Abstract

For more than two decades, a network of face-selective brain regions has been identified as the core system for face processing, including occipital face area (OFA), fusiform face area (FFA), and posterior region of superior temporal sulcus (pSTS). Moreover, recent studies have suggested that the ventral route of face processing and memory should end at the anterior temporal lobes (i.e., vATLs), which may play an important role bridging face perception and face memory. It is not entirely clear, however, the extent to which neural activities in these face-selective regions can effectively predict behavioral performance on tasks that are frequently used to investigate face processing and face memory test that requires recognition beyond variation in pose and lighting, especially when non-Caucasian East Asian faces are involved. To address these questions, we first identified during a functional scan the core face network by asking participants to perform a one-back task, while viewing either static images or dynamic videos. Dynamic localizers were effective in identifying regions of interest (ROIs) in the core face-processing system. We then correlated the brain activities of core ROIs with performances on face-processing tasks (component, configural, and composite) and face memory test (Taiwanese Face Memory Test, TFMT) and found evidence for limited predictability. We next adopted an multi-voxel pattern analysis (MVPA) approach to further explore the predictability of face-selective brain regions on TFMT performance and found evidence suggesting that a basic visual processing area such as calcarine and an area for structural face processing such as OFA may play an even greater role in memorizing faces. Implications regarding how differences in processing demands between behavioral and neuroimaging tasks and cultural specificity in face-processing and memory strategies among participants may have contributed to the findings reported here are discussed.

Highlights

  • Reviewed by: José Salvador Blasco Magraner, University of Valencia, Spain Andrea Lami, Catholic University of Valencia San Vicente Mártir, Spain

  • We adopted an multi-voxel pattern analysis (MVPA) approach to further explore the predictability of face-selective brain regions on Taiwanese Face Memory Test (TFMT) performance and found evidence suggesting that a basic visual processing area such as calcarine and an area for structural face processing such as occipital face area (OFA) may play an even greater role in memorizing faces

  • We reported the results of prediction accuracy based on the binary support vector machine (SVM) classification on the TFMT, where prediction accuracy was defined as an unweighted average of hit and correct rejection on classifying TFMT performance, with the evaluation scheme of leaving-one-participant-out crossvalidation

Read more

Summary

Introduction

Reviewed by: José Salvador Blasco Magraner, University of Valencia, Spain Andrea Lami, Catholic University of Valencia San Vicente Mártir, Spain. Recent studies have suggested that the ventral route of face processing and memory should end at the anterior temporal lobes (i.e., vATLs), which may play an important role bridging face perception and face memory It is not entirely clear, the extent to which neural activities in these face-selective regions can effectively predict behavioral performance on tasks that are frequently used to investigate face processing and face memory test that requires recognition beyond variation in pose and lighting, especially when non-Caucasian East Asian faces are involved. To address these questions, we first identified during a functional scan the core face network by asking participants to perform a one-back task, while viewing either static images or dynamic videos. While not attempting to resolve the controversy here, we are more inclined to take the view advocated by Rossion (2008, 2009, 2013), McKone (2010; see McKone and Yovel, 2009), Tanaka and Gordon (2011, see Tanaka and Farah, 1993, 2003), and Richler and Gauthier (2013, 2014), where holistic processing entails integrated processing of all features, nameable and non-nameable, as well as a variety of detailed metric relationships between and among them (McKone and Yovel, 2009), which results in a single face representation created by simultaneous integration and combination (see Kimchi, 1992, for an in-depth discussion on distinguishing between holistic and global properties in face or other hierarchically structured visual patterns)

Objectives
Methods
Results
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call