top of page

TALKS
2024-2025

The group addresses questions, such as what’s the difference between a posed and a spontaneous expression? how fast can we perceive a face or an emotional expression?, what strategies do radiologists employ to detect breast cancer and is this skill trainable? how do clinical conditions, such as depression, autism, affect face recognition? To address these questions, researchers in the collaborative employ a variety of empirical techniques involving psychophysics, cognitive experiments, eye tracking, neural imaging (fMRI, EEG), and computer modeling.
 
Catch up on the 2024-25 DMC talk season here!
​
______________________________________________________________________________________
 
September 11th, 2024
​

Speaker: Isabelle Boutet 

Title: Does semantic information influence scanning and recognition of newly learned faces and names?

​

Abstract: I will start my presentation with a short review of my (unusual) academic trajectory and my research program. I will then present a study that examines the influence of person-specific semantic information on learning new faces. Background: As we become familiar with new people, we are repeatedly exposed to different perceptual images of their face which, over time, leads to storage of a perceptually invariant representation in memory (Bruce & Young, 1986; Burton, 2013, Johnston & Edmonds, 2009). During familiarisation with a new person, we also acquire semantic information about this person such as their name, occupation, preferences, etc. (Bruce & Young, 1986). Given that familiarization with new people involves both perceptual and semantic information, researchers have speculated that the presence of person-specific semantic details might affect encoding and recognition of newly learned faces. Accordingly, several studies have shown that associating faces with unique semantic information such as names or hobbies can improve their recognition (e.g., Gordon & Tanaka, 2011; Weise & Schweinberger, 2015; Schwartz & Yovel, 2016). In the present study, we examined how person-specific semantic information influences scanning, encoding and recognition of newly learned faces and names. Methods. During a Familiarization phase, participants were repeatedly exposed to a set of faces that varied in lighting across repetitions. Each face was associated with a name and half the faces were presented with a unique hobby (e.g., likes to read). This was followed by a Consolidation phase, where memory for the face-name associations was tested until participants reached a learning criterion. Finally, recognition and naming of the newly learned faces was tested immediately after the Consolidation phase, and after 1-week and 2-week delays. Faces were presented in a different viewing angle during recognition. Eye tracking technology was utilized to measure face scanning patterns throughout all phases of the experiment. Results. I will present preliminary results from the study at the DMC talk.  

​

Link to Zoom recording of Isabelle's Talk

​

​​

___________________________________________________________________________________

October 9th, 2024
​

Speaker: Joe DeGutis & Alison Campbell

Title: Perceptual, memory, and neural mechanisms underlying developmental prosopagnosia

​

Abstract: Developmental prosopagnosia, lifelong severe face recognition deficits, is a heterogeneous disorder and the cognitive and neural mechanisms are highly debated. In a relatively large DP sample, we first compare DP face recognition deficits to novel objects (Ziggerins). Next, using face matching tasks, we examine whether DP face perception deficits are continuous vs categorical and then examine the relative contribution of holistic and feature processing deficits to DPs' perceptual deficits. We next examine dysfunctional memory mechanisms in DPs, focusing on recollection vs familiarity as well as face associative memory (e.g., face-name recall and face-scene memory). Finally, using fMRI during rest and while watching short video clips of faces and objects, we examine DP vs control group differences in face network selectivity and resting-state connectivity within and beyond the face network. Together, these studies provide novel insights and open up several new DP future research directions.

​

Link to Zoom recording of Joe & Alison's Talk

___________________________________________________________________________________

​​

November 13th, 2024
​

Speaker: Geraldine Jeckeln 

Title: Unlocking the power of partnership: How humans and algorithms can work together to improve face recognition

​

Abstract: Human-machine “collaboration” is common in applied face recognition (e.g., forensic settings). Individual differences between people and machines, however, affect whether collaboration improves or degrades accuracy in any given case. We establish the circumstances under which combining human and machine face identification decisions improves accuracy. Using data from expert and non-expert face identifiers (Phillips et al., 2018; White et al., 2015), we simulated human-human and human-machine collaborations. The benefits of collaboration increased as the difference in baseline accuracy between collaborators decreased---following the Proximal Accuracy Rule (PAR). This rule predicted collaborative (fusion) benefit across a wide range of baseline accuracy, from people with no training to those with extensive training. Using the PAR, we established a critical fusion zone in which a human's less accurate judgment would benefit the machine via fusion. Human-machine fusion was beneficial up to a surprisingly large machine advantage. We implemented “intelligent fusion” by selecting individual people based their  potential to increase the accuracy of a high-performing machine. Intelligent human-machine collaboration was more accurate than the machine operating alone and more accurate than combining all human and machine judgments. The results demonstrate a meaningful role for both humans and machines in assuring accurate face identification. This study offers an evidence-based road map for the intelligent use of AI in face identification.

​

Link to Zoom recording of Gerie's Talk​

___________________________________________________________________________________

​

December 11th, 2024
​

Speaker: Karla Evans 

Title: Different Neuronal Signatures of Attentional Allocation for two Processes Supporting Visual Awareness of Complex Scenes​

​

Abstract: Visual awareness of complex scenes is supported by gist processing for rapid access to image identity and orienting in complex images, and a slower selective process providing individuation of objects within complex scenes. We tested the time course of interactions between the two, using the attentional blink (AB) paradigm, while recording event related potentials.

The focus was on T2-locked subcomponents P3a and P3b as well as multivariate pattern analysis of EEG data across the scalp when observers perform two different tasks during rapid visual serial perception (RVSP) of real scenes. The selective task required identification and localization of an object to a side in the scene and the gist task required categorization of the scene itself. In different blocks, observers are asked to perform two selective or two gist tasks or two different combinations of these two tasks, spaced with two different lags.

An attentional blink is observed when performing two selective tasks in close succession, with significant amplitude reduction in P3a and P3b. A different neuronal signature is observed with only in lag 1 dip in performance when two gist tasks are executed despite task difficulty being matched. Deeper and more prolonged attentional blink is observed when task switching is required relative to when performing two tasks of the same selectivity indicating interaction between the two types of processing. Based on neuronal markers as well as performance on the tasks this interaction seems to indicate these two processes happen in a serial fashion.

​

Link to Zoom recording of Karla's Talk​

___________________________________________________________________________________

 
Janurary 8th, 2025
​

Speaker: Michelle Greene 

Title: Coming soon!​

​

Abstract: Coming soon!

​

Link to Zoom recording of Michelle's Talk: Coming soon!

___________________________________________________________________________________

​​

February 12th, 2025
​

Speaker: Heida Sigurdardottir 

Title: Coming soon!​

​

Abstract: Coming soon!

​

Link to Zoom recording of Heida's Talk: Coming soon!

​

___________________________________________________________________________________

​

March 12th, 2025
​

Speaker: Clare Sutherland 

Title: Coming soon!​

​

Abstract: Coming soon!

​​

Link to Zoom recording of Clare's Talk: Coming soon!

___________________________________________________________________________________

​

June 11th, 2025
​

Speaker: Fang Jiang 

Title: Coming soon!​

​

Abstract: Coming soon!

​​

Link to Zoom recording of Fang's Talk: Coming soon!

​

___________________________________________________________________________________

​

July 9th, 2025
​

Speaker: Kami Koldewyn 

Title: Coming soon!​

​

Abstract: Coming soon!

​​

Link to Zoom recording of Kami's Talk: Coming soon!

​

___________________________________________________________________________________

bottom of page