Date of Award

Fall 12-21-2020

Level of Access Assigned by Author

Open-Access Thesis

Degree Name

Doctor of Philosophy (PhD)




Shawn Ell

Second Committee Member

Thane Fremouw

Third Committee Member

Mollie Ruben

Additional Committee Members

Jordan LaBouff

Benjamin Guenther


Testing has been shown to improve long-term memory retention by decreasing the amount of material forgotten, a phenomenon known as the testing effect. This positive impact of testing has been shown using direct tests of memory that require declarative memory, things like memorizing word-pairs and single-word lists. This dissertation is the first research to investigate how testing impacts nondeclarative memory using three experiments. The first and second experiment utilize the word fragment completion task to measure the effect that testing has on words learned via methodology thought to recruit either declarative or nondeclarative memory. The third experiment utilizes a probabilistic category learning task known as the weather prediction task. This task requires that participants learn the associations between cues and weather patterns and can be completed with either declarative or nondeclarative memory strategies. This research is the first of its kind investigating the role that testing has on nondeclarative memory. The results from Experiment 1 showed an overall benefit of testing relative to studying for both declarative and nondeclarative memory, but not a testing effect. Experiment 2 expanded upon these findings and assessed whether participants were using declarative or nondeclarative memory during the word fragment completion task. A testing effect was shown for the participants using declarative and nondeclarative memory, overall, there was less forgetting for the word fragments that were tested on as compared to ones studied. Experiment 3 showed no decrease in performance over two days for those using nondeclarative memory regardless of if they learned from testing or studying. Taken together, these results show that testing can decrease forgetting for nondeclarative memory as well as declarative memory. This is important for multiple reasons, education being one of them. Learning in the classroom involves both declarative and nondeclarative memory, understanding the most effective way to improve nondeclarative memory has important implications for education. In addition, this research expands the large testing effect literature by expanding it to another memory system.