complex problems than individuals. This distribution advantage has been shown in the domain of cognitive brain research, where the capacity of the brain was increased by dividing the processing of complex tasks between the two hemispheres (interhemispheric), in stead of using one hemisphere ( Maertens & Pollmann, 2005). In this study it is therefore hypothesized that the more complex the task (i.e., the higher the intrinsic cognitive load), the more efficient it will become for individuals to cooperate with other individuals in a fashion that reduces this load. By contrast, we predict that less complex tasks that can be solved by an individual will lead to less efficient learning in groups than in individuals, because the required communication and coordination process (i.e., transaction costs) are not effective for the learning process and as such impose an extraneous cognitive load. Thus, the basic assumption of this study is that collaborative learning can only become more efficient than individual learning if the cognitive costs associated with learning the task (i.e., the intrinsic cognitive load) plus the cognitive costs associated with the communication and coordination of the knowledge between the group members (i.e., the extraneous cognitive load) exceed the cognitive resources that an individual can supply. In this study it is therefore also hypothesized that the lower the level of task complexity (i.e., the lower the intrinsic cognitive load) the more efficient it will become for learners to learn individually than in a group. Combining the two hypothesis this study could in addition give an indication of the level of task complexity at which it becomes more effective for learning to assign tasks to groups rather than to individuals.

A randomized 2 (Cognitive Capacity: individual vs. group) x 3 (Task Complexity: low, medium, high) factorial design with repeated measures on the latter factor is used to study the learning efficiency (Paas & Van Merriënboer, 1993) of 80 participants, either learning individually or as a member of a triad. The learning tasks are in the domain of inheritance (biology). Every problem solving task consists of a number of information elements that need to be combined to give a correct answer to a posted question. Each piece of information is relevant but insufficient for solving the problem, but combined with the other information it allows for solving the problem. This format allows for a clear differentiation in three task complexity levels (low, medium, high), which can be identified by Sweller and Chandler’s (1994) method based on the number of interactive information elements in a task. A task with a low level of task complexity contains three interrelated information elements, a task with a medium level of task complexity contains six interrelated information elements, and a task with a high level of task complexity contains nine interrelated information elements. In the learning phase, participants have to solve a low, medium and high complexity task, and rate the amount of invested mental effort on a 9-point cognitive load scale (Paas, 1992). Video recordings are used for qualitative analysis of the groups’ learning process. An individual transfer test and subjective ratings of mental effort in the test are used to calculate the learning efficiency. Data will be collected in December 2006.

References

Cowan, N. (2001). The magical number 4 in short-term memory: A reconsideration of metal storage capacity. Behavioral and Brain Sciences, 24, 87-114.

Kester, L. & Paas, F. (2005). Instructional interventions to enhance collaboration in powerful learning environments. Computers in Human Behavior, 21, 689-696.

Maertens, M., & Pollmann S. (2005). Interhemispheric resource sharing: Decreasing benefits with increasing processing efficiency. Brain and Cognition, 58, 183-192.

Paas, F. (1992). Training strategies for attaining transfer of problem-solving skill in statistics: A cognitive load approach. Journal of Educational Psychology, 84, 429-434.

Paas, F., & Van Merriënboer, J. G. (1993). The efficiency of instructional conditions: an approach to combine mental effort and performance measures. Human Factors, (35)4, 737-743.

Paas, F., Renkl, A., & Sweller, J. (2003). Cognitive load theory and instructional design: Recent developments. Educational Psychologist, 38, 1-4.

Sweller, J. (2004). Instructional design consequences of an analogy between evolution by natural selection and human cognitive architecture. Instructional Science, 32, 9-31.

Sweller, J., & Chandler, P. (1994). Why some material is difficult to learn. Cognition & Instruction, 12, 185-233.