The Impact of Guided Discovery vs. Didactic Instruction on Learning

Initially posted in

Previous research has identified didactic instruction an effective approach for learners who lack prior knowledge. The evidence suggests that the degree of guidance should vary with the age of learners. Direct instruction can be more beneficial for younger learners (e.g., elementary and middle school children), whereas older ones gain more with non-directive guidance or guided discovery [1]. Different research findings indicate that guided discovery is more effective than lecture-based instruction in that learners develop a deeper understanding of concepts and their underlying structure.

In their study, Schwartz and colleagues [2], proposed an alternative approach to didactic instruction in teaching simple concepts in classes. They argued that didactic teaching undermines the processes of discovery, but withholding the explicit instruction and allowing learners to discover by themselves enhance deep learning and increase transfer. They conducted two experiments to investigate this theory.

In experiment 1, the authors compared two groups of students: the first group invented solutions with contrasting cases to come up with the ratio of density before receiving instruction on the concept of density; the second group received the instruction first and then practiced with the same cases. The contrasting cases in this activity consisted of three levels of structure: (1) surface features which were irrelevant to the concept, (2) density or deep structure, and (3) ratio structure, which was the invariant under transformation across all the levels.

The authors conducted the study in two phases. In phase 1, Day 1, the first group received a directive asking them to invent an index and some examples of it. They were then asked to do the contrasting cases worksheets. The second group received the one-page instruction on the concept of density, with its formula and some worked examples. In Day 2, both groups were tested on the recall of information by drawing the structures in the worksheet within 10 minutes. The authors found that the second group did not recreate the deep structure, which indicated their understanding of the ratio structures, while the first group performed better. Further, both groups performed the same in surface feature recall.

In the subsequent days that included Phase 2, both groups were given a variety of tasks on ratio structures until Day 5, and their immediate transfer of learning was measured on Day 8. On Day 5, the first group received a lecture on density following the completion of the activities and test of immediate transfer. To measure immediate transfer, both groups were asked to describe some diagrams. On Day 11, they were given a word problem test. On Day 21, their delayed transfer was measured by determining the stiffness of trampoline fabrics.

Results showed that the first group demonstrated a better transfer of learning in terms of applying the density ratio to other tasks. However, in recall of the ratio structure, the groups were not different. This means remembering the formula did not entail mastery of the phenomenon. The authors concluded that the first group learned the concept on their own by looking at multiple instances to develop the ratio formula. Their learning strategy led to better transfer of learning compared to the second group who used the formula for each case separately without considering the use of its concept across multiple cases.

In the second study, the authors wanted to find out what was transferred and what led to it in the first group. They repeated the same steps with two other groups and this time video-taped 24 students from both groups to record their strategies in solving the cases. Findings showed that the treatment effect was strong even for low-achieving students of the first group. They ensured that higher achievement and lower achievement as a covariate did not interfere with the results. In effect, they found that the first groups (even lower achieving) outperformed the second groups (higher achieving) in deep structure recall. They also showed better transfer of understanding of the ratio.

Findings of a recent study [2] extended the previous study by showing that the guided discovery aids students in developing a deeper understanding of complex concepts compared to those who receive lecture-based instruction. These results indicate that guided discovery approach help students build on their prior knowledge while giving them ownership of the problem-solving process. In this approach, they can test their ideas against alternative perspectives and develop a contextual understanding of concepts.
What does it Mean to Educators and Learning Designers?

Firstly, guided discovery should not be mistaken with discovery learning. In guided discovery, learners have access to prompts and domain knowledge from experts. They are not left on their own to explore and discover ideas, but are guided through well-designed activities and questions that can prompt them to build the necessary knowledge.

Secondly, learners will still receive instruction but at a later time. So this does not mean that you should withhold information or instruction from them completely.

Lastly, if you want your learners to apply what they have learned to different contexts, guided discovery may be a better approach. Results of these two studies showed that learners may become dependent on concept or problem without recognizing its applicability when they receive direct instruction. In such instances, didactic teaching may limit learners’ understanding of deep structure of a concept, whereas guided discovery may result in better learning retention and transfer.

These findings are aligned with extensive review of literature by Mayer [3], who argued that discovery learning must incorporate guided instruction and corrective feedback. Therefore, engaging learners in cognitive activities coupled with guidance promotes meaningful learning.

[1] Ramanujan, D., Zhou, N., & Ramani, K. (2019). Integrating environmental sustainability in undergraduate mechanical engineering courses using guided discovery instruction. Journal of Cleaner Production, 207, 190-203.
[2] Schwartz, D., Chase, C., Oppezzo, M., & Chin, D. (2011). Practicing versus inventing with contrasting cases: The effects of telling first on learning and transfer. Journal of Educational Psychology, 103(4), 759-775.
[3] Mayer, R. (2004). Should there be a three-strikes rule against pure discovery learning?: The case for guided methods of instruction. American Psychologist, 99(1), 14-19.