Doctor of Philosophy (PhD)
Faculty of Science
Jeffery A. Jones
William E. Hockley
Todd R. Ferretti
Memory is undoubtedly one of the most important processes of human cognition. A long line of research suggests that recognition relies on the assessment of two explicit memory phenomena: familiarity and recollection. Researchers who support the Dual Process Signal Detection (DPSD) model of recognition memory link the FN400 component (a negative ERP deflection peaking around 400 ms at frontal electrodes) with familiarity; however, it is currently unclear whether the FN400 reflects familiarity or implicit memory. Three event-related potentials (ERP) studies were conducted to determine whether implicit memory plays a role in setting up encoding strategies, and how these encoding strategies influence recognition.
Experiment 1 consisted of two phases; an encoding/study phase and recognition/test phase. During the encoding phase, participants viewed pictures of common objects and later during a recognition test phase they made remembered/not-remembered judgments about previously seen (old) pictures and new pictures. ERP analysis of the encoding phase compared subsequently-remembered and subsequently-not-remembered stimuli and revealed marginally significant subsequent memory effects for the FN400 and LPC components. Because participants first saw the pictures during the encoding phase, the FN400 effect during this phase suggested that it was driven by conceptual fluency. Additionally, the fluency ERP (a positive ERP deflection during the time window ~200 - 400 ms) during the encoding phase significantly distinguished subsequently-remembered stimuli from subsequently-not-remembered stimuli, indicating that processing during encoding determined the stimuli to-be-remembered during the recognition test. During the recognition test, the FN400 component correlated with the behavioral indicators of recollection and appeared to benefit from repetition.
Experiment 2 was similar to Experiment 1 except that participants saw meaningless novel stimuli (fractals). ERP results from recognition indicate that the FN400 effect did not capture repetition-based familiarity, however, the fluency ERP appeared to gain from the repetition of the stimuli. These results suggest that the FN400 potentials were driven by conceptual implicit memory during encoding, whereas during recognition, the behavioural indicators of recognition linked with the perceptual implicit memory, suggesting that explicit memory is not the only source of familiarity and the neural correlates of perceptual (fluency ERP) and conceptual (the FN400 component) implicit memory can influence decisions made by explicit memory.
Experiment 3 manipulated perceptual fluency, conceptual fluency, and repetition-driven familiarity. Participants viewed primed and unprimed, blurred and clear images of common objects that were presented once, twice or three times. Based on recognition performance, ERPs were back-sorted into their corresponding conditions. Fluency and FN400 components correlated with the behavioral indicators of recognition. Additionally, a conceptual implicit priming effect was significant over anterior and right frontal electrodes and perceptual implicit priming was significant at the occipital electrodes.
Conclusion: Collectively, the behavioural and ERP results add support the idea that the FN400 is “multiply determined” and may reflect familiarity (explicit memory driven) or conceptual fluency (implicit memory) depending on the task and stimulus, revealing that performance on recognition is not explicit memory driven. The Discrepancy Attribution Hypothesis may provide a better understanding of the heuristics of familiarity, however, further research is needed to better examine the processes that underlie recognition.
Gul, Asiya, "What Makes an Image Memorable? Effects of Encoding on the Mechanism of Recognition" (2020). Theses and Dissertations (Comprehensive). 2263.
Available for download on Wednesday, October 14, 2020