Children with fetal alcohol spectrum disorders (FASD) have deficits across many cognitive, behaviour and social domains. However, despite social difficulty being proposed as a main deficit following prenatal alcohol exposure, the nature of their deficient social behaviour is largely unknown. One process that may underlie difficulties in social functioning is poor social cognition, which refers to one’s understanding of the self, others and social world. The primary goal of this dissertation was to determine whether social cognitive deficits represent a core disability underlying the socio-behavioural problems of FASD using a bottom-up approach. The first level of this approach is represented by face processing. Global and independent face feature processing was compared between FASD and normal controls (NC) using experimental and clinical measures. Eye gaze processing was investigated next using experimental and clinical tasks. At the highest level of the bottom-up approach, social perspective taking, including theory of mind and empathy were examined, along with how these abilities related to parent-rated behaviour. Finally, the lowest level consisted of specific aspects of the social neural network. White matter in three limbic pathways was investigated using diffusion tensor imaging (DTI). Results generally supported the bottom-up approach of social cognition in FASD. These children showed impaired processing of face features, when matching mouth shapes and partially occluded identity, compared with NC. The FASD group was slower to process gaze and arrow cues, suggesting impaired attention shifting. Children with FASD also showed impairments in social perspective taking, including understanding false beliefs and empathy, and these impairments were related to parent-rated attention and social problems, and autistic-like traits. Deficits in theory of mind got worse with age in FASD and empathy showed distinct sex-related differences. Although no group differences were observed on DTI indices, groups did show different age-related changes in white matter. In conclusion, deficits at each level of the current bottom-up approach may underlie the social impairments in FASD and may contribute to their broader social behavioural phenotype. The results from this dissertation have potential to inform clinical practice and lead to more effective diagnostic and treatment approaches in FASD.

Paediatric posterior fossa tumors are often effectively controlled with a combination of radiation, chemotherapy and surgery. However, therapeutic craniospinal radiation has been associated with widespread cognitive late effects. Working memory is one such cognitive ability that has yet to be fully examined in this clinical population. Bilateral tracts connecting the cerebellum with the DLPFC were delineated using DTI tractography in all participants, replicating the cerebrocerebellar pathway outlined in an animal model. There were observable differences in white matter integrity (quantified by DTI measures of anisotropy, and mean, axial, and radial diffusivity) of the cerebellum-DLPFC pathway in patients versus controls. Additionally, working memory deficits that were found in patients were correlated with DTI indices pertaining to the cerebellum-DLPFC pathway. Therefore, this thesis is the first to explore the possible relations between white matter integrity of this pathway following treatment for paediatric posterior fossa tumors and working memory function.

Context: The study of expertise in medical education has tended to follow the traditions of describing either the analytic processes or the non-analytic resources that experts acquire with experience. We argue that a critical function of expertise is the ability to transition from the automatic mode to the more effortful mode when required – a transition referred to as ‘slowing down when you should’. Objectives: To explore the phenomenon of ‘slowing down when you should’ in operative surgical practice and its role in intra-operative surgical judgment, and to develop conceptual models of the factors involved in the display of this transition in surgical operative practice. Design: In Phase 1A, 28 surgeons were interviewed about their views of surgical judgment in general and their perceptions of the role of this phenomenon in operative judgment. In Phase 1B, a subset of surgeons from Phase 1A was re-interviewed to explore their perceptions of automaticity in operative practice. In Phase 2, observational sessions (and brief interviews) were conducted of surgeons in the operating room to explore the nature of this phenomenon in its natural environment. Results: The surgeons in this study recognized the phenomenon of ‘slowing down’ in their operative practice and acknowledged its link to surgical judgment. Two main initiators were described and observed: proactively planned ‘slowing down’ moments occurring intra-operatively initiated by critical events anticipated pre-operatively and situationally responsive ‘slowing down’ moments initiated by emergent cues intra-operatively. Numerous influences of this transition were uncovered. A control dynamic emerged as surgeon’s negotiated ‘slowing down’ moments through trainees in their supervisory academic practice. Numerous manifestations of this phenomenon were observed in the operating room and considered using a cognitive psychology attention capacity model. Conclusions: This study offers a conceptual framework for understanding the role of ‘slowing down when you should’ in operative surgical practice, providing a vocabulary that will allow more explicit consideration of what contributes to surgical expertise. Consideration of this framework with its consequent ability to make surgical practices more explicit has implications for self-regulation in practice, surgical error, and surgical training.

In the study of language production in aging, an important question relates to the relationship between lexical retrieval and syntactic production. Studies have reported changes in syntactic production across the lifespan, but their underlying cause remains unclear. In younger adults, it has been suggested that lexical factors, such as an item‟s semantic or phonological representation influence syntactic production; however, the full nature of this influence remains unclear. Studies investigating the type of sentence produced have found semantic facilitation and phonological interference (e.g., Bock, 1986, 1987), but studies investigating response time (e.g., Meyer, 1996) have found the opposite effects. This investigation sought to examine the influence of lexical level information on sentence production in younger and older adults. This was accomplished by concurrently examining reaction time and sentence type effects. In Experiment 1, 61 adults (mean age: 21.8 years) were presented with pictures and distractor words (unrelated, or semantically or phonologically related). Three stimulus onset asynchronies (SOA) were used (-1000ms, -150ms and 150ms). Participants were required to describe each picture. Using an analysis of variance, response time was compared across the different conditions and using generalized estimating equations, the type of sentence produced and the position of the primed word were compared. In Experiment 2, phonological distractors were excluded, and one SOA (-150ms) was used. Testing involved 83 younger adults (mean age: 22.9 years) and 56 older adults (mean age: 74.7 years). In Experiment 1, semantic distractors resulted in related nouns being produced more often in the subject position. This effect was observed in the analysis of the position of the target noun, but not in the analysis of the type of sentence produced. There were no effects of phonological distractors. In Experiment 2, semantic distractors influenced the type of sentence produced for both age groups. The groups differed only in error production. No reaction time effects were observed in either experiment. This investigation successfully demonstrated an influence of lexical level information on the syntactic productions of younger and older adults. The two groups were similar in their productions, suggesting that aspects of syntactic production are preserved in older adults.

The efficiency of human visual information processing is supported by numerous attentional resources. These resources ensure that behaviourally relevant information within visual scenes is selected for detailed processing, while behaviourally irrelevant information is ignored. One of these attentional resources—reflexive visuospatial attention—operates by prioritising locations in visual space in response to the appearance of salient stimuli. The purpose of the present dissertation was to examine how this type of attention contributes to the efficiency of visual processing by asking: How is processing altered for information presented at the location of attention? To develop some initial evidence of the stage of processing affected by reflexive visuospatial attention, Chapters 1 to 6 assessed whether this attentional resource is related to four other stimulus-driven effects that are each associated with a specific stage of visual processing: identity processing, object filtering, visual working memory (VWM), and response generation. Based on the observation that only the stimulus-driven effects on VWM are related to reflexive visuospatial attention (i.e., only those effects were contingent on attentional control settings), a VWM model of reflexive visuospatial attention was proposed in Chapter 7, and tested in Chapters 8 to 11. According to this model, reflexive visuospatial attention alters visual processing by triggering VWM to update. Thus, the effect of reflexive visuospatial attention is to speed the encoding of attended information into VWM. As a result, this information is more likely than unattended information to bias our behaviour, in particular those behaviours that depend on VWM. Further, by biasing VWM, reflexive visuospatial attention can interact with other attentional resources that have also been associated with VWM. In this way, these attentional resources can coordinate in optimising the process of selection, thus, contributing to the efficiency of the human visual system.

Despite the ubiquity of music and its importance to one’s identity, there has been limited research on individual differences in music preferences. The aim of this dissertation was to examine how music preferences vary across individual differences in stable traits (e.g., gender, personality) and states (e.g., mood) over three studies. The focus of Study 1 was on the influence of the listener’s mood on emotion-based music preferences. The typical preference for happy- over sad-sounding music was found to be mood-dependent, evident after happy and neutral but not sad mood inductions. When the music was emotionally ambiguous (i.e., with cues to both happiness and sadness), happy listeners liked the music more than other listeners, whereas sad listeners perceived it to be more sad-sounding. Study 2 examined how emotion-based music preferences develop. Adults and children 5, 8, and 11 years of age listened to short pieces expressing emotions that varied in arousal and valence: happiness (high, positive), scariness (high, negative), peacefulness (low, positive), or sadness (low, negative). Adults preferred pieces with a positive valence (happy and peaceful), whereas children preferred excerpts depicting high-arousal emotions (happy and scary). Identification accuracy was predictive of a positive-valence bias among 5- and 8-year-olds. A number of other findings related to accuracy and gender differences were also evident. Study 3 examined whether personality differences affect the influence of exposure on liking. Undergraduates completed the Big Five Inventory and provided liking ratings for novel music excerpts as well as for excerpts they heard 2, 8, or 32 times. Higher scores on Openness-to-Experience were related to greater liking for novel excerpts and more rapid satiation. In sum, these three studies highlight that music preferences vary as a function of individual differences and the listening context.

This thesis addresses the computation and organization of conceptual knowledge. Specifically, it focuses on the recruitment of concrete knowledge during single word reading using behavioural and electrophysiological methodologies. Chapters 1 and 2 assess how number of visual semantic features listed by participants as being part of a given concept influence the speed of word meaning computation, and its neural underpinnings, providing evidence for modality-specific neural organization. Chapter 3 assesses the flexibility of knowledge activation as a function of specific task constraints, suggesting a multi-faceted approach to semantic richness is needed. Chapter 4 describes a novel application of recent statistical advances to the analysis of real-time electrophysiological data, and highlights some limitations of standard analytical approaches. Chapter 5 assesses the real-time influence of several types of knowledge on the neuroelectric activity underlying concrete word meaning computation. A timecourse of sensory-based knowledge type activation is outlined. Finally, Chapter 6 describes a novel approach whereby real-time electrophysiological brain activity is used to predict the speed of semantic decision making, providing further evidence of a highly flexible, but finely structured, human semantic memory system.

Becoming a face expert takes years of learning and development. Many research programs are devoted to studying face perception, particularly given its prerequisite role in social interaction, yet its fundamental neural operations are poorly understood. One reason is that there are many possible explanations for a change in facial appearance, such as lighting, expression, or identity. Despite general agreement that the brain extracts multiple layers of feature detectors arranged into hierarchies to interpret causes of sensory information, very little work has been done to develop computational models of these processes, especially for complex stimuli like faces. The studies presented in this thesis used nonlinear generative models developed within machine learning to solve several face perception problems. Applying a deep hierarchical neural network, we showed that it is possible to learn representations capable of perceiving facial actions, expressions, and identities, better than similar non-hierarchical architectures. We then demonstrated that a generative architecture can be used to interpret high-level neural activity by synthesizing images in a top-down pass. Using this approach we showed that deep layers of a network can be activated to generate faces corresponding to particular categories. To facilitate training models to learn rich and varied facial features, we introduced a new expression database with the largest number of labeled faces collected to date. We found that a model trained on these images learned to recognize expressions comparably to human observers. Next we considered models trained on pairs of images, making it possible to learn how faces change appearance to take on different expressions. Modeling higher-order associations between images allowed us to efficiently match images of the same type according to a learned pairwise similarity measure. These models performed well on several tasks, including matching expressions and identities, and demonstrated performance superior to competing models. In sum, these studies showed that neural networks that extract highly nonlinear features from images using architectures inspired by the brain can solve difficult face perception tasks with minimal guidance by human experts.

To help circumvent reading disability (RD) decoding difficulty, Text-To-Speech (TTS) software can be used to present written language audibly. Although TTS software is currently being used to help RD students, there is a lack of empirically supported literature to inform developers and users of TTS software on best practices. This dissertation investigated two methods to determine whether they increase the effectiveness of TTS for RD and typically-developing students. The first method compared low and high quality TTS voices in regards to understanding. TTS voice quality was identified by having 40 university students listen to and rate the quality of 10 commonly used TTS voices and 2 human voices. Three voices were chosen for the subsequent study based on the ratings; one low quality TTS, one high quality TTS, and one natural voice (Microsoft Mary, AT&T Crystal, and Susan, respectively). Understanding was assessed with tests of intelligibility and comprehensibility. Forty-five grade 6 to 8 students who were identified as having a RD were compared to same-age typically-developing peers. Results showed high quality TTS and natural voice were more intelligible than the low quality TTS voice, and high quality TTS voice resulted in higher comprehensibility scores than low quality TTS and natural voice. The second method investigated whether it is possible to increase a student’s comprehension when using TTS by modifying the presentation style of the TTS voice. The presentation style was manipulated in two ways: varying the speed at which the TTS presented the materials (120, 150, 180 words per minute) and the presence of pauses varied (no pauses inserted, random pauses inserted, or 500 millisecond pauses at the end of noun phrases). Due to a floor effect on the comprehension of the texts the expected results were not obtained. A follow up analysis compared the participants’ prosodic sensitivity skills based on whether they had a specific language impairment, (SLI) a reading impairment (RI), or were typically-developing. Results suggested that SLI has significantly less auditory working memory then RI impacting their auditory processing. Recommendations for future research and the use of TTS based on different learning profiles are provided.

Deception is not only part of daily life for adults, but it is also part of typical development throughout childhood. Research has shown that the ability to deceive improves with age and has been found to be related to the development of cognitive skills such as executive functioning processes, theory of mind, and intelligence. However for some children, lie-telling becomes problematic and atypical. For children with conduct problems, deception is one of the most common presenting symptoms and has been found to be pervasive across childhood and into adolescence. While most studies analyze the lying behaviours of these children based on parent and teacher reports, no studies have yet empirically evaluated the actual deceptive behaviours of children with severe conduct problems. The current study investigated the antisocial and prosocial deception rates and abilities of children with and without severe conduct problems. Additionally, cognitive measures and parental reports of lying and parenting styles were considered in order to determine how such variables may be related to deception. A total of 66 children participated in the current study with half of the sample originating from a community agency for children with behaviour problems and the other half consisting of an age- and gender-matched control sample. Results demonstrated that compared to their typically developing counterparts, children with conduct problems were more likely to behave antisocially by committing antisocial transgressions, and subsequently lying. These children were also less likely to tell prosocial lies. Consistent with previous research, many cognitive measures, such as Executive Functioning (e.g., inhibition) and Theory of Mind (1st and 2nd Order) were found to be related to deception for the typically developing sample, though fewer and different cognitive measures were found to be predictive of deception within the clinical sample. Parent ratings of lie-telling frequency were not predictive of antisocial or prosocial deception, though they were predictive of antisocial lie-telling sophistication. Parenting styles were minimally predictive of deception. Results suggest that the mechanisms used by children with conduct problems during deception differ from a typically developing sample. Limitations as well as implications are discussed.