Chi’s ICAP (Interactive, Constructive, Active, Passive) (2009) provides a framework on the hierarchical organization of activities, their social levels and their learning effectiveness. Essentially, it explains how different learning activities at different social levels (individual, pair, group, plenary) might mediate learning, i.e., invoke different cognitive processes and facilitate different learning outcomes.
Essentially, the ICAP model hypothesizes that interactive activities are better than constructive activities, which in turn might be better than active activities, which are better than passive activities. However, it is also important to note that Chi also cautioned that not all intended activity modes maintain its status quo throughout the learning process. An interactive activity is no longer interactive if partners are not responding to each other’s contribution. Likewise, interaction with an expert in instructional dialogues (such as explaining, providing corrective feedback and scaffolds) is regarded as guided-construction, and is therefore, considered as an interactive activity when students respond in a meaningful and substantial way.
The ICAP framework provides two helpful insights in the design of classroom learning environments:
- It focuses our attention on the roles of overt learning activities and their relationship to the internal processes.
- By making a distinction between interactive, constructive, active and passive activities, the framework compares and contrasts one learning activity with another learning activity during a learning phase from a learner’s perspective
This is to say that the correct classification of an activity rests on the analyses of the content of the outputs. On this selfsame note, Chi and Wylie (2014) foregrounds some inherent limitations and caveats to the taxonomy of the four modes of activities and the hypothesized knowledge change processes and the eventual cognitive outcomes. Hence, it is possible that two interactive learning activities can engender different cognitive processes and outcomes if they are premised on differing task structures and/ or assessment measures. For instance, when the assessment does not align with the activity mode – the desired learning gains are not measured: test items/ assessments are superficial and too general for the detection of differences in the learning outcomes.
In sum, what essentially qualifies an activity as interactive, constructive, active or passive, is contingent both on the task and assessment structure to bring about the desired learning outcomes of that said activity mode.