Electronic Thesis and Dissertation Repository

Thesis Format

Integrated Article

Degree

Doctor of Philosophy

Program

Neuroscience

Supervisor

Köhler, Stefan

Abstract

Familiarity is a type of memory signal that can support recognition of prior occurrences without retrieval of associated contextual information. It is typically probed with respect to recent laboratory exposure in recognition-memory studies involving human participants. This line of work has revealed several neural correlates including event-related potentials (ERPs) and blood-oxygenation-level-dependent (BOLD) activity in several regions. However, few studies have examined familiarity accumulated outside of laboratory settings through lifetime experience. Hence, it is currently unclear whether similar neural correlates are involved. The fluency-attribution framework decomposes familiarity judgement into automatic and decision-related processes. Since recent and lifetime familiarity are phenomenologically and experimentally dissociable for meaningful stimuli, another question is whether certain neural correlates track both types of familiarity regardless of task relevance --- as a marker of automaticity, and whether they can be distinguished from other neural correlates that are decision-related. To answer these questions, I conducted an ERP and an fMRI study using a common paradigm in which degree of recent and lifetime familiarity could be compared in both task-relevant and -irrelevant conditions. In Chapter 2, I focused on ERP (FN400/N400 and LPC) responses and found that the LPC tracked both lifetime and recent familiarity when they were relevant to the task, while the N400 tracked both types of familiarity regardless of task relevance. The FN400 was sensitive only to task-relevant recent familiarity. In Chapter 3, I focused on BOLD activity in PrC and found that the left PrC tracked both types of familiarity regardless of task-relevance, while a set of frontoparietal regions tracked only task-relevant familiarity. In Chapter 4, I attempted to further delineate the decision-related neural correlates in familiarity judgement by combining the fMRI data collected in Chapter 3 with drift-diffusion modelling (DDM). A model comparison procedure showed that familiarity effects in medial frontal regions were most strongly involved in decision-making, followed by PrC, then by medial parietal regions. Overall, these results revealed temporally (ERP) and spatially (fMRI) distinct neural correlates corresponding to the automatic and decision-related processes in both recent and lifetime familiarity judgement. Furthermore, a hierarchy exists among the decision-related neural correlates.

Summary for Lay Audience

We can make memory judgements on different time scales. For example, we can judge whether we have seen an apple during our last visit to the supermarket, and we can judge how familiar we are with apples in general. These two types of judgement rely on different information. Scientific studies of memory have mostly focused on the first kind. This line of research has revealed several neural correlates including electrical potentials in the brain recorded with electroencephalogram (EEG) and changes in blood oxygenation level related to neural activity recorded with functional magnetic resonance imaging (fMRI). However, two questions are unclear. First, do these neural correlates also track information supporting the second kind of memory (i.e. how familiar we are with apples in general)? And second, given that we can make both types of judgement when presented with an apple, if one is asked to make one versus the other type of judgement, do these neural correlates behave differently depending on which judgement is being made? In Chapter 2, I showed that for three brain potentials recorded with EEG, one tracked both types of memory information regardless of which judgement was required, another tracked only the information that was relevant to the judgement, and a third selectively tracked information supporting the first kind of memory judgement when a judgement of that kind was required. In Chapter 3, I showed that fMRI activity in a region called perirhinal cortex (PrC) tracked memory information supporting both types of judgement, regardless of which judgement was required. fMRI activity in other regions tracked only the information that was relevant to the judgement being made. In Chapter 4, I combined the fMRI data with a mathematical model and showed that activity in PrC and other regions contributed differently to memory decision. Overall, these results revealed temporally (EEG) and spatially (fMRI) distinct markers corresponding to automatic (i.e. present regardless of if the information is relevant to the judgement) and decision-related (i.e. only present when the information is relevant) processes in both types of familiarity judgement. Furthermore, a hierarchy exists among the decision-related neural markers.

Creative Commons License

Creative Commons Attribution 4.0 License
This work is licensed under a Creative Commons Attribution 4.0 License.

Download

Share

COinS