Theses and Dissertations

Issuing Body

Mississippi State University


Eakin, Deborah K.

Committee Member

Pratte, Michael S.

Committee Member

Moss, Jarrod

Committee Member

Winer, Eric Samuel

Date of Degree


Document Type

Dissertation - Open Access


Cognitive Science

Degree Name

Doctor of Philosophy (Ph.D)


College of Arts and Sciences


Department of Psychology


False recollection refers to the retrieval of contextual information associated with an event that has not occurred. For instance, during a recognition task, one might identify a nonstudied word presented at test as old because she remembers the font color of the word during study. Although instances such as this are rare and typically occur at a varying rate of 0-5%, current models of recognition such as the Complementary Learning Systems (CLS) model and the Dual-Process Signal-Detection (DPSD) model do not contain a mechanism to account for their occurrence. Although both the CLS and DPSD models have support from studies demonstrating functional dissociations, neurophysiological dissociations, and behavioral findings of process dissociation, their ability to explain false memories has been more elusive; neither theory specifically addresses false recollection. Instead, such models have ignored false recollection as inconsequential noise in the data. The purpose of this dissertation was to determine whether the false recognition effect obtained by the Payne-Eakin paradigm was due to false recollection or familiarity. The Payne-Eakin paradigm is based on the PIER2 model, which theorizes that targets implicitly activated during study lead to the falser recognition of a false-target pair. Using a modified version of the Payne-Eakin paradigm, we investigated the nature of the false recognition effect using a priori behavioral analyses and statistical modeling. The findings of this dissertation provide a step toward a more solid understanding of the cognitive mechanisms involved in the recognition of nonstudied items. This dissertation demonstrates that modeling false recollection is possible. The results of this dissertation suggest that, because current models of recognition do not provide a mechanism to account for false recollection, our understanding of recognition is not fully understood. The results highlight that the current understanding of how false recollection contributes to recognition performance is an area in need of further development.