Investigating the Neural Bases of Autobiographical Memory Retrieval with FMRI and Overt Speech Restricted; Files Only

Ferris, Charles (Summer 2021)

Permanent URL:


Autobiographical memory (AM), episodic memory from one’s past, is critically important to day-to-day life. AM retrieval involves both episodic (events) and semantic memory (facts) and requires the orchestration of dynamic networks of brain regions. AM retrieval’s protracted duration is advantageous for fMRI studies. Several fMRI studies of the temporal dynamics of AM retrieval have compared initial memory search processes to later elaboration processes, during which rich episodic details are retrieved. Nearly all fMRI studies of AM retrieval utilize silent retrieval to avoid speech-related motion artifacts. Recent technical advances have begun to counteract motion related problems, allowing AM content and temporal dynamics to be measured using overt retrieval. Herein, two experiments addressed three key aims. The first aim was to compare the neural correlates of narrated and silently retrieved AMs, bridging between the silent AM retrieval literature and new studies using overt retrieval. The second aim was to compare AM search to elaboration to determine how retrieval modality may alter AM temporal dynamics. The third aim was to utilize moment-to-moment overt memory content to identify patterns of activation supporting fine-grained retrieval of memory detail. In both experiments, adults generated unrehearsed AMs during fMRI scanning (half silent, half overtly narrated). Experiment 1 used a univariate fMRI activation approach to identify brain regions associated with the three aims outlined above. Experiment 2 used a combination of functional connectivity approaches to address the same broad experimental questions. Consistent with predictions, Experiment 1 revealed that both narrated and silent AM retrieval activated key AM retrieval regions. Further, narrated retrieval relative to silent retrieval activated speech processing networks as well as memory and cognitive control regions (posterior hippocampus, DLPFC). Silent retrieval relative to narrated retrieval activated regions associated with recollection (anterior hippocampus, angular gyrus, PCC). Contrasting search and elaboration phases, we replicated previous findings and detected a novel whole-brain AM connectivity pattern. Finally, when participants spontaneously retrieved object and scene words they activated corresponding content-specific cortex (LOC, RSC), consistent with theoretical models of neural reactivation. These findings provide new insights into cognitive processes, content-specific activation, and the role of retrieval modality during AM retrieval.

Table of Contents

Table of Contents

1.     General Introduction. 1

1.1.      Autobiographical Memory. 2

1.1.1.        What is Autobiographical Memory?. 2

1.1.2.        Phases of AM Retrieval 2

1.1.3.        Duration of AM Retrieval 4

1.2.      Functional Neuroimaging of AM Retrieval 4

1.2.1.        Temporal Dynamics of AM Retrieval 7

1.2.2.        Narrated AM Retrieval 9

1.2.3.        Content-Specific Reactivation. 12

1.3.      Summary. 13

2.     Experiment 1, Dynamic activation of content specific cortex during silent and overtly narrated autobiographical memory retrieval 15

2.1.      Abstract. 15

2.2.      Introduction. 16

2.3.      Methods. 23

2.3.1.        Participants. 23

2.3.2.        Procedure. 24

2.3.3.        Search and Elaboration Phases. 28

2.3.4.        fMRI Scanning Methods. 30

2.3.5.        Data Analysis. 31

2.3.6.        Activation Contrasts. 32

2.3.7.        Analysis of Reliving Ratings. 33

2.3.8.        Transcription and Content Analyses. 35

2.3.9.        Analysis of Memory Content. 37

2.4.      Results. 38

2.4.1.        Behavioral 38

2.5.      fMRI Results. 41

2.5.1.        Silent Retrieval 41

2.5.2.        Narrated AM Retrieval 46

2.5.3.        Silent AM Retrieval > Narrated AM Retrieval 52

2.5.4.        Narrated AM Retrieval > Silent AM Retrieval 55

2.5.5.        Reliving Analysis. 58

2.5.6.        Content Analysis. 59

2.6.      Discussion. 60

2.6.1.        Silent AM retrieval activates core autobiographical memory regions. 62

2.6.2.        Narrated AM retrieval activates both memory regions and speech regions. 63

2.6.3.        Differences between silent and overtly narrated retrieval activations. 65

2.6.4.        Retrieval of object and scene content words differentially activate the lateral occipital complex vs. retrosplenial cortex. 67

2.6.5.        Limitations of the present study. 67

2.6.6.        Advantages and limitations of the overtly narrated retrieval method. 68

2.6.7.        Future Directions. 69

2.6.8.        Conclusion. 70

3.     Experiment 2: Network features and dynamic connectivity of overtly narrated autobiographical memory retrieval 71

3.1.      Abstract. 71

3.2.      Introduction. 72

3.3.      Methods. 76

3.3.1.        Participants. 76

3.3.2.        Procedure. 77

3.3.3.        Search and Elaboration Phases. 80

3.3.4.        fMRI Scanning Methods. 83

3.3.5.        Data Analysis. 84

3.3.6.        Regions of Interest. 85

3.3.7.        Network Based Statistic. 87

3.3.8.        Transcription and Content Analyses. 88

3.3.9.        Generalized Psychophysiological Interaction. 89

3.3.10.     Graph Theory. 90

3.4.      Results. 92

3.4.1.        Behavioral 92

3.4.2.        Silent AM Retrieval 95

3.4.3.        Narrated Retrieval 101

3.4.4.        Silent AM Retrieval vs. Narrated AM Retrieval 109

3.4.5.        Content Analysis. 113

3.5.      Discussion. 118

3.5.1.        The AM core network is densely interconnected to the right and left anterior hippocampus, mPFC, PCC, and left angular gyrus. 118

3.5.2.        Whole brain connectivity during silent retrieval shows stronger occipito-parietal connectivity during search, and stronger MTL connectivity to frontal and occipital cortex during elaboration. 119

3.5.3.        Narrated Search elicits temporal and occipital connectivity, narrated elaboration elicits MTL and whole brain connectivity. 120

3.5.4.        Silent AM retrieval connects midline AM core hubs; Narrated AM retrieval connects parietal regions 121

3.5.5.        Content retrieval elicits connectivity from the anterior hippocampus to brain regions supporting corresponding content processing. 122

3.5.6.        Limitations and Future Directions. 123

3.5.7.        Conclusion. 124

4.     General Discussion and Conclusions. 126

4.1.      Silent and Narrated AM Retrieval 126

4.2.      Temporal Dynamics of AM Retrieval 128

4.3.      AM Content. 130

4.4.      Future Directions. 131

4.5.      Conclusion. 133

5.     References. 134

About this Dissertation

Rights statement
  • Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.
Subfield / Discipline
  • English
Research Field
Committee Chair / Thesis Advisor
Committee Members
Last modified Preview image embargoed

Primary PDF

Supplemental Files