Karpicke: Expanding retrieval schedules produced mixed results

From supermemo.guru
Jump to navigation Jump to search

Three decades since the birth of spaced repetition there is still a degree of confusion in scientific community. Poorly designed experiments can produce mixed results in spaced repetition. Two component model of long-term memory can be used to show that expanding, equal, and/or contracting schedules can all be made superior given appropriate timing. My own failed experiment from 1985 is a simple example. Moreover, different memory mechanisms are involved for spaces measured by time, intervening items, or trials.

Researchers can test individually hand-picked schedules till cows come home and never come with a clear conclusion. Without mathematical modelling, studying memory reminds the study of planets before the days of Copernicus.

The notion that expanding retrieval schedules would produce the best long-term retention gained acceptance after it was first proposed (for reviews, see Balota, Duchek, & Logan, 2007; Roediger & Karpicke, 2010). However, the results of several experiments have been mixed. Some experiments have obtained positive effects of expanding retrieval schedules relative to appropriate equally spaced control conditions (e.g., Cull, Shaughnessy, & Zechmeister, 1996, Experiments 1 and 4; Storm, Bjork, & Storm, 2010, Experiments 2 and 3). Others have found no differences between expanding and equally spaced conditions (e.g., Carpenter & DeLosh, 2005; Cull, 2000; Cull et al., 1996, Experiments 2, 3, and 5; Karpicke & Roediger, 2010; Logan & Balota, 2008; Pyc & Rawson, 2007; Storm et al., 2010, Experiments 1 and 3). Still other experiments have obtained advantages of equally spaced retrieval over expanding retrieval schedules (e.g., Cull, 2000; Karpicke & Roediger, 2007a; Logan & Balota, 2008)

Title: Spaced retrieval: absolute spacing enhances learning regardless of relative spacing

Link: http://www.ncbi.nlm.nih.gov/pubmed/21574747

Backlink: Why spaced repetition research kept failing?

This reference is used to annotate "History of spaced repetition" (2018) by Piotr Wozniak