About
Two people can watch the same scene and come away with different memories of it. The difference begins before memory does, in what each person selected, what they ignored, and the state they were in while looking. Those selections are not made fresh each time. They are tuned by everything the person has seen before. I study this cycle, in which experience shapes attention and attention decides what becomes experience.
Much of my current work is on the first half. Ignoring looks like doing nothing, but it is something people learn to do, and I want to know what exactly is learned. Does suppressing a distractor require having first been captured by it? Is the learning tied to a particular feature and context, or is it closer to a skill that transfers?
The second half asks how attention writes memory. What we select determines what is encoded at all, and what we prioritize determines how richly. Because attention is never constant, moment-to-moment fluctuations in attentional state help explain why some experiences last and others fade. I approach this through the evidence people sample with their eyes, the attentional state they are in at encoding, and the boundaries they perceive in ongoing experience.
I came to these questions through memory. My undergraduate dissertation at the University of Edinburgh, with Prof. Andrew Lawrence, asked whether cultural differences in thinking style change how people segment and remember a film. It left me thinking that to understand what people remember, I first had to understand what they attend to. At Wake Forest University I am advised by Dr. Brad Stilwell and collaborate with Dr. Anthony Sali; the projects below describe my collaborations at Duke and UCLA. I use behavioral and eye-tracking experiments, fMRI, and statistical and computational modeling.
Current research
Learned distractor suppression and attentional templates
With Dr. Brad Stilwell · Wake Forest University
How do people learn to ignore distracting information, and does that learning require first looking at the distractor? Using visual search with eye tracking, I test whether learned suppression depends on initial oculomotor capture and how far it generalizes across visual contexts. So far, we have found that oculomotor suppression of a learned distractor emerges whether or not the distractor was first fixated, which challenges the view that suppression has to be built on capture. Ongoing work uses continuous report and Bayesian mixture modeling to ask how negative attentional templates are held in working memory after guiding search.
Attentional states as context for memory
With Dr. Anthony Sali · Wake Forest University
Can an internal attentional state act like a context cue for long-term memory? With attentional shift and hold paradigms, we examine whether matching the attentional state at encoding and retrieval produces context-dependent memory and priming effects. Our results to date show that repetition priming is larger when an item is re-encountered in the same attentional context in which it was first seen, but only for items encoded before performance had stabilized, which suggests that an attentional state can serve as an internal retrieval context.
Gaze dynamics and memory for ambiguous images
With Prof. Roberto Cabeza and Dr. Maxi Becker · Duke University
Recognizing an ambiguous image is something the eyes work at, not something that simply happens. I ask whether the way people search a two-tone (Mooney) image for diagnostic evidence predicts when they recognize it and whether they remember it later, combining human-annotated diagnostic regions with fixation dynamics, time-resolved gaze similarity, and image-level measures. Early sampling of diagnostic regions appears to predict recognition, whereas late convergence of gaze tracks the decision to respond rather than its accuracy; early sampling has no direct effect on memory five days later, and its link to memory runs through successful recognition.
Posterior medial network and event transitions
With Dr. Maureen Ritchey and Dr. Valentina Krenz · University of California, Los Angeles
Memory for an experience depends not only on what happens during it but on what the brain does at the moment it ends. This project asks whether engagement of the posterior medial network at event boundaries during encoding predicts how well multi-feature events are later remembered; I contribute fMRI quality control and univariate and multivariate subsequent-memory analyses.