Face recognition is vital for many of our day-­‐to-­‐day social interactions. For the most part we can effortlessly recognise the faces of family, friends, casual acquaintances and famous people, ignoring the new hairstyle or fashionable sunglasses that they might be sporting and attending to diagnostic features that reveal their identity. From a psychological or computational standpoint, our flair for face recognition is particularly interesting because, unlike the many other categories of object that we routinely encounter, faces must be individuated. That is to say, faces must be differentiated at the within-­‐category level, placing a unique demand on the visual system’s ability to rapidly and accurately discriminate a large number of visually similar patterns. The ease with which we recognise familiar faces belies the computational complexity of the task, rendering us unaware of the dramatic image variance caused by changes in viewing angle, lighting conditions and partial occlusion, and leaving us with an illusion of stability that is characteristic of our impression of the visual world. Clearly, with some 120 million retinal cells semi-­‐ independently encoding the various aspects of a visual scene (Palmeri & Cottrell, 2010), the problem of face recognition is not (normally) one of visual acuity. Rather, the problem of face recognition is one of dimensional reduction, representing faces in a way that ignores the gross image-­‐level variance across different instances of the same person while still maintaining the ability to discriminate between different faces (see Figure 1.1)

A top-down attentional set allows selective processing of the most informative aspects of a scene by biasing attention towards task-relevant stimuli and away from task-irrelevant stimuli on the basis of task demands. The work in this thesis explored the characteristics of the attentional set and top-down control by measuring the persistence of a set. That is, the carry-over of a set from a task in which it is appropriate to a subsequent task in which it is inappropriate. Twelve experiments were completed, employing three different methodologies in order to provide converging evidence for the persistence of attentional set. The first method was a rapid serial visual presentation task, the second was a change detection task that was preceded by a visual search task, and the third was a visual search of natural scenes following an unrelated search through letter strings. All three methodologies provided strong evidence for the carry-over effect, whereby the allocation of attention in a second task was influenced by the top-down settings from the first task. This shows that an attentional set is not established solely based on current task demands but is also influenced by previous experience. Carry-over appears to be contingent upon the level of control invested in the task; too much control over the initial top-down set will enhance carry-over, but a high level of control in the second task will attenuate carry-over. In addition, a lack of executive control over the set will also lead to carry-over when the set is highly practiced because task performance will not be monitored, and a change in task demands will not be accompanied by a change in attentional set. Carry-over provided evidence for three different types of attentional set; a location-based set, a feature-based set, and a feature-value-based set. It also indicated that the attentional orienting system can be configured at more than one level according to the task demands, implying that top-down control is more flexible than previously suggested. The work ultimately led to the development of a general model of attentional set (G-MAS) which attempts to explain the current results and account for pertinent findings from the literature.

Previous research suggests that facial characteristics associated with body mass index (BMI) play an important role in health and attractiveness judgments of faces. However, very little work has investigated the factors that predict individual differences in preferences for facial cues of adiposity or how these individual differences are related to social outcomes. In light of the above, the first two empirical chapters of this thesis investigated the relationships between individual differences in preferences for facial cues of adiposity and (1) the BMI of men’s and women’s actual romantic partners and (2) disgust sensitivity. Analyses suggested that people with particularly strong preferences for slim-looking faces were more likely to have partners with low BMI and that men, but not women, who scored higher on pathogen disgust showed stronger aversions to faces displaying cues associated with high BMI. The third chapter investigated how people integrate information from shape cues of adiposity and information from skin color when judging the health and attractiveness of faces. Analyses showed that preferences for cues of low BMI were particularly strong when assessing faces displaying skin color cues associated with the absence of illness. These results suggest that integrating information from shape cues of adiposity and information from skin color could allow people to distinguish between individuals with low BMI because they are healthy and those with low BMI due to illness. Most research investigating the role of facial cues of adiposity in social perception has focused on the possible role of facial adiposity as a health cue. However, it is also possible that facial cues of adiposity contain other types of information, such as information about a person’s reported sociosexual orientation (openness to short-term, uncommitted sexual relationships). To explore this issue, the fourth empirical chapter of my thesis investigated the relationship between facial correlates of BMI and women’s sociosexual orientation. Although analyses suggested that slimmer women reported greater openness to short-term, uncommitted sexual relationships, the observed relationships were weak and, thus, unlikely to play an important role in social interactions. Together these studies support the claim that responses to facial cues of adiposity are related to romantic partner choice and function to distinguish between healthy and unhealthy individuals.

Time is an elemental dimension of human perception, cognition and action. Innumerable studies have investigated the perception of time over the last 100 years, but the computational basis for the processing of temporal information remains unknown. This thesis aims to understand the mechanisms underlying the perceived timing of stimuli. We propose a novel Bayesian model of when stimuli are perceived that is consistent with the predictive coding framework – such a perspective to how the brain deals with temporal information forms the core of this thesis. We theorize that that the brain takes prior expectations about when a stimulus might occur in the future (prior distribution) and combines it with current sensory evidence (likelihood function) in order to generate a percept of perceived timing (posterior distribution). In Chapters 2-4, we use human psychophysics to show that the brain may bias perception such that slightly irregularly timed stimuli as reported as more regular. In Chapter 3, we show how an environment of irregularity can cause regularly timed sequences to be perceived as irregular whilst Chapter 4 shows how changes in the reliability of a signal can cause an increased attraction towards expectation.

Social perception (i.e., the formation of impressions based on perceivable cues) of both faces and bodies is an integral part of social interaction and can influence and can be influenced by many variables, such as motivational salience (i.e., the amount of effort an individual will expend to continue viewing faces and bodies) and hormone levels of the perceiver. The first empirical chapter (i.e., Chapter 2) investigated social perception of faces and bodies using multiple trait ratings. First, participants rated face and body stimuli on the same 13 traits as those used in the seminal article on social perception of faces. Replicating previous work, I found that social perception of faces can be summarized by the two-component pattern of valence (i.e., intent to cause harm) and dominance (i.e., ability to cause harm). Social perception of bodies, though, can be summarized by one main component. Therefore, social perception of faces and bodies followed different, distinct patterns. The second empirical chapter (i.e., Chapter 3) investigated the relationship between the social perception components established in Chapter 2 and motivational salience. I assessed motivational salience using a standard key-press task in which participants could increase or decrease stimulus viewing time by pressing specified keys on the keyboard. Replicating previous work, valence and dominance positively and independently predicted the motivational salience of faces. Additionally, the one main social perception component of bodies positively predicted the motivational salience of bodies. The third empirical chapter (i.e., Chapter 4) investigated the relationship among the previously established social perception component of bodies, motivational salience of bodies, and hormone levels of the perceivers. I used the passive drool method of hormone measurement to determine exact hormone levels at five weekly test sessions. Similar to studies of faces, motivational salience of bodies was greater when testosterone was higher. While the one social perception component for bodies positively predicted motivational salience separately for male and female bodies, there was no interaction between testosterone and the social perception component, failing to conceptually replicate previous interactions between testosterone and stimulus valence. Overall, I first replicated the two-component social perception pattern of valence and dominance for faces before finding a different, one-component social perception pattern for bodies. In turn, each of these social perception components predicted motivational salience of faces and bodies. Additionally, motivational salience of bodies was greater when testosterone was high, but this effect was not qualified by the main social perception component for bodies. I conclude by discussing the similarities and differences between faces and bodies in this and other work on social perception and motivational salience.

The notion that unconscious Bayesian inference underlies perception is gaining ground. Predictive coding approaches posit that the state of the world is inferred by integrating, at each level of the perceptual hierarchy, top-down prior beliefs about sensory causes and bottom-up prediction errors. In this framework, percepts correspond to a top-down stream of beliefs that best 'explain away' sensory signals. Although such frameworks are gathering empirical support, subjective facets of perception remain unexplained from these perspectives. This thesis combines behavioural, computational and neuroimaging methods to examine how subjective visual confidence can be accounted for in a predictive coding framework. Experiment one shows that, behaviourally, perceptual expectations about target presence or absence both liberalise confidence thresholds and increase metacognitive accuracy. These results are modelled in a signal detectiontheoretic framework as low-level priors shifting the posterior odds of being correct. Using EEG, experiment two reveals that influence of expectations on decision and confidence oscillates with the phase of pre-stimulus alpha oscillations. This means that prior to target onset, both objective and subjective decisions have been rhythmically biased by the periodic recruitment of expectations to visual areas. Using fMRI, experiment three shows that in the post-stimulus period, expectations and sensory signals are integrated into confidence judgements in right inferior frontal gyrus (rIFG). Furthermore, this process recruits orbitofrontal cortex and bilateral frontal pole, which represent top-down influences, and occipital lobe, which represents bottom-up signals. Together, these results suggest that expectations shape subjective confidence by biasing the posterior probability of the perceptual belief.

Within-person variability has largely been neglected in face processing research, with research typically focusing instead on between-person variability. In experimental settings between-person variability often becomes between-image variability with research using one image to represent a face. This implies that an image is an adequate representation of a face, however one image cannot illustrate the variability that can occur in facial appearance. This thesis argues that overlooking within-person variability is a fundamental flaw in face processing research, as within-person variability is surprisingly large. The experiments in this thesis illustrate the effect of within-person variability in different face processing contexts: face identification, face perception and image memory. Experiments 1 – 7 demonstrate the difficulty of identifying familiar and unfamiliar faces across within-person variability using an image-sorting task. Experiment 8 illustrates the within-person variability that occurs in personality perception, and Experiments 9 and 10 illustrate the within-person variability that occurs in the perception of facial attractiveness. Lastly Experiments 11 and 12 introduce within-person variability to memory recognition and demonstrate the difficulties of remembering multiple images of the same face. From the results of Experiments 1 – 12 it is concluded that within-person variability is highly influential in all the discussed areas of face processing and therefore needs to be taken seriously in face processing research.

Facial attractiveness is a particularly salient social cue that influences many important social outcomes. Using a standard key-press task to measure motivational salience of faces and an old/new memory task to measure memory for face photographs, this thesis investigated both within-woman and between-women variations in response to facial attractiveness. The results indicated that within-woman variables, such as fluctuations in hormone levels, influenced the motivational salience of facial attractiveness. However, the between-women variable, romantic relationship status, did not appear to modulate women’s responses to facial attractiveness. In addition to attractiveness, dominance also contributed to both the motivational salience and memorability of faces. This latter result demonstrates that, although attractiveness is an important factor for the motivational salience of faces, other factors might also cause faces to hold motivational salience. In Chapter 2, I investigated the possible effects of women’s salivary hormone levels (estradiol, progesterone, testosterone, and estradiol-to-progesterone ratio) on the motivational salience of facial attractiveness. Physically attractive faces generally hold greater motivational salience, replicating results from previous studies. Importantly, however, the effect of attractiveness on the motivational salience of faces was greater in test sessions where women had high testosterone levels. Additionally, the motivational salience of attractive female faces was greater in test sessions where women had high estradiol-to-progesterone ratios. While results from Chapter 2 suggested that the motivational salience of faces was generally positively correlated with their physical attractiveness, Chapter 3 explored whether physical characteristics other than attractiveness contributed to the motivational salience of faces. To address this issue, I first had the faces rated on multiple traits. Principal component analysis of third-party ratings of faces for these traits revealed two orthogonal components that were highly correlated with trustworthiness and dominance ratings respectively. Both components were positively and independently related to the motivational salience of faces. While Chapter 2 and 3 did not examine the between-woman differences in response to facial attractiveness, Chapter 4 examined whether women’s responses to facial attractiveness differed as a function of their romantic partnership status. As several researchers have proposed that partnership status influences women’s perception of attractiveness, in Chapter 4 I compared the effects of men’s attractiveness on partnered and unpartnered women’s performance on two response measures: memory for face photographs and the motivational salience of faces. Consistent with previous research, women’s memory was poorer for face photographs of more attractive men and more attractive men’s faces held greater motivational salience. However, in neither study were the effects of attractiveness modulated by women’s partnership status or partnered women’s reported commitment to or happiness with their romantic relationship. A key result from Chapter 4 was that more attractive faces were harder to remember. Building on this result, Chapter 5 investigated the different characteristics that contributed to the memorability of face photographs. While some work emphasizes relationships with typicality, familiarity, and memorability ratings, more recent work suggests that ratings of social traits, such as attractiveness, intelligence, and responsibility, predict the memorability of face photographs independently of typicality, familiarity, and memorability ratings. However, what components underlie these traits remains unknown, as well as whether these components relate to the actual memorability of face photographs. Principal component analysis of all these face ratings produced three orthogonal components that were highly correlated with trustworthiness, dominance, and memorability ratings, respectively. Importantly, each of these components also predicted the actual memorability of face photographs.

Recent work in defence of direct realism has concentrated on the representationalist and disjunctivist responses to the arguments from illusion and hallucination, whilst relatively little attention has been given to the argument from causation which has been dismissed lightly as irrelevant or confused. However such charges arise from an ambiguity in the thesis which is being defended and the failure to distinguish between metaphysical and epistemological issues and between factual and conceptual claims. The argument from causation, as an argument against the conjunction of metaphysical direct realism and an explanation of the perceptual process in terms of a naturalistically understood causal chain of events, has not been answered in the philosophical literature. Moreover when the process of perception is fleshed out in terms of contemporary cognitive science, the difficulties are compounded. Neither representation-friendly mainstream cognitive science, nor representation-averse radical embodied cognitive science, is compatible with a theory of perception which is at the same time both direct and robustly realist.

Following the work of Dixon (1971) some experiments were performed to investigate two aspects of perception without awareness, the handling of emotional stimuli and the relation between subliminal perception and selective attention. Apparatus was designed which utilized the phenomenon of binocular rivalry, so that an image (above identification level when presented alone) could be masked by a brighter image to the other eye and thus perceived without awareness. An experiment of Smith et al. (1959) was replicated with improved controls. It was shown that responses to words presented outside of awareness tended to be meaning-related, the same words yielding structure-related responses when presented supraliminally. Spence (1967) proposed an explanation of perceptual defence in terms of the interaction of arousal and memory. Some experimental support for this idea was obtained. Further experiments on the handling of emotive stimuli led to the conclusion that individual differences in perception are an important factor to be controlled. Similarly, further to Brown (1965,1971) it was concluded that the stimulus characteristics of emotive words used as experimental stimuli need to be better controlled. An explanation of word association phenomena in terms of the interaction of arousal and attention was discussed and the perceptual defence and WAT situations contrasted. Finally, two brief experiments illustrated aspects of the selective attention paradigm relevant to perception without awareness: pre-attentive processes (Neisser, 1967) and incidental stimulation (Eagle et al, 1966). Following a review of selective attention experiments, including evidence of unattended channel processing, some tentative proposals were made which might encompass the material presented. Utilizing a model suggested by MacKay (1972) it was proposed that the phenomena of perception without awareness represent the functioning of an early stage in the normal perceptual process essential both to the handling of emotive stimuli and the selection of inputs to awareness.