Dr. Larsen emphasizes that “greater retrieval effort in the practice tests produces greater long-term retention.” In other words, a question that challenges learners and requires them to work to dig the answer from their memory is going to generate greater long-term retention than a question that focuses on regurgitating a simple answer (e.g what is the capital of Japan?). Structuring multi-layered scenario-based questions is an excellent means to achieve this ‘desirable difficulty’.
Repeated testing (retrieval practice) facilitates the process of memory reactivation and updating that is a critical part of memory consolidation and reconsolidation. He confirms our research findings that there are diminishing returns after several repetitions of a question. For this reason, Qstream is structured to limit the number of times that questions are presented to learners (usually max 2-3).
Short spacing intervals lead to improved retention of knowledge and skills over a short retention interval. Longer spacing intervals are needed to retain information over longer periods of time. My favorite research article that addresses this was published in the journal Psychological Science by Drs. Nicholas Cepeda and Harold Pashler of York University and UCSD entitled ‘Spacing Effects in Learning: A Temporal Ridgeline of Optimal Retention’. Based on their impressive research findings, they estimate that the optimal spacing interval for longer-term retention is about 5-10% of the desired duration of optimal retention. I love their quote in the discussion: “However, for practical purposes, the results also reveal a sobering fact: The optimally efficient gap between study sessions is not some absolute quantity that can be recommended, but rather depends dramatically on the (retention interval).” Of note, this desire for ever longer spacing intervals needs to be counterbalanced by the need to construct Qstreams that can be completed in a manageable duration and provide a meaningful end-goal for learners (progression and completion dynamics). The default settings for Qstream aim to balance all of these important yet competing goals.
While feedback is not necessary for the testing effect (“retrieval practice”) to be effective, studies have shown that feedback can dramatically improve longer-term learning of knowledge and skills, in one study by more than 100%. There are two other benefits of feedback: (1) it allows learners to more accurately assess their performance and consequently modify their learning approaches, and (2) it allows for learners who answer a question incorrectly to focus on the correct answer and to not strengthen their memories of the incorrect answer. At Qstream, we feel that concise and targeted feedback is a critical component of an effective learning process.