Date of Award

Winter 12-15-2021

Author's School

Graduate School of Arts and Sciences

Author's Department


Degree Name

Doctor of Philosophy (PhD)

Degree Type



Forgetting reflects the decreased likelihood of memory retrieval over time. With a few exceptions, although the mean likelihood of retrieval (or retention) across experimental conditions may differ markedly, rates of forgetting across those conditions do not differ. Similarly, although groups (e.g., young and old adults) may differ in the amount retained at a given point in time, the rates of forgetting tend not to differ across groups. In contrast, recent work suggests that individual differences in rates of forgetting may emerge when more sensitive statistical analyses are used on person-level performance. Some person-level variables purported to influence forgetting rate include working memory capacity (WMC), fluid intelligence (gF), verbal fluency (VF), learning strategies, and learning rate. Synthesizing these findings has been challenging due to discrepancies in how long-term forgetting is measured; specifically, a common approach for estimating long-term forgetting rates (over days and weeks) is to measure recall across multiple short delays within a single experimental session (single-session forgetting; SSF). This shortcut approach may introduce confounding influences of short-term and working memory. Experiment 1 (N = 112) examined the influence of individuals’ WMC on their forgetting rates obtained from both the SSF procedure and a long-term multi-session procedure spanning one week. Greater WMC was related to better recall in both procedures but WMC was only related to forgetting in the SSF procedure. Additionally, forgetting rate estimates obtained from the two procedures were not significantly correlated. This outcome suggests that findings obtained from the single-day SSF procedure should not be interpreted as a measure of forgetting rate across longer intervals of hours and days. Experiment 2 (N = 245) extended these findings by including a measure of learning rate for the to-be-remembered material, and assessed the relative contributions of gF, VF, learning strategies, demographic variables (sex, age), and exploratory variables (pre-encoding sleep quality) to both learning rate and forgetting rate across one week. Both gF and VF were predictive of learning rate and learning rate in turn was the best predictor of forgetting rate. VF was related to forgetting rates but in the opposite hypothesized direction, such that higher VF was related to faster forgetting. An analysis of response times suggests those with higher VF were more susceptible to interference at later delays relative to those with lower VF. In regard to demographic variables, women learned more quickly than men and younger adults learned more quickly than older adults, and these differences resulted in better retention across one week but did not influence forgetting rates. Ultimately, the mechanisms that allow for quicker learning of information may not facilitate slower forgetting of that information over time.


English (en)

Chair and Committee

Kathleen B. McDermott

Committee Members

Mark A. McDaniel