Judging a person as a friend or foe, a mushroom as edible or poisonous, or a sound as an l or r are examples of categorization problems. Because people never encounter the same stimulus twice, they must develop categorization schemes that capture the useful regularities in their environment. Key research challenges include how humans acquire and represent categories. This project tackles the broader challenge of elucidating the nature of the learning system or systems, these systems' neural underpinnings, how such systems develop, how they differ across species, and how they interact. To answer these fundamental questions, a space of category learning models is defined to allow for formal evaluation of the theories these models encode. Although no one study can explicate the nature, developmental trajectory, evolution, and neural underpinnings of all category learning mechanisms, the results from numerous studies can when coupled with powerful analysis techniques. By defining a space of models, data from numerous studies (developmental - Project 1, as well as comparative and neuroscientific - Project 2) can be jointly evaluated to determine the most likely theories given the data. This approach incorporates key task variables, such as proposed relationships between formal mechanisms and brain regions, and how various system capacities and biases can vary across development and evolution. Thus, the developed theories (in the form of formal models) not only specify computational mechanisms, but how these mechanisms change over development, vary across species, their neural underpinnings, how genetic variations shape individual differences, and how task variables (e.g., secondary task load) affect their operation. Bayesian Model Selection (BMS) procedures will evaluate candidate models on a vast array of data collected within Projects 1 and 2 to determine which models are most likely to be valid (i.e., generalize to novel studies). Preliminary results will help guide efforts in Projects 1 and 2 to determine the most theoretically informative study designs. Model fits may prove useful for gauging what constitutes normal development and for directing interventions for populations suffering from disease or other difficulties.
| O'Leary, Allison P; Sloutsky, Vladimir M (2018) Components of metacognition can function independently across development. Dev Psychol : |
| Darby, Kevin P; Castro, Leyre; Wasserman, Edward A et al. (2018) Cognitive flexibility and memory in pigeons, human children, and adults. Cognition 177:30-40 |
| Ahlheim, Christiane; Love, Bradley C (2018) Estimating the functional dimensionality of neural representations. Neuroimage 179:51-62 |
| Castro, Leyre; Wasserman, Edward A; Lauffer, Marisol (2018) Unsupervised learning of complex associations in an animal model. Cognition 173:28-33 |
| Kim, Jangjin; Castro, Leyre; Wasserman, Edward A et al. (2018) Dorsal hippocampus is necessary for visual categorization in rats. Hippocampus 28:392-405 |
| Bobadilla-Suarez, Sebastian; Love, Bradley C (2018) Fast or frugal, but not both: Decision heuristics under time pressure. J Exp Psychol Learn Mem Cogn 44:24-33 |
| Mack, Michael L; Love, Bradley C; Preston, Alison R (2018) Building concepts one episode at a time: The hippocampus and concept formation. Neurosci Lett 680:31-38 |
| Guest, Olivia; Love, Bradley C (2017) What the success of brain imaging implies about the neural code. Elife 6: |
| Sloutsky, Vladimir M; Yim, Hyungwook; Yao, Xin et al. (2017) An associative account of the development of word learning. Cogn Psychol 97:1-30 |
| Palmeri, Thomas J; Love, Bradley C; Turner, Brandon M (2017) Model-based cognitive neuroscience. J Math Psychol 76:59-64 |
Showing the most recent 10 out of 16 publications
