Electronic Thesis and Dissertation Repository

Thesis Format

Integrated Article


Doctor of Philosophy




John Paul Minda


Objectives. Individuals can differ in their strategic approach in learning the same categorization task, researchers have sought to study what specific stable individual differences traits can help explain these differences. This dissertation first surveyed extant literature on the impact of trait differences on category learning then examined the effect of temperament traits on these dependent variables. Chapter 2 (scoping review): This scoping review synthesized the past literature that examined the relationship between sources of stable individual differences and category learning performance and strategy use outcomes. Five database platforms were searched to identify relevant articles, cross-referencing was also performed. Sixty-nine studies met inclusion criteria with 3 major sources of individual differences identified: (1) developmental, (2) aging, (3) working memory. The results of this scoping review suggest that (1) children tend to show both performance and task-appropriate strategy-use disadvantage in both rule-based and similarity-based category learning tasks compared to young adults. (2) Older adults also showed a performance disadvantage, but results were less consistent with regards to whether they used different strategies than young adults. (3) Working memory was associated with better performance on both types of tasks, but it was not associated with strategy choice on rule-based tasks, and results were inconsistent in terms of strategy choice on similarity-based tasks. Chapter 3 (two studies): In two studies, I examined affective temperament traits to see whether the tendency to experience negative and positive affect is predictive of category learning performance and strategy use. Temperamental effortful control and working memory were measured as covariates. There were minimal effects of affective temperament traits and temperamental effortful control may be negatively associated with learning on both types of category learning. Working memory may be positively associated with learning on both types of category learning. However, these findings were not consistent across studies. The results may either reflect a lack of relationship or low data quality due to the pandemic. Conclusions: Neither previous studies nor the present dissertation provided a firm answer to the mystery behind individual differences in category learning strategy use. Future research should replicate the studies in Chapter 3 of this dissertation in the laboratory to see whether temperament effects would emerge.

Summary for Lay Audience

People use a variety of strategies when learning the same category, despite the fact that only one strategy yields high performance. This curious phenomenon has spurred the interest of researchers to want to understand if certain characteristics can predict the specific strategy a person tends to use. This dissertation explores the effect of temperament traits as potential predictive characteristics through two projects. In the first project, an overview of existing research studies that examined the impact of different characteristics on category learning was provided. Three characteristics were particularly studied: brain development of children, healthy aging of adults, and the ability to hold and manipulate information in the brain. I found that children tend to use a strategy that gives them lower performance compared to young adults. Older adults also suffer from lower performance than young adults, but the two groups may not differ in their choice of strategy. Higher ability to hold and manipulate information in the brain has been associated with better performance but not consistently with the better strategy. In the second project, participants self-rated on a questionnaire which measures their temperament traits and they also completed category learning tasks. Three temperament traits were of interest: tendency to experience positive emotions, tendency to experience negative emotions and the ability to focus and control attention. I also measured the ability to hold and manipulate information since it was found to lead to better performance. I did not find a consistent relationship between any temperament traits or the ability to hold and manipulate information with category learning. These results suggest either an actual lack of relationship or that data collected during the pandemic may have had low quality and hid the real effect. More research is needed to further study the effects of temperament traits on predicting category learning performance and strategy.