Degree

Doctor of Philosophy (PhD)

Department

Psychology

Document Type

Dissertation

Abstract

Abstract

Researchers in the eyewitness literature have extensively used retrospective confidence judgments as opposed to prospective confidence judgments. This choice is because historically, prospective confidence judgments have shown a weak to no relationship with accuracy. Recently two studies have been published showing the conditions where prospective confidence judgments have a strong relationship with accuracy. This relationship emerged by manipulating encoding conditions and utilizing more modern analytical methods. The researchers showed that prospective judgments’ relationship with accuracy relies on the variability in encoding conditions, suggesting that prior research was too homogenous. The goal of the current studies was to better understand the impact of methodological choices in the eyewitness literature and reconcile the differences relative to findings in metacognition. Experiment 1 extended the work of Molinaro et al. (2021) through introducing more conditions to investigate the impact of the memory inventory and the potential impact of anchoring effects on calibration curves. Experiment 2 was designed to better reconcile the results of McKinley et al. (2023) by utilizing a similar retention interval for delayed judgments of learning (JOLs) found in metacognition. The results of my study suggest that pre-identification confidence judgments are predictive of future accuracy, without compromising the confidence-accuracy relationship for post-identification judgments. Furthermore, reducing the retention interval utilized in McKinley et al., (2023) improved the confidence-accuracy relationship for delayed pre-identification judgments, suggesting that pre-identification judgments, regardless of delay, in my study were made from long-term memory.

Date

1-9-2025

Committee Chair

Emily Elliott

Download

Available for download on Sunday, January 09, 2028

Share

COinS