Search

How might faces we have learned be represented in our memory? Researchers believe that our memory for faces is based on building a robust averaged representation comprised of the stable aspects of the face (i.e., eyes, nose, mouth). However, anecdotal evidence suggests this one size fits all approach to face representations may not be correct. A new theory suggests our representation for faces is instead based on a dynamic weighting, wherein what is seen as most diagnostic during learning will be encoded to a greater extent than other features in the face. One factor that may be especially important for a weighted representation is the context in which a face is initially viewed. Dependent on the context of learning, certain features may appear more distinctive than others and therefore be deemed diagnostic and receive representational weight. The current study had participants learn four faces with one manipulated to appear distinctive in the experimental context by having a unique hair colour (Experiment 1), or eye colour (Experiment 2) compared to the other faces. Participants then completed a recognition task where the feature of interest (i.e., hair or eye colour) was either available or unavailable (i.e., bald and eye closed conditions) for recognition. Findings suggested recognition was disrupted when the diagnostic feature was unavailable compared to when that feature was available, across both distinctive and typical faces. Interestingly, Experiment 2 showed a distinctiveness performance advantage compared to Experiment 1, most likely because neighbouring features may be more diagnostic than others during recognition. In addition, further exploratory analysis showed the order of the test could further affect what was encoded.

Individuals with developmental dyslexia, considered as a group, perform poorly on tasks that involve phonological analysis, such as applying sight-sound rules to read new words, or analysing words into their component sounds (De Groot, Huettig, & Olivers, 2016; Temple & Marshall, 1983). However, dyslexia is also associated with other types of difficulties. For example, in some individuals, reading latencies increase disproportionately with the length of the word (De Luca, Barca, Burani, & Zoccolotti, 2008; Spinelli et al., 2005) suggesting they may have difficulties recognising familiar words as whole units (“whole word” processing). This thesis examined the relationship between the word length effect and overall reading proficiency in a diverse sample of 49 adolescents. We found that the length effect was a unique predictor of reading proficiency, even after factoring out variance in phonological skills (measured using a nonword reading task). We also tested the recent hypothesis that visual attention span - the number of letters a reader can capture in a single glance - is important for efficient whole word reading (Bosse, Tainturier, & Valdois, 2007). Contrary to this hypothesis, we found no association between the word length effect and scores on a standard measure of visual attention span (a partial report task). We also explored whether reading-delayed adolescents could benefit from an intervention targeting their specific cognitive profile. Five cases demonstrating a selective difficulty with either “phonological” or “whole word” skills completed two interventions. One targeted phonological skills: participants were trained to recognise and apply common sight-sound correspondences. The other targeted whole word skills: we reasoned that training participants to recognise commonly-occurring letter redundancies (e.g. ogue: rogue, synagogue, dialogue) could reduce the load on parallel letter processing. Only one of the five cases showed greater improvement in (un

A prominent debate in visual perception centers on the nature of mechanisms underlying face processing. One side of this debate argues that faces are processed by specialised mechanisms that are not involved in any form of object processing. By contrast, the other side argues that faces are processed by generic mechanisms common to all objects for which we are experts. To distinguish between these two hypotheses, I investigated whether participants with impaired face processing (developmental prosopagnosia) can acquire expertise with novel objects called greebles. To do so, I recruited 10 developmental prosopagnosics and 10 neurotypical control participants. All participants completed a standard training program for developing expertise with greebles, as well as two similar training programs with upright faces and inverted faces. Prosopagnosics were able to acquire expertise with greebles to the same extent as controls but were impaired when learning upright faces. These results demonstrate that deficits for face processing in individuals with prosopagnosia are dissociated from their ability to gain expertise with objects. Overall, the results support the hypothesis that face processing relies on specialised mechanisms, rather than generic expertise mechanisms. Despite their deficits, though, prosopagnosics still showed some evidence of learning with upright faces and showed better learning with upright faces than inverted faces. These findings suggest that prosopagnosics have face-specific mechanisms that are somewhat functional, and that training could be a useful rehabilitation tool in developmental prosopagnosia. Finally, I found substantial heterogeneity among the patterns of performance of the prosopagnosics, suggesting that further investigations into the subtypes of prosopagnosia are warranted.

Individuals with developmental prosopagnosia experience lifelong deficits recognising facial identity, but whether their ability to process facial expression is also impaired is unclear. Addressing this issue is key for understanding the core deficit in developmental prosopagnosia, and for advancing knowledge about the mechanisms and development of normal face processing. In this thesis, I report two online studies on facial expression processing with large samples of prosopagnosics. In Study 1, I compared facial expression and facial identity perception in 124 prosopagnosics and 133 controls. I used three perceptual tasks including simultaneous matching, sequential matching, and sorting. I also measured inversion effects to examine whether prosopagnosics rely on typical face mechanisms. Prosopagnosics showed subtle deficits with facial expression, but they performed worse with facial identity. Prosopagnosics also showed reduced inversion effects for facial identity but normal inversion effects for facial expression, suggesting they use atypical mechanisms for facial identity but normal mechanisms for facial expression. In Study 2, I extended the findings of Study 1 by assessing facial expression recognition in 78 prosopagnosics and 138 controls. I used four labelling tasks that varied on whether the facial expressions were basic (e.g., happy) or complex (e.g., elated), and whether they were displayed via static (i.e., images) or dynamic (i.e., video clips) stimuli. Prosopagnosics showed subtle deficits with basic expressions but performed normally with complex expressions. Further, prosopagnosics did not show reduced inversion effects for both types of expressions, suggesting they use similar recognition mechanisms as controls. Critically, the subtle expression deficits that prosopagnosics showed in both studies can be accounted for by autism traits, suggesting that expression deficits are not a feature of prosopagnosia per se. I also provide estimates of the preva

Face perception depends on a network of brain areas that selectively respond to faces over non-face stimuli. These face-selective areas are involved in different aspects of face perception, but what specific process is implemented in a particular region remains little understood. A candidate processisholistic face processing, namely the integration of visual information across the whole of an upright face. In this thesis, I report two experimentsthat examine whether the occipital face area (OFA), a face-selective region in the inferior occipital gyrus, performs holistic processing for categorising a stimulus as a face. Both experiments were conducted using online, repetitive transcranial magnetic stimulation (TMS) to disrupt activity in the brain while participants performed face perception tasks. Experiment 1 was a localiser in which participants completed two face identification tasks while receiving TMS at OFA or vertex. Participants’ accuracy decreased for one of the tasks as a result of OFA but not vertex stimulation. This result confirms that OFA could be localised and its activity disrupted. Experiment 2 was a test of holistic processing in which participants categorised ambiguous two-tone images as faces or non-faces while TMS was delivered to OFA or vertex. Participants’ accuracy and response times were unchanged as a result of either stimulation. This result suggests that the OFA is not engaged in holistic processing for categorising a stimulus as a face. Overall, the currentresults are more consistent with previous studies suggesting that OFA is involved in processing of local face features/details rather than the whole face.