Normal aging is known to be accompanied by loss of brain substance. The present study was designed to examine whether the practice of meditation is associated with a reduced brain age in meditators. Specific focus was directed at age fifty and beyond, as mid-life is a time where aging processes are known to become more prominent. We applied a recently developed machine learning algorithm trained to identify anatomical correlates of age in the brain translating them into one single score: the BrainAGE index (in years). Using this validated approach based on high-dimensional pattern recognition, we estimated and compared the brain ages of 50 long-term meditators and 50 control subjects. We observed that, at age fifty, brains of meditators were estimated to be 7.5 years younger than those of controls. In addition, we examined if the brain age estimates change with increasing age. While in controls, brain age estimates varied only little, significant changes were detected in meditators: for every additional year over 50, meditators’ brains were estimated to be an additional 1 month and 22 days younger than their chronological age. Altogether, these findings seem to suggest that meditation is beneficial for brain preservation, effectively protecting against age-related brain atrophy, with even increasing effects (i.e., augmented conservation) in later years of life.

While research suggests a relationship exists between essential fatty acid (EFA) status and cognition, there is little research within young adult populations with attention deficit hyperactivity disorder (ADHD). As omega-3 fatty acids have been found to improve ADHD symptoms, the current study used fMRI analysis to investigate the relationship between EFA status, symptoms of ADHD and performance on a Stroop task in a young adult population. The analysis focused on the anterior cingulate cortex (ACC), an area commonly associated with cognitive control and attention, and a region identified in functional abnormalities associated with ADHD. The study aimed to investigate both the presence of ADHD and the effect of supplementation on brain activation measured with fMRI while completing a modified Stroop task.
In a double-blinded, placebo-controlled, randomised parallel study, 46 participants (26 control and 20 ADHD) were divided into three diet groups, an EPA-rich diet, DHA-rich diet and a soya oil placebo treatment for 12 weeks. Participants underwent fMRI scans before and after supplementation. Activation patterns associated with three different Stroop task conditions were analysed using SPM software.
At baseline, differences were detected between the brain activation of the ACC between the control and the ADHD participants for the Stroop task (p<0.05). After the supplementation period, no significant differences between ADHD and control were detected for any of the supplementation groups. Post-hoc analyses revealed no interaction between supplementation and ADHD status. The interpretation of these changes is explained in terms of the neural efficiency theory.

The intraparietal sulcus (IPS), located in the posterior parietal cortex, is active during spatial transformation. The IPS contains several regions (human intraparietal regions 1, 2 and 3; hIP) that show a topographic organization for visual–spatial information, similar to retinotopic maps in the visual cortex. Functionally segregating the IPS can aid in consolidating the difficulties encountered with brain maps and IPS regions of interest (ROI). A possible way to explore functional segregation is through fMRI neuroimaging studies while administering tasks that emphasise the role of the IPS in spatial transformation. Antisaccade and mental rotation tasks require two different forms of spatial transformation: antisaccades require inversion of the vector for an eye movement from the target location to its mirror opposite location; mental rotation requires transformation of the presented stimulus to its correct orientation. Overlap of functional activity in the IPS between the two tasks will demonstrate a non-context dependent role of this region in spatial transformation. We hypothesised that medial and posterior medial IPS regions would show overlap of activation. Fourteen healthy participants (M= 29.07 years, SD= 5.85) completed antisaccade and mental rotation tasks. fMRI conjunction analysis showed overlapping activity within the medial and posterior medial IPS regions. Probabilistic cytoarchitectonic maps (Eickhoff et al., 2005) revealed that 70% of the cluster was within the cytoarchitectonically defined IPS areas hIP3 and hIP2. In particular, the conjunction of these tasks activates areas hIP3, BA2 and SPL 7PC. These results suggest that these IPS regions are important for spatial transformations independent of current context.

Our brain does not process all arriving information equally. Instead, our current brain state provides an important constraint on how we react to information. Fluctuations in brain states can be tracked by measuring oscillations in ongoing neural activity using electroencephalography (EEG). In particular, variations in pre-trial alpha oscillations have been associated with modulations in our sensitivity to incoming sensory information. Here we investigated how fluctuations in pre-trial alpha influence our ability to encode numerical information and to engage in complex reasoning using that information. Thirty-five undergraduate students completed a fraction reasoning task while EEG data where collected. Participants where presented with two fractions, separated by a brief delay, and asked to indicate which of the two fractions represented the greater proportion. Across two conditions fractional quantities were represented either pictorially (as dot clouds) or symbolically (using standard fraction notation). Behavioural results demonstrated a fractional distance effect, whereby participants took longer to make the proportion judgment, the closer the numerical distance between the two fractions. This fractional distance effect was also present in event-related potentials (ERPs) recordings, which showed greater p3 amplitude for fraction pairs that had a small numerical distance between them. This effect was largest for fractions presented as dot clouds. There was also a significant correlation between pre-trial alpha amplitude and the amplitude of the p3 distance effect, suggesting that fluctuations in our readiness to receive information has important consequences for high level reasoning.

The highly specific prediction, based on physiological properties of the magnocellular inputs to the dorsal visual stream, that eye movements will be affected by rapid, non-conscious encoding of low-luminance peripheral landmark information was tested. Participants fixated centrally, and then moved their eyes left or right to discriminate a peripheral (12.7 deg) target object. Low luminance (6.5 cd/m2) or very low luminance (2.1 cd/m2) landmark cues (X + T) were presented briefly (33ms) and bilaterally at the same eccentricity as the target, 66ms prior to its onset. The spatial arrangement of landmarks predicted target location (p = .8), but participants were not informed of this; a post-task questionnaire evaluated awareness of the relationship. Participants who remained unaware of the predictive utility of landmarks initiated saccades more rapidly when targets appeared at predicted compared to non-predicted locations. A small number of anticipatory saccades, initiated in the brief interval after landmark onset but prior to target encoding, were observed. The direction of these saccades was biased towards the location predicted by landmark stimuli. In a second task, participants were able to discriminate consciously between the two possible spatial arrangements of landmark stimuli with high levels of accuracy. It is concluded that: (1) Eye movements can be influenced by rapid encoding of low luminance, peripheral landmark cues; (2) This encoding may be described as preconscious; (3) Preconscious encoding of peripheral landmark cues probably relies on the dorsal visual stream.

The ability to flexibly orient to goal-relevant stimuli necessarily requires the interaction of feedforward and feedback cortical activity. Recent work in non-human primates has characterised distinct oscillatory frequencies as carrier signals for feedforward and feedback information. Theta and gamma oscillations have been identified as preferentially associated with the feedforward flow of information, while alpha/beta oscillations have been associated with feedback information. Here we investigated the roles of theta and alpha oscillations in goal-directed visual attention, and asked whether their involvement is consistent with their putative roles as carriers of feedforward and feedback information. We had participants respond to a target of a particular colour among heterogeneously coloured distractors. Prior to the appearance of the target display we cued one location with a nonpredictive cue that was either target- or nontarget-coloured. To map alpha and theta activity in response to these cues we recorded EEG and behavioural responses. We found that theta oscillations became lateralised early and in response to all cues. This lateralisation was larger, however, if the cue matched the target colour, consistent with feature-based enhancement of the feedforward signal. Alpha oscillations became lateralised later, and only in response to target-coloured cues, consistent with what is known about the behaviour of spatial attention under these conditions. These results demonstrate the involvement of theta and alpha oscillations in goal-directed attention, and are consistent with the hypothesised roles of theta and alpha oscillations as carrier frequencies for feedforward and feedback signals, respectively.

The ability to focus on current goals without being distracted by goal-irrelevant information is highly adaptive. Emotional information captures our attention to a greater extent than neutral information; therefore ignoring emotional distractions may require a more effective control strategy. Using behavioural and EEG methods, we study the mechanisms used to control emotional distractions. In Experiment 1, participants performed a simple letter discrimination task at fixation, while irrelevant positive, negative and neutral images were presented peripherally. When distractors occurred on 25% of trials, emotional images were more distracting than neutral images. However, when distractors occurred on 75% of trials, both emotional and neutral distractors were effectively ignored, showing that we can control emotional distractions when they are expected. The attenuation of distraction in the high frequency condition may result from participants adopting a strategy of greater proactive control (i.e., anticipating and preparing to ignore distractors prior to onset) when distractors are expected. In Experiment 2, we are currently testing this hypothesis, recording EEG during task performance to measure pre-stimulus alpha suppression, an established index of anticipatory attention. Distractor valence (positive, negative, or neutral) is blocked, so that participants expect emotional or neutral distractors. If participants adopt proactive control in the high, but not low, distractor frequency condition, there should be greater pre-stimulus alpha suppression in the high frequency condition than in the low frequency condition. In addition, if ignoring emotional distractors requires greater proactive control, pre-stimulus alpha suppression should be greater during emotional distractor blocks than non-emotional distractor blocks.

Recent work using singleton distractors has called into question the veracity of a single-route account of visual search, while also demonstrating that search prior to deployment of focal attention can still localize objects. The present study extends on this work by asking participants to indicate the location of a target using a computer mouse. Using spatial accuracy as a dependent variable, and displaying stimuli for only short durations, results show that participant’s spatial accuracy is largely unchanged by across-feature singleton distractors, varying frequency of singleton distractors, and varying salience of across-feature singleton distractors. By contrast, within-feature singleton distractors reduced spatial accuracy. These results are compatible with a dual-route account of visual search; in addition, we will discuss potential limitations of the spatial task and brief displays.

In anticipation of a critical event, expectation increases over time. When the onset of a critical event is uncertain, reaction times typically reflect the evolving probability of stimulus onset as a function of elapsed time. Increasing probability over time (the hazard rate) has been a long-standing explanation for the inverse relationship between reaction time and foreperiod length. Multi-process accounts of temporal preparation suggest that this relationship is the result of controlled monitoring of elapsed time, relying upon available cognitive resources such as working memory. We investigated whether limiting available cognitive resources via a working memory load task would affect the inverse relationship between hazard rate and response speed. One-hundred and one participants were tested across two experiments manipulating the hazard rate of auditory target stimuli while concurrently completing N-back visual tasks manipulating working memory load. Limiting the available cognitive resources appeared to selectively affect temporal preparation when the hazard rate increased steadily with foreperiod length. Our data showed an inverse relationship between the hazard rate and response speed even under conditions of high working memory load. This might indicate that temporal preparation is not limited by available cognitive resources and may rely on an automatic mechanism.

Emotional information is important and attracts attention better than neutral information. However, emotional stimuli are sometimes irrelevant to current goals, and must be ignored. In a series of behavioural and ERP experiments we examined the mechanisms that allow us to ignore emotional information. Across experiments, participants performed a simple letter identification task near fixation. On some trials a distracting image (negative, neutral or positive) was presented peripherally. Experiment 1 showed that irrelevant emotional images were more distracting than neutral images. Experiment 2 showed that emotional distractors can be suppressed as effectively as neutral distractors when anticipated (i.e., when distractors were presented on 75%, compared to 25%, of trials). Experiment 3 showed that frequent emotional distractors are not merely suppressed via proactive suppression of potential distractor locations. In experiment 4, ERPs were measured during task performance. An early (EPN) and late (LPP) index of emotional processing was used to determine how increased frequency reduces distraction from emotional images. For positive images, increased distractor frequency was associated with attenuation of the EPN, suggesting that expectations affected image processing at early stages. Paradoxically, positive images produced similar LPPs in both conditions. Results were less clear for negative images, because they did not show an EPN under either frequency. Thus, behavioural evidence for effective control of emotional distractors was not clearly explained by these two ERP components. Other components that are sensitive to emotion (such as N2pc or C1) may be useful alternatives in further explorations of the mechanisms used to control emotional distraction.

Pain is an experience that motivates self-protective behaviour when threat to body tissue is detected. This threat protection is useful when pain is acute, however when pain becomes chronic this process becomes maladaptive. Among other negative effects, people experiencing chronic pain report cognitive deficits. These subjective experiences correspond to performance deficits on cognitive tasks assessing memory, attention, and executive function. According to a motivated cognition view, these deficits may not be widespread impairments in cognitive function; rather they may reflect attentional biases to pain-relevant information at the expense of neutral information. Some support for a motivated cognition account comes from dot-probe and emotional-Stroop studies, but these effects are small and not robust in people experiencing pain. In three experiments, we used the attentional blink task as an alternative way to assess attention to pain-relevant and neutral information under capacity limits in people not experiencing pain. Target 1 was always neutral but target 2 could be pain-relevant or neutral. In Experiments 1 and 2, pain-relevant information was less susceptible to the attentional blink, and in Experiment 2, interfered with processing of target 1. However in Experiment 3, these effects were not replicated. Together, findings suggest a small but perhaps inconsistent bias to pain-relevant information in healthy participants. In ongoing research we are extending our experiments to examine attentional biases in people with chronic pain.

Boundary extension, the tendency to remember scenes as having extended boundaries, is a robust memory error. Reduced attention to a scene appears to attenuate boundary extension. Importantly, attentional allocation has spatial asymmetries: neurotypical individuals pay more attention to the left side of space compared to the right (pseudoneglect). Could this attentional asymmetry affect boundary extension? Recently, Dickinson and Intraub (2006) found that participants misremembered extended boundaries more often on the right side of an image compared to the left. These data suggest that pseudoneglect attenuates boundary extension. However, we also know people tend to incorrectly estimate that a bisected line is longer on the left than on the right (landmark task; e.g., Cavezian, et al., 2012). Thus, on the one hand, people might make fewer boundary extension errors for images processed in the left hemifield because they overattend to this side of space. On the other hand, people might make more boundary extension errors for images on the left because they overestimate space on the left. In the present study we presented a neutral image to participants’ left and right visual fields concurrently. We instructed participants to focus on a fixation cross that appeared immediately before and with the images. Then, we asked participants to choose between the original and an extended version of one of the presented images. Participants made more boundary extension errors for images presented on the left than on the right, consistent with literature showing people tend to overestimate the size of space on the left.

The relationship between attention and emotion has been a topic of increasing importance in the study of affective cognition. Theories have highlighted the importance of emotion to goal-relevant cognition, leading to a focus on threatening stimuli and the relationship between threat and anxiety in attracting attention. One of the main tools used to examine this is the “dot-probe” task; however, mixed findings have led to questions about its reliability. This study examined the relationships between threat bias, anxiety and dot-probe performance by using previously unexamined dynamic emotional stimuli. The aim was to assess whether the unreliability of the dot probe task could be explained by the stimulus control and unrealistic facial stimuli. Participants completed a dot-probe task in which to both static and dynamic facial stimuli were used as cues for spatial attention. Questionnaires were also administered to assess participants’ anxiety levels. We expected that dynamic stimuli would potentiate the capture of attention by threatening emotional expressions, with stronger effects in the high-anxiety participants. Preliminary data suggest that while a response bias was not demonstrated towards threatening stimuli in general, faster responses were evoked by dynamic stimuli. Furthermore, when taking participant anxiety into account, significant differences were found within the emotion type, dot-congruency and stimulus type interactions. This study suggests that the use of static versus dynamic stimuli may affect responses to the dot-probe task, especially when accounting for anxiety, and that the reliability of the dot-probe task may need further examination, especially in non-anxious participants.

Background: Spatial attention is modulated by a person’s level of vigilance and alertness. There is evidence that a decrease in vigilance/alertness is associated with a rightward shift of attention. Levels of alertness fluctuate over the course of a day, and peak times of alertness differ between individuals. Some individuals feel most alert in the mornings; others feel more alert in the evenings. Our aim was to investigate the influence of an individual’s peak time (chronotype) and time of testing on spatial attention.

Method: Drawn from Amazon Mechanical Turk, 531 right-handed participants filled in questions from the Ostberg Morningness/Eveningness Scale and performed a spatial attention task (landmark task). For the landmark task, participants indicated whether the left or right segment of pre-bisected lines was longer. We calculated a response bias for each participant by subtracting the ‘number of left responses’ from the ‘number of right responses’, and dividing by the number of trials.

Results: Controlling for participants’ age, the response bias was submitted to an ANOVA with the factors ‘chronotype’ and ‘time of testing’ (morning = 6am-2.30pm, evening = 2.30pm-11pm). There was a significant interaction between chronotype and time of testing (p<0.05). Posthoc tests showed a relative rightward shift of attention from peak to off-peak times of testing for morning, but not evening chronotype groups.

Conclusions: The deployment of spatial attention was modulated by an individual’s peak time and time of testing. This novel observation corroborates previous research demonstrating a close coupling between non-spatial and spatial attentional functions.

While working memory and selective attention have traditionally been viewed as distinct processes in human cognition, there is now a growing body of literature which demonstrates significant overlap between these two constructs. One line of evidence for the presence of this relationship comes from between-subject designs in which individuals low in working memory capacity show greater interference in standard attention-based tasks. The specific neural mechanisms involved in this interaction are still unknown, however three lateralised ERP components have been identified as ideal candidates for studying the neural underpinnings of attention processes. N2pc, Ptc, and SPCN have been associated with object selection, attentional disengagement, and short term maintenance of target features respectively. In this study we measured these components during the delay period of a working memory task in order to see whether they could predict performance on the working memory task. The results showed that under high load, working memory accuracy is predicted by N2pc amplitude whereas response times were predicted by the amplitude of the Ptc component. Under low load however, both working memory accuracy and working memory response times are predicted by Ptc amplitude. In all cases, increases in component amplitudes predicted improvements on the behavioural measure. These results provide direct neural evidence which is consistent with suggestions that large working memory capacity is a consequence of more efficient distractor disengagement processes.

Predictive coding posits that the human brain is a predictive machine that continually monitors the environment for regularities and detects inconsistencies. It is unclear, however, what effect attention has on these processes. Attention is known to enhance neural responses, whilst prediction attenuates them; with little research and contradictory findings examining the interaction between the two. Here, we investigated the effects of selective spatial attention and predictability on brain dynamics and behaviour. We designed a novel binaural auditory oddball task where participants were required to attend to (and detect targets in) an auditory stream. Oddball tones were either present in the side of the attended ear, or played to the contralateral ear. Participants were asked to attend to the auditory streams in either one or both ears. Behavioural results showed we successfully manipulated the attentional load with faster reaction times and more errors made in the bilateral attention condition. We observed mismatch negativity (MMN) across each and all attentional manipulations and source reconstruction revealed a fronto-temporal network for MMN across all attentional levels. Furthermore, we found a trend for an interaction between attention and deviance in the inferior temporal lobe. Our results demonstrate that attention modulates the neural representations of prediction errors and are in keeping with theoretical ideas of predictive coding.

In two experiments, participants shifted spatial attention in response to bilaterally presented cue letters. In the landmark procedure, two different letters were presented on every trial, and targets were likely to appear near to a specific ‘landmark’ (e.g. the letter X). In the identity cueing procedure, two identical letters were presented on every trial, and participants shifted attention left or right in response to cue identity. The proposal that spatial cueing effects driven by landmark features, and those driven by cue identity, rely on cue encoding within the dorsal and ventral visual streams respectively, was investigated in two experiments. Experiment 1 tested two predictions, derived from the transient and sustained response characteristics of the magnocellular and parvocellular inputs to the dorsal and ventral visual streams. These were, firstly that landmark cueing effects would remain robust under conditions of brief cue exposure time, and secondly that identity cueing effects would collapse when cue exposure time was reduced. Both predictions were confirmed. Experiment 2 tested two further predictions, derived from the varying contrast sensitivities of the magnocellular and parvocellular inputs to the dorsal and ventral visual streams: Firstly, that landmark cueing effects would remain robust when the luminance contrast of cues was low; and secondly that identity cueing effects would collapse under conditions of low cue luminance contrast. Again, both predictions were confirmed. We interpret these results as suggesting that there may be a relatively straightforward relationship between dual-stream models of vision and two-process theories of attention shifting.

As amblyopia is a developmental disorder characterized by reduced visual acuity, and impaired binocular function, but without any observable retinal anomaly, we have previously suggested that visual attention, and eye movement driven shifts in attention may be deficient through the amblopic eye of strabismic amblyopes. To test this hypothesis we compared average speed of saccades involuntarily induced by a rapidly moving motion stimulus, through each eye of 8 adult amblyopes and 8 matched controls. FMRI activation to the involuntary eye movement task and a prescribed visual pursuit task, were also compared between the two eyes of amblyopes and controls. Both involuntarily and voluntarily driven eye movements were significantly slower when driven through the strabismic eye than through the non-deviating eye of strabismic amblyopes or through either eye of normal subjects. The intensity of BOLD activation induced through the amblyopic eye, in ROIs known to be associated with visually driven eye movements including primary visual cortex and intraparietal sulcus and frontal eye fields, was also weaker than through the fellow eye or either eye of normal subjects during both involuntary bottom-up and top-down associated pursuit movements. The parieto-frontal attention network activated by the amblyopic eye also showed abnormal activation to the motion driven task, but normal activation to the pursuit task. Thus both the oculomotor and fMRI results demonstrate that amblyopia is associated with deficits in visual attention.

Relatively few studies have examined the event-related potentials (ERPs) associated with failing to inhibit a response. Examining the ERPs elicited following failed inhibitions provides additional insight into the functional significance of the frontocentrally distributed no-go N2 and no-go P3 components. Using a modified go/no-go task, we examined ERPs elicited following partial inhibitions, which is when participants initiate, but do not complete the response to a no-go stimulus.
In a sample of young adults, we examined the amplitude and latency of the no-go N2 and no-go P3 components elicited by correct go trials, successful inhibitions and partial inhibitions, as well as response onset time. Response onset for partial inhibitions occurred significantly earlier than go responses. No-go N2 amplitudes were largest following partial inhibitions, and the latencies were later for partial compared to successful inhibitions. No-go P3 amplitudes were larger for successful inhibitions, compared to partial inhibitions and go responses. No-go P3 amplitudes did not differ between partial inhibitions and go responses.
Our results indicate that failures of response inhibition are reflected by substantial modulations to both neurophysiological and behavioural measures, highlighting the value of using the response paradigm employed in this modified go/no-go task. Furthermore, examining the additional information obtained by measuring response onset, in conjunction with ERP components elicited during partial inhibitions provides new perspective for investigating the functional significance of the no-go N2 and P3. Implications of the N2 and reduced P3 effect for our understanding of inhibitory processes in the go/no-go task are discussed in detail.

Theta (4-8 Hz) oscillatory activity over the midline frontal scalp (midfrontal theta; MFT) is increased in tasks that require cognitive control to detect conflict and resolve contextual uncertainty. MFT has been associated with a number of ERP components typically modulated by conflict and uncertainty, such as the N2 and the error-related negativity. Both MFT and associated ERP components have been linked to activity in the midcingulate cortex (MCC), leading to recent models suggesting that MFT may reflect a common mechanism for identifying the need for adaptive control and communicating this need across control networks (e.g., Cavanagh & Frank, 2014; TCIS:18:414-421). Other recent work also implicates the MCC in the perception of and response to negative affect (e.g., Shackman et al. 2011; NatRevNeurosci:12;154-167) and reports that MFT vary with individual variability in dispositional anxiety. In this study, we investigate whether the MFT is elicited under proactive and reactive control conditions in a cued-trials task switching paradigm (Karayanidis et al, 2009; CABN:9:202-215). MFT is expected to be associated with the target-locked N2 which varies in amplitude across fully and partially prepared trials. We investigate whether MFT will also be associated with the switch-positivity, an ERP component which varies as a function of switch-related preparation, and/or the pre-target negativity, an ERP component which varies with task-related preparation. Finally, we examine whether MFT elicited proactively or reactively varies as a function of cognitive control ability and psychological distress in late adolescence and young adulthood (n=215,15-25y,98 male). Implications for the adaptive control hypothesis are discussed.

Cognitive control ability has been shown to predict adaptive functioning later in life. Therefore, effective measurement of cognitive control ability in early life may prove important in identifying individuals who would benefit from early interventions to improve their level of functioning and later outcomes. Cognitive control ability can be assessed using performance-based neuropsychological and experimental methods as well as self-report/informant questionnaires. In this study, we examine how scores on a well-normed measure of cognitive control (self-report Behavior Rating Inventory of Executive Function: BRIEF) are related to both neuropsychological (i.e., Trail Making Test: TMT, Stockings of Cambridge: SOC) and an experimental (i.e., cued-trials task-switching) tests of cognitive control. We also examine whether self-report, neuropsychological, and experimental tasks are equally able to predict current outcome behaviours (i.e., self-reported alcohol use, suicidality, and quality of life). Participants included 191 healthy adolescents and young adults (21.23 years+4.91 years, 87 females). Self-reported cognitive control ability was only weakly correlated with performance on neuropsychological and experimental tasks, but more strongly correlated with self-report measures of affect and psychological distress. Linear regression showed that the BRIEF was the strongest predictor of outcome measures, followed by TMT and task-switching performance (SOC significantly predicted no outcomes). These preliminary analyses suggest that performance is more related to method than the construct under investigation (i.e. people who self-report poor cognitive control will likely also self-report poor outcomes). The analyses also suggest that assessment method needs to be considered when examining the relationship between cognitive control and outcome behaviours.

Background: We know very little about risk judgements outside the gambling and finance fields, despite the ubiquity of risk judgements we make everyday – do we cycle or drive, or go skydiving? We have a one-in-a-million chance of dying each day from an accident, with this acute probability being termed one Micromort. Activities we choose can increase our risk, e.g. skydiving is 10 Micromorts, where for every million jumping, 10 are expected to result in death. This study aimed to investigate psychophysiological indices of everyday risk perception, relative to objective metrics (Micromorts) and subjective judgements.
Method: We tested 27 participants (56% female), aged between 18-36. Participants judged whether activities conveyed approximately the same risk, significantly more risk, or significantly less risk than either skydiving or going under general anaesthesia (both 10 Micromorts). Event-related potentials (ERPs) were calculated relative to objective risk (Micromorts) and subjective risk, in three categories: high, medium/same, and low risk.
Results: The P3 showed a significant main effect of objective risk category, such that larger P3 amplitudes were found for the high-risk category, followed by the low-risk and medium-risk categories. Relative to subjective risk responses, the P3 showed a similar pattern of results, however this did not reach significance.
Conclusion: The P3 ERP component appears sensitive to risk magnitude, however, the effect is stronger for objective rather than subjective risk. To our knowledge, this is the first psychophysiological study to illustrate risk-related judgement effects outside of gambling and finance, in healthy adults.

33% of all those who try cigarette smoking are expected to eventually become nicotine-dependent. While several neurocognitive markers reliably predict dependence and treatment outcomes, less is known about indicators of risk for the transition from occasional to dependent nicotine use. Substance-dependent individuals show both heightened sensitivity to reward and diminished sensitivity to punishment, which is associated with an aberrant pattern of enhanced learning from rewarding by comparison with punishing feedback. Critically, few studies have examined occasional use groups, making it unclear whether learning differences observed in dependent individuals are premorbid. The present study examined baseline patterns of learning from rewarding and punishing feedback in a sample of 56 non-daily, non-dependent occasional smokers and 57 non-smoking controls using a novel associative learning task that directly assesses the effects of immediate, performance-dependent feedback on subsequent learning. A repeated-measures ANOVA indicated all participants learned more effectively from rewarding feedback, with no significant main effect of group or significant interaction between group and feedback condition. Our results highlight the heterogeneity of the occasional smoking population, with preliminary evidence of a relationship between duration of nicotine use and learning performance. Follow-up assessments over the next two years will examine potential differences in reward and punishment learning that may exist between groups, and whether these differences predict transition to dependence among this sample of occasional smokers.

Cognitive inflexibility is a prominent feature of autism spectrum disorder (ASD), and is thought to contribute to the “restricted and repetitive patterns of behaviour” symptom cluster. Cognitive flexibility can be modelled experimentally using the reversal learning paradigm, which involves a visual discrimination task where subjects must learn to respond according to the opposite, previously irrelevant, stimulus–reward pairing. This paradigm appears to be underpinned by neural circuitry comprising dorsomedial and ventrolateral prefrontal cortices (dmPFC and vlPFC), although the precise contribution from each is somewhat unclear. The aim of the study was to determine, in a healthy adult population, whether the disruption of dmPFC and vlPFC using high definition transcranial direct current stimulation (HD-tDCS) affects the performance in cognitive reversal learning skills. Forty healthy individuals underwent one session in which HD-tDCS was applied to temporarily disrupt dorsomedial prefrontal cortex (n=10), ventrolateral prefrontal cortex (n=10), or sham (placebo) stimulation (n=20). Reversal learning was assessed immediately after HD-tDCS. It is anticipated that the results of this study will help to clarify the respective roles of dmPFC and vlPFC in cognitive flexibility. From a clinical perspective, modulating of cognitive flexibility via prefrontal HD-tDCS raises the possibility of therapeutic brain stimulation applications to improve restricted and repetitive patterns of behaviour in ASD.

The ‘SNARC’ (Spatial Numerical Association of Response Codes) effect refers to the finding that speeded decisions about numbers – e.g. is the number ‘3’ odd or even? – are made more quickly with the left hand for (relatively) smaller numbers and with the right hand for larger numbers. Subsequent research demonstrated analogous effects when judgements were indicated via foot pedal responses or leftward/rightward saccades, demonstrating that the SNARC effect is independent of effector modality. Previous work by our group has revealed a SNARC effect for non-symbolic numbers – i.e. participants were presented with two dot cloud stimuli in sequence and asked to judge whether the second stimulus contained more or less dots than the first, faster responses were obtained when ‘less’ responses were assigned to the left hand and ‘more’ responses to the right hand compared to the alternative response assignment. Here, we extend this finding, showing that the non-symbolic number SNARC effect can also be observed for foot pedal responses. We conclude that the non-symbolic number SNARC effect is also independent of response modality. However, the effect for foot pedal responses was not as robust as that for manual responses.

Predictive coding is the theory that brains are essentially hypothesis testers: they establish top-down generative models that attempt to predict and ‘explain away’ bottom-up sensory input, leaving only what was not predicted—prediction-error, to be processed by propagating forward within the cortical hierarchy. Despite the ubiquitous role that prediction-error signals are thought to play, remarkably little is known about some of their most basic properties. For example, it is not yet known whether the same neural signal is used to encode prediction-errors across all levels of the hierarchy. To address this, we used two electrophysiological brain signatures of prediction-error: the mismatch negativity (MMN), an event-related potential (ERP) sensitive to deviance detection, and the N400, an ERP sensitive to symbolic meaning. To elicit MMN and N400 in a single trial, we presented five prediction-establishing sounds (e.g., “meow”) in rapid succession, followed by either a standard (e.g., “meow”) or deviant (e.g., “woof”), then a written word that was either congruent (e.g., “CAT”) or incongruent (e.g., “DOG”) with the standard (or deviant). We found a MMN—deviants were more negative than standards, at fronto-central electrodes at about 150 ms, and a N400—incongruent words were more negative than congruent words, at centro-parietal electrodes at about 400 ms. Critically, we found that N400 was bigger when preceded by a deviant than a standard, meaning that the MMN prediction-error signal modulates the N400. These results are consistent with prediction-error signals being the brain’s lingua franca: the common computational principle underlying interrelated processes of perception and cognition.

Background. We have around a one in a million chance of dying from an accident every day, and this acute risk is quantified as one MicroMort. We increase our risk through activity choices, and MicroMorts index their relative objective risk of death. We aimed to investigate the accuracy of peoples’ acute risk judgements for activities such as walking and giving birth, and relationships to mental representations of numerical magnitudes.
Methods. The Amazon Mechanical Turk platform: n=284, 40% female, 19-68 years. Participants had to indicate the positions of (1) numerical magnitudes on a horizontal-line ranging from 0 to 1000, and (2), the relative death risk of activities on a horizontal-line with ‘very low risk’ and ‘very high risk of death’ as left and right anchors, respectively. The difference in distance (pixels) between the indicated position and the objective position was calculated for all number and activity/MicroMort stimuli.
Results. Accuracy on the MicroMort task (risk judgements for activities) was strongly related to accuracy on the symbolic-number mapping task. Participants under-estimating the location of a numerical magnitude also underestimated the risk of an activity, and vice versa for over-estimations. On the whole, participants over-estimated risk for low-risk activities such as walking, and under-estimated risk for high-risk activities such as basejumping; prior experience was associated with lower risk judgements.
Conclusion. An individual’s perception of risk for any given activity is related to the mental number line. The research raises the question under which circumstances basic conceptions of numbers contribute to everyday judgments and decisions.

Healthy ageing is associated with changes in the efficiency with which response processes involved in preparation and selection of an appropriate response are engaged. These response processes can be measured by manipulating the information provided by and the relationship between cues and targets. These processes have been extensively studied using behavioural and event-related potentials measurements. However, we know less about how these processes are affected by age-related changes in the frequency domain. In this study, we examined whether oscillatory activity associated with response preparation, selection and execution changes with ageing. We assessed task-based oscillatory activity of 24 younger and 33 older participants while they completed a cued go/nogo task with informative and uninformative cues. Time frequency analysis was performed across a power band spectrum from 2-30Hz, encompassing the upper delta, theta, alpha and beta frequencies. Behavioural results showed that younger adults responded approximately 100ms faster than older adults. However, there was no interaction with cue condition, indicating that ageing does not differentially affect prepared and unprepared motor responses. Time-frequency analysis showed that older adults elicited increased delta-theta and alpha power across both response preparation (cue-locked) and response selection/execution (post-target) time windows. However, comparison between the cue conditions revealed that there was no effect of preparation on the oscillatory results. These findings are discussed in the light of age-related changes in ERP components during both the preparation and the response execution interval.

Chronic cannabis users are known to have impaired decision-making abilities, and have been shown to be less influenced by the magnitude of loss than non-users. However, to date, studies investigating brain activity associated with risk-based decision making have not typically dissociated probability of risk from magnitude of reward. In addition, research investigating whether the outcome of a previous decision affects future performance has been limited.

This study will use a novel cognitive task during electroencephalography (EEG) to determine whether there are any task-related differences in 15 cannabis users compared to 15 non-drug using control participants. EEG activities were recorded whilst subjects completed a novel risk-based task, using a 64-channel electrode cap. Event-related potentials (ERPs) and event-related spectral perturbations (ERSPs) were computed. The two groups were compared in respect to their EEG data and behavioural performance on the task.

EEG data were analysed using EEGLAB. Reaction time (RT) was compared between groups using ANOVA in SPSS. The influence of previous outcomes on future RT was also assessed. Data analysis showed a significant difference in cannabis using participants in comparison to non-using subjects in oscillatory activity. There were no significant differences in overall RT between groups; however, the effect of a positive or negative outcome on future responses did show differences.

This study shows differences in activity during the evaluation of risk between cannabis using and non-using control subjects. This finding is in line with previous studies that have shown that chronic cannabinoid exposure can alter the ability to generate neural oscillations.

Background: Inhibitory control has been linked to risk-taking behaviours. We aimed to investigate the association between inhibitory control, as indexed by N2 and P3 ERP components, and behavioural measures of risk-taking.

Methods: Thirty healthy participants (16 female, mean age = 23.56, SD = 4.85) completed a go/no-go task (30% no-go stimuli) during EEG acquisition. ERPs were calculated relative to successful go and no-go trials, and mean amplitudes of the N2 and P3 extracted. Participants completed the Domain Specific Risk-taking Scale (DOSPERT), the Barratt Impulsiveness Scale (BIS-11) and the Cambridge Gambling Task (CGT). Linear mixed-effects modelling was used.

Results: As expected, the nogo-N2 and nogo-P3 were significantly larger than the go-N2 and go-P3. DOSPERT measures (risk taking, risk perception, and expected benefits) did not significantly associate with ERP responses during a go-nogo task. Furthermore, there were no associations between ERP responses and the CGT or the BIS-11. Analyses revealed no significant correlations between the DOSPERT subscales, the CGT or the BIS-11.

Conclusion: We saw no associations between behavioural measures of risk-taking and psychophysiological measures of inhibitory control. Despite consistent suggestions that poor inhibitory control may contribute to risk-taking behaviour, this psychophysiological study provides little support for a relationship between ERP correlates of inhibitory control and two behavioural measures of risk-taking.

Twenty-four participants responded to red, green, or yellow traffic lights using foot pedals or verbal commands during single or dual task conditions. An analysis of reaction time revealed a main effect of Task [F(1,25) = 4.58, p <.05] and a Modality × Traffic Light interaction [F(2,50) = 17.42, p < .05]. The former was explained by slower responses during dual versus single task. The latter was explained by slower responses to yellow relative to red relative to green lights during foot responses (all p < .05) but not during verbal responses (all p > .22). To analyse people’s tendency to stop and the associated variability in their responses, we first calculated the probability of stopping for each of the different traffic lights and then fitted this data into a logistic function. After the logistic function was calculated for each individual, we then determined the point of subject equality (PSE), which served as an index of conservativeness, as well as the width between the 25th and 75th percentile of the logistic function (ω), which served as an index of uncertainty. Although no differences were found in PSE between conditions [χ²(3) = 7.47, p = .06], greater levels of uncertainty were found for foot pedal relative to verbal responses in the single (p < .05) but not in the dual (p = .21) task. Our results show how people are slower and more indecisive when they respond to yellow traffic lights with foot pedals compared to verbalising commands.

The time taken to specify the number of dots in a group increases linearly for set sizes above four. Eye movement patterns during enumeration have been proposed to reflect counting strategies that utilise natural clumping in the randomly arranged stimuli used in number research. However, eye movement strategies are ignored as a contributing factor to dot enumeration times. We investigated the contribution of eye movements to enumeration times by manipulating the ability to group dot arrays. 60 undergraduate participants enumerated sets of dots numbered from 1 to 16. Stimulus dots were arranged randomly (unconstrained grouping), in a hexagonal lattice (ungroupable) or constrained in a quad arrangement (dots distributed evenly to one of four clusters). We expected the response times to be greatest for ungroupable stimuli because there were no clusters to guide eye movements. We expected response times for stimuli arranged into four clusters to increase minimally with set size, assuming the quad grouping would promote a county strategy using four eye movements. Instead, we found that enumeration times for ungroupable stimuli were similar to randomly arranged dots. Furthermore, enumeration times for quad clustered stimuli increased non-monitonically with set size. Local response time minima were observed when quad clusters had the same number of dots. Local maxima were observed for most variable quad cluster sizes. Our data show that the form of the enumeration time curve depends on stimulus arrangement and that enumeration times are dependent on the degree to which eye movement strategies are constrained.

Advances in the neurocognitive understanding of autism spectrum disorder (ASD) can be made by drawing on Bayesian theories of brain function. An emerging hypothesis in this regard is that ASD is characterised by an increased reliance on new sensory data relative to prior expectations in the information processing that underlies perception and cognition. In predictive processing models of brain function, this equates to a greater weighting of prediction error in updating predictions about the environment. Bayesian theories can account for a variety of cognitive and sensory characteristics of ASD; however, empirical data that directly tests the principal hypotheses of these accounts has only just begun to emerge. The present study examines behaviour in a spatial prediction task in which prediction errors and prediction updates can be directly quantified. Participants were 40 adults recruited from the general population, rated in terms of their autistic traits using the Autism-Spectrum Quotient. We report evidence that autistic features across the general population are not related to differences in the average weighting of prediction errors, nor in how this weighting is adjusted as the precision of feedback data changes. These findings are valuable in constraining the aspects of predictive processing mechanisms that are likely to be atypical in ASD. One implication is that differences in information processing mechanisms that characterise autistic perception may not extend to more explicit statistical learning; another prospect is that the gain on prediction errors in the autistic brain is characterised by a more subtle atypicality than chronic over-weighting.

Individuals who can quickly and accurately enumerate visual arrays, especially of small numerosities (subitizing), tend to have greater mathematical proficiency; however, the precise nature of this relationship is not well understood. One plausible hypothesis is that differences in visual enumeration abilities, and ipso facto math abilities, reflect differences in the capacity to efficiency scan visual arrays for small sets. To test this claim we analysed eye movement patterns of 21 adults as they completed two visual enumeration tasks. In separate blocks, participants reported either the number of elements in a visual array or the apparent number of sets of elements. Arrays comprised 1 to 12 elements presented at locations sampled with equally probable density across a central circular region (15º diameter, >1º element spacing). Arrays of each set size were presented at four orthogonally rotated orientations. We examined spatial and temporal gaze distributions to determine whether scan patterns were element based or invariant to global image statistics. Behavioural results confirmed that higher saccadic frequency in the small number range was related to greater response variability for larger numerosities. Comparison of fixation density maps revealed salient locations based on perceptual clusters rather than global image locations. Geometric scanpath analyses showed individual differences in the efficiency and similarity of scanning behaviours. Simulated gaze patterns were derived from image statistics and parameters based on plausible assumptions of the visual system, which were compared to neurocognitive models of other strategic perceptual tasks such as visual search.

Cognitive impairment is a core feature of schizophrenia (SZ) that is unrelated to psychotic symptoms, exists in non-ill relatives of patients, and is not markedly improved by antipsychotic medication. Deficits are consistently observed in executive control functions that present independently from a general intellectual impairment. In this study we used magnetoencephalography (MEG) to probe dysfunction of a core executive control function, response inhibition, by comparing SZ patients to healthy matched controls (HC). Response inhibition was operationalized using the stop-signal task that required participants to make reaction time responses to visual ‘go task’ stimuli and the attempt to stop when stop-signals (tones) were presented at a brief ‘stop-signal delay’ after go stimulus onset. Participants performed the stop-signal task during 4 blocks of stimulus presentation (160 trials per block, 25% stop-signal trials) that included dynamic adjustment of the stop-signal delay to yield a 50% inhibition rate. MEG data were recorded on an ELEKTA Neuromag TRIUX machine at Swinburne University of Technology. Previous work has shown that SZ have slowed inhibition processing speed, termed the stop-signal RT (SSRT), perhaps deriving from an inability to engage the right inferior frontal gyrus (rIFG) that is critical for efficient stopping. Hence the focus of this work is upon induced neuromagnetic activity in rIFG occurring around SSRT. Preliminary analysis of one HC data set has revealed a strong broadband response at a sensor over rIFG peaking about 30 ms before estimated SSRT (238ms).

Neurometabolites measured by 1HMRS have been identified to vary in relationship to cognition. The biochemical processes of cognition have not been extensively explored and may elucidate underlying subtle functional changes before any clear structural brain changes occur in healthy ageing. Preliminary examination between cognition and metabolites in a cross-sectional study of 13 healthy adults (60-75yrs) participating in the Australian Research Council Longevity Intervention study was conducted. The Mini Mental State Exam (MMSE), Wechsler Abbreviated Scale of Intelligence (WASI) and a computerised cognitive test battery sensitive to the effects of ageing were employed. 1HMRS was used to measure metabolite concentrations in the left dorsolateral prefrontal cortex (LDLPFC) and posterior cingulate cortex (PCC). Preliminary results identified no significant correlations between brain metabolites and tasks from the computerised cognitive test battery. However, glutamate (Glu) in the PCC increased with lower IQ scores as measured by the WASI (r = -.564, p < .05). Possibly high levels of Glu may be interfering with task attention, resulting in poor performance. N-acetyl-aspartate (NAA) in the left DLPFC increased with higher scores on the MMSE (r = .627, p < .05). This may be indicative of good axonal integrity. Regression analyses were performed to investigate the effects of the metabolites against WASI and MMSE scores over and above the effects of age, producing ambiguous results. With restricted data in the preliminary results, the outcomes are limited to global assessments of participants’ cognitive function. Once more participant data is available, conclusive relationships between cognition and neurometabolites are expected.

Despite the fact that crying is a ubiquitous form of emotional expression, its functions are still largely misunderstood. It is proposed that the help-soliciting functions of tears are largely dependent on interpersonal factors. In order to understand this phenomenon, it is critical to investigate the anticipated social reactions to adult crying and the discrepancy between person perception and behavioural responding. Participants (N = 263) were randomly assigned to a context condition (employment versus home) and responded to two vignettes of opposing severity. Although the effects of gender have been exemplified in previous research, no work has been conducted utilising androgynous stimuli. Participants ascribed greater gender-stereotypical female traits of emotionality and decreased levels of competence to persons who cried in employment conditions. Furthermore, participants reported that they would offer less emotional support, and ascribe greater negative characteristics to persons who were crying in employment contexts. These results signify the importance of investigating interpersonal factors when determining the help-soliciting functions of tears.

Background: Empathy can assist healthcare professionals in their interactions with patients. This study examined the how views on empathy and pain vary in undergraduate university students completing health and non-health degrees.

Methods: One hundred and four undergraduate students (20 males, M = 21 years old, SD = 1.57) from a major Australian university completed an online survey about mood, empathy for pain, patient-centeredness, and views on health professionals attitude towards chronic pain sufferers. Forty-one participants were completing non-health (NH) degrees; 28 were undertaking health-oriented degrees, but had not undertaken clinical placements (H-NP); and 35 had commenced clinical placements (H-CP).

Results: H-CP students reported significantly less affective distress towards observing persons in pain (p = .005), had lower anxiety (p = .001) and depression (p < .001), and were significantly more patient-centric (p < .001) compared to NH students. Within H-CP students, depression severity was positively correlated with the emotional exhaustion component of burnout (p = .002); subjective competence in treating complex clients was correlated with higher sense of personal achievement (p = .017), and greater emotional sensitivity toward complex clients was associated with greater patient-centred values (p = .006).

Discussion: This study demonstrates that undergraduates in clinical training degrees place high value on patient-centeredness, and are less emotionally distressed at seeing others in pain, compared with students who are studying other degrees (e.g., engineering). These findings emphasise that emotional sensitivity, self-confidence, and empathy for pain are all positively associated in clinical trainees when they contemplate treating complex clients in pain.

Individuals with ASD display difficulties in detecting subtle changes in affective prosody. This may stem from auditory-perceptual deficits, and thus there is new focus on trialling sound amplification devices with this population. Studies using sound amplification have demonstrated improved processing of speech.
So far, this research has involved five children with ASD, six typically developing (TD) children, and twenty TD adults. Monosyllabic neutral utterances as the standard stimulus, and angry, happy, and sad utterances as deviant stimuli, were used to evoke a mismatch response (MMR) to changes in prosody. The children with ASD were assessed before and after a trial period wearing hearing aids.
ERPs from adults showed clear P1, N1, and P2 components and a negative MMR at N2. ERPs for TD children differed from the adult waveform and showed a large positivity, followed by a large negativity. Pre-hearing aid trial, ASD children showed later and larger responses compared to TD. Difference waveforms for emotional utterances overlapped in the ASD sample more than in the TD children. Post-trial, children with ASD showed an enhanced P2, and MMR differentiated more clearly between the three emotions.
Larger and later ERPs suggest hypersensitivity and delayed processing of emotional variation in children with ASD. The lack of MMR differentiation across emotions is consistent with reduced behavioural discrimination of affective differences in speech. Post-hearing aid trial increased emotion differentiation and enhanced P2 in the ERP waveforms suggests that sound amplification coupled with discrimination training made prosodic factors more salient for the children with ASD.

Visual perception of emotional facial expressions has been associated with a range of event related potential (ERP) components. Despite the large corpus of research in this area, the electrophysiology of facial expression processing remains poorly understood. Here, we aimed to investigate the way in which emotional expressions are processed along the expression trajectory. We recorded EEG while participants viewed either faces that were neutral in expression or morphed faces that were 20%, 40%, 60%, 80%, or full-intensity happy or angry faces. Participants were asked to indicate whether each face displayed an emotional or a neutral expression. Preliminary results show that faces rated as emotional elicited a greater positivity in the frontal regions in the 160 – 330ms time window and greater negativity in the occipital regions in the 200 – 500 ms time window. There was also a graded modulation of the P2 in the occipital regions as a function of emotion intensity (20% emotion had the highest amplitude and 100% had the lowest amplitude). Happy images elicited greater positivities than angry ones in the 150 – 250 ms window in both central and frontal regions, as well as in the 400 – 600 ms window in the parietal regions. These data suggest that ERP modulations for different basic expressions are not equal. Furthermore, they indicate that there are systematic differences in ERPs that are modulated as a function of the intensity of emotional facial expressions.

The current study is investigating music- and language-syntactic processing in congenital amusia using electroencephalography (EEG). For the music experiment, 80 western five-note-melodies were created with a piano timbre. They were randomly mixed with the same 80 melodies ending with an out-of-key note. Another 40 melodies were included that contain one note with a deviant instrument (i.e. guitar). Participants were asked to detect these timbre-deviants. For the language experiment, five-word English sentences were presented orally. The final word was either syntactically incorrect, semantically incongruent, or syntactically / semantically “correct”. To ensure they attended to the stimuli, participants were occasionally required to answer questions on randomly selected trials related to the sentence they just heard. Brain activity was recorded using 64-channel EEG. The results showed that syntactic violations in both music and language elicited similar brain responses in normal controls [early right anterior negativity (ERAN) and N5 for the music task and left anterior negativity (LAN) and P600 for the language task]; Amusics with music-syntactic deficit showed impairment to some extent in ERAN and LAN respectively reflecting the early stage of music and language syntactic processing whereas no group difference were observed in later stage (i.e. N5 for music processing and P600 for language processing). Further, amusics showed normal N400 reflecting semantic processing.

Children with specific language impairment (SLI) have difficulties with various aspects of language. SLI affects both comprehension and production, and can affect various components of language such as vocabulary knowledge, morphology, syntax and pragmatics.
The present study was designed to investigate which components of grammar are challenged in English-speaking children with SLI by examining aspects of their answers to WH-questions. Answers to two types of WH-questions (i) What’s the boy doing? and (ii) Who’s wearing a hat?) were elicited from 54 children. Eighteen children with SLI (mean age = 5;3); 18 typically-developing children matched on mean length of utterance (MLU) (mean age = 3;4); and 18 age-matched children (mean age = 5;3) participated.
The experimental findings revealed that children with SLI answered questions using an appropriate syntactic structure across both question types. This implied unimpaired comprehension of the wh-question and its targeted extraction site. However, in contrast to the control groups, children with SLI produced more full sentence answers in contexts when a fragment answer is sufficient. This suggests that the SLI group is not as facile with pragmatic norms as the control children. The greatest difference between the SLI group and the control children was in the omission of the grammatical morpheme for the verb ‘BE’. The SLI group omitted the ‘BE’ morpheme 27% of the time in answers to questions like (i) and 29% for (ii). These findings are consistent with previous research that proposes that the most affected component of the grammar is the morphology.

Recent research has examined how predictive coding, a possible unifying principle of brain function (Friston, 2010), may be affected by age–related changes. Here, we measured ERPs to examine age-related changes in prediction generation versus internal model adaptation following prediction errors. Older adults (60–81 years; n=40) read sentences of the form "The opposite of black is white/yellow/nice". Replicating previous work in young adults (Roehm et al., 2007), results showed a target-related P300 for the expected antonym ("white" – ANT; an effect assumed to reflect a prediction match), and a graded N400 for the incongruous conditions (i.e. larger N400 amplitude for the incongruous continuation not related to the expected antonym, "nice" – NONREL, versus the incongruous associated condition, "yellow" – REL). These effects were followed by a late positivity (NONREL > REL > ANT). Linear mixed-effects models showed that the target-related P300 and the N400 were both modulated by age, suggesting that age-related changes affect both prediction generation and model adaptation. However, age effects were outweighed by the interindividual variability of ERP responses, as reflected in the high proportion of variance captured by the inclusion of by-condition random slopes for participants and items. We conclude that – at both a neurophysiological and a functional level – the notion of general differences between language processing in young and older adults may only be of limited use, and that future research should seek to better understand the causes of interindividual variability in the ERP responses of older adults and its relation to cognitive performance.

Voluntary behaviour can be conceptualised as choosing and scheduling courses of action. When the consequences of actions are uncertain, experiential learning can be used to guide behaviour. Operant conditioning via positive reinforcement is a fundamental process of such learning. We investigated how the environmental context can affect the development of preference. Using a two-alternative-choice procedure, we arranged relative reinforcer probabilities of 1:3 and 3:1. Participants learned the contingencies in two contexts. The Events context consisted of the delivery of contingent stimulus presentations with no intrinsic value. The Gifts context had additional non-contingent reinforcers. Based on the information-theoretic conception of contingency, the Gifts should degrade the retrospective contingency between reinforcer and response while the Events should have no impact on the association. High density electroencephalography was used to assess the differences between the Event-Related Potentials (ERP) evoked by the Events, Gifts and reinforcers. Previous research has identified the reward-positivity as well as the P300 as neural correlates of reward processing, with the amplitude of the reward-positivity linearly predictable from the participants preference, as measured by the log odds of staying versus switching between alternatives. Response preference and stability was quantified using a change point procedure. Preliminary results indicate the reward-positivity evoked by Gifts was more negative than reinforcers and Events, with no evidence for a difference between reinforcers and Events or for the modulation of the P300.

Background: Advancing age is accompanied by a deterioration in memory performance that can profoundly influence activities of daily living. However, the physiological processes responsible for this age-related memory decline are poorly understood. Neuroplasticity is an important factor mediating learning and memory, and studies in animals have shown evidence for reduced neuroplasticity with advancing age. While similar findings have been shown in humans using various plasticity-inducing non-invasive brain stimulation techniques, these previous studies have been limited to measuring plasticity in the primary motor cortex, and so the relevance to memory performance is unclear. Here, we used transcranial magnetic stimulation (TMS) in combination with electroencephalography (EEG) to assess age-related changes in neuroplasticity in the dorsolateral prefrontal cortex (DLPFC), an association area of the human cortex important for memory and other cognitive functions.
Methods: TMS-evoked cortical potentials were recorded before and following intermittent theta burst stimulation (iTBS) applied to the left DLPFC in healthy young (18-30 years) and older (50-85 years) adults. In a separate session and for older participants only, memory performance was assessed using a Paired Associates Learning (PAL) task (Cambridge Neuropsychological Test Automated Battery).
Results: While iTBS modulated the amplitude of TEP peaks in young participants, only modest effects were observed in older participants. Additionally, iTBS-induced changes in TEP amplitudes correlated with enhanced memory performance.
Conclusions: These findings suggest that DLPFC neuroplasticity is reduced with advancing age, and this is may be an important factor mediating age-related memory decline.

Humans experience mental time travel by remembering past events and imagining potential future events. These experiences are represented linearly, on a mental timeline, with past events on the left and future events on the right. Valence also influences how an event is remembered/imagined. Interestingly, valence is lateralised, such that negative information is associated with the left, whereas positive information is associated with the right. We investigated whether valence effects occur when only one side of the mental timeline is represented. In Experiment 1, participants generated positive and negative events, associated with a keyword, occurring in either the past or future. They indicated, using a horizontal slider, when each event occurred, ranging from present to birth/death. Consistent with previous research, negative events were placed further from present and most events occurred in the near past/future. In Experiment 2, we reduced the timeline to the past/next 5 years. Negative events were placed to the left of positive events for both past and future. This result contrasts prior research wherein future negative events are judged as temporally distant (to the right of positive events). We found an identical pattern of results in Experiment 3, when we reduced the timeline to the past/next 1 year. These data have implications for how past and future autobiographical events are spatially represented; specifically, how our mental timeline is influenced by the association of valence with side of space, and by psychological distance, whereby people tend to feel psychologically close to positive events and distant from negative events.

Existing research using event-related potentials (ERP) suggests that emotional content enhances memory recall. By engendering better recall performance, emotional versus neutral stimuli modulate the well-known late parietal ERP effect for old versus new stimuli (e.g. Weymar et al., 2009). Emotional content is also known to affect memory consolidation during sleep (e.g. Payne & Kensinger, 2010). However, little is known about the influence of sleep and emotional information on the neurophysiological correlates of memory recall. The current study pilots a paradigm to be used for a larger physiological study investigating interactions of sleep, memory and emotion. A repeated-measures counterbalanced design was used with 14 adults aged 18-41 years (9 female). Participants underwent a standard old/new memory task using emotionally valenced (positive, neutral, negative) stimuli, before and after periods of sleep and wake. Polysomnography was used to quantify sleep. Overall, the total number of stimuli correctly identified decreased after periods of wake and sleep. Change in number of targets missed from immediate to delayed recall was greater for negative targets compared to positive and neutral, but only in the sleep condition. Aspects of sleep neurophysiology associated with performance measures will be assessed. This pilot data suggests that sleep in healthy young adults acts to reduce learning of negatively salient content compared to neutral and positive. We predict that this effect will manifest itself as a differential modulation of the late positive old/new effect such that the general differentiation between emotional and neutral stimuli only holds for recall in the absence of sleep

Background:
Changes to the neural processes supporting working memory have been observed across the lifespan. This study explored the effects of increasing working memory load on Steady-State Visually Evoked Potential (SSVEP) amplitude and latency, in addition to studying event-related partial coherence, during a spatial working memory task in younger and older adults.
Methods:
Healthy younger (n = 20, mean age 23.15 years) and older adult males (n = 18, mean age 61.94 years) completed a graded spatial working memory task during 60-channel EEG recordings of the 13 Hz SSVEP induced by a task-irrelevant visual flicker.
Results:
Age-related differences were observed in behavioural performance of the task, with older adults demonstrating a significantly greater decrement in response accuracy from low memory load to high memory load compared with younger adults. During the encoding period, older adults showed greater SSVEP amplitude and latency reductions in frontal regions in the high memory load condition, a pattern which was associated with relatively preserved accuracy. Analysis of coherence showed a diffuse pattern of elevated connectivity in the older participants for the higher memory load compared to low memory load. Among older adults, the greater this pattern of increased coherence in the high demand condition, the greater decrement in task performance.
Conclusions:
These findings further implicate frontal brain regions in compensatory neural processes associated with increasing load during working memory task performance in older adults, whilst suggesting recruitment of additional resources via more diffuse increases in functional connectivity are in fact associated with poorer behavioural outcomes.

Hippocampal theta oscillations are integral to spatial learning in animals. A growing body of translational evidence from virtual reality tasks indicates that the human hippocampus also exhibits theta oscillations (4-8 Hz) during goal-directed navigation. Its functional importance in human spatial learning has not been firmly established. Here we aimed to determine with greater specificity the function of human hippocampal theta in a spatial learning context. Healthy participants (N=17, 9 men) navigated a virtual Morris water maze in search of a hidden platform while neuromagnetic activity was recorded with a whole-head 306-channel magnetometer. The availability of allocentric cues (i.e., fixed platform location with a fixed configuration of distal landmarks) to efficiently locate the hidden platform was manipulated across three conditions to determine whether increases in navigation-related hippocampal theta are associated with the kind of spatial cognitive strategies used. Adaptive spatial filtering (i.e., beamforming) source analyses were carried out to estimate regional changes in theta oscillatory power across the brain. Left hippocampal theta was robustly induced during navigation both when allocentric cues were available to guide navigation and when an egocentric strategy could be used. Right hippocampal theta power was more selectively increased during allocentric-based navigation. These results provide some of the clearest evidence to date of the functional importance of human hippocampal theta in spatial learning.

A common single nucleotide polymorphism (SNP) within the gene for brain derived neurotrophic factor (BDNF) is known to impact performance on recognition memory tasks. Previous research has identified two subcomponents of recognition memory, familiarity and recollection. While the neural circuits that underpin each of these processes are related, one major distinction is that recollection is dependent upon the hippocampus. Given that the highest concentration of BDNF is expressed in the hippocampus, it is of interest to examine how genotype for BDNF interacts with each aspect of recognition memory individually. Here we use EEG to dissociate the familiarity and recollection components of recognition memory using event related potentials (ERPs) gathered during a source recognition task. This task is comprised of an old-new discrimination and a source discrimination, allowing us to examine both familiarity and recollection processes. Familiarity has previously been associated with modulations to the N400, while recollection has been associated with a late positive component (LPC). Our preliminary results show differences between the genotype groups for the overall patterns of these two ERP components. Val homozygotes show greater amplitudes of the N400 compared to Met allele carriers, despite both groups showing similar patterns of increased positivity for correctly identified old vs new items. The amplitudes of the LPC is also larger in Val homozygotes compared to Met allele carriers. We therefore suggest that Val homozygotes appear to show greater amplitudes for ERP components that index both familiarity and recollection.

The catechol-O-methyltransferase (COMT) val158met polymorphism, which affects dopamine degradation rates, has been associated with a variety of neurophysiological and behavioural outcomes. Some of the effects of this polymorphism appear to be sex-specific and oestrogen has been shown to down-regulate the activity of the COMT enzyme. Prior studies have reported a female advantage for face recognition, an ability that may be linked to dopaminergic function. This study investigates the effects of sex and COMT genotype on face recognition in a sample of healthy university students (n = 111) genotyped and assessed on the Faces subtest of the Wechsler Memory Scale - Third Edition (WMS-III). Face recognition scores were subjected to a factorial ANOVA with sex and COMT genotype as the independent variables. There was a significant main effect of gender on face recognition (p = .003), with male participants obtaining lower scores than female participants. There was also a significant two-way interaction between sex and COMT genotype (p = .007). Simple effects tests with Bonferroni adjustments were performed. Of the male participants, COMT val homozygotes had significantly lower scores than those with a copy of the COMT met allele (p = .017). Male val homozygotes had significantly lower scores than female val homozygotes (p = .001). The gene-sex interaction found in the current study contributes to the accumulating literature documenting sex-specific effects of the COMT polymorphism on cognition and is consistent with a role for dopamine in face recognition.

Background: Behavioural findings suggest attention influences long-term memory processes; however, few studies have investigated the physiological correlates of attention at encoding of emotional memory, particularly for delayed (>30mins) recall intervals. As part of a larger study investigating the role of sleep on emotional memory, we piloted a paradigm to assess the role of attention during encoding on recall of emotional information. Method: 12 (4 male) healthy adults, aged 18-41 years (M=27.33, SD=7.45), participated in a within-subjects experimental design consisting of encoding and recall sessions involving emotionally valenced stimuli. Attention was indexed at encoding using the Conjunctive Continuous Performance Task – Visual. Results: A repeated measures analysis of covariance revealed a significant attention by recall interaction, p=.03. Follow-up regression analyses revealed attention was moderately correlated with performance at immediate recall, but not at delayed recall. Further, attention was associated with enhanced recall of neutral stimuli, R2=.36, p=.02, but not recall of emotionally valenced stimuli (positive, negative). Conclusions: Enhanced baseline attention was important for recall of neutral, but not emotional, information. This is consistent with recent electrophysiological findings that demonstrate post-stimulus neural processes are important for stimuli that receive heightened attention, such as emotionally valenced stimuli, whereas pre-stimulus attentional processes may be more critical for determining encoding of less salient stimuli. Further analyses will determine whether alpha oscillatory activity serves as a robust physiological index of attention, which may provide insights into how oscillatory dynamics modulate perception and attention, facilitating information into long-term memory.

Researchers are increasingly applying surface Laplacian transformations to electroencephalography (EEG) data to minimise volume conduction effects. These transformations, whilst helpful, have the possibility to change topographical representations of neural signatures. This can make it difficult to interpret commonly extracted ERP components that are usually obtained from different referencing schemes. One such example is task switching, where event-related potentials (ERPs) are typically referenced against either averaged mastoids or common average. However, studies examining time-frequency power and inter-trial phase coherence properties in these same data typically use surface Laplacian transformations. As these transformations may affect the topography of common task switching ERP components, it becomes difficult to compare across different analyses modes. In this study, we examine how different referencing montages affect typical ERP measures in a cued-trials task switching paradigm. We compare several common referencing schemes used with ERPs (e.g., vertex [Cz], common average, average mastoids) to those obtained with surface Laplacian transformations. We also examine how ERP measures quantified from these different references relate to time-frequency power and inter-trial phase coherence data extracted from the same data. This work has the potential to encourage researchers to apply surface Laplacian transformations to task switching data with confidence.

Clinical investigations are crucial to the understanding and treatment of psychiatric disorders; however, most studies draw conclusions from group-based analyses and fail to identify key markers that are relevant to individual patients. We explored the use of spiking neural networks to classify people with schizophrenia based on their response to the antipsychotic clozapine using diffusion tensor imaging (DTI) and resting-state functional magnetic resonance imaging (rs-fMRI).
Methods: Twelve people with treatment-resistant schizophrenia who responded to clozapine monotherapy and 7 people who had previously failed treatment with clozapine (ultra-treatment-resistant schizophrenia; UTRS) and were instead receiving two antipsychotic drugs were included in the analysis.
Rs-fMRI and diffusion-weighted images were obtained using a Siemens 3T Skyra scanner. Functional and diffusion images were pre-processed using the FSL toolbox and transformed into the same dimensions using weight averaged subsampling. A matrix of fibre connections was created from the principal Eigen component, which provided the artificial neural network connection settings for the NeuCube. Classification of participants was then performed using the NeuCube-Multimodal, derived from the NeuCube (1) architecture.
Results: All experiments were repeated 10 times with 50:50 training: testing ratio. An average accuracy of 73.3% was achieved overall, with a standard deviation of 5%.
Conclusion: These preliminary findings highlight the potential use of multimodal imaging for classification. Here, we used DTI outputs to establish initial connections in the NeuCube-M, providing a unique model for each individual, into which their fMRI data was placed. A spiking neural network then classified each participant based on their individual brain data.

Partial least squares (PLS) is a multivariate analysis technique that is highly sensitive to whole-brain patterns of BOLD signal associated with task. However, current versions of software implementing PLS analyses do not allow for controlling of behavioural confounds. Using a dataset in which BOLD differences between two future imagination conditions was mirrored in a number of behavioural variables (subsequent memory, plausibility, detail, novelty and difficulty), we present a technique that regresses out any effects of behavioural scores from the functional data.
To account for potentially confounding effects, we removed any between-subject variance from the fMRI data that was shared by i) differences in BOLD signal and ii) differences in behavioural scores between the two conditions. For each subject, we calculated difference scores for each TR at each voxel, as well as difference scores for the behavioural scores. Each behavioural difference score was then regressed out of the BOLD differences, resulting in a difference between the two conditions for each voxel at each TR that was independent of any undesirable behavioural effects. These residualised difference scores were projected back onto the original data matrix and used for PLS analysis.
Activations before and after residualising were generally similar; however, BOLD differences in some structures (e.g. hippocampus) disappeared after controlling for behavioural scores, suggesting that these effects were dependent on behavioural differences and not the experimental manipulation per se.
Our results highlight the importance of controlling for behavioural confounds in fMRI analyses and we presented a way to achieve this for PLS analyses.

Introduction
We sought to demonstrate resting state functional connectivity in the alpha band using magnetoencephalography (MEG) with ICA and Granger-Geweke Causality (GGC) using the Source Information Flow Toolbox (SIFT) pipeline.
Methods
Participants: This is a three-subject case study. Our participants are three of the 22-25 year-old males who participated in the Human Connectome Project (HCP).
Procedure: MEG data acquisition was performed on a whole head MAGNES 3600 (4D Neuroimaging, San Diego, CA) system at the Saint Louis University (SLU) medical campus (248 magnetometer channels together with 23 reference channels at 509 Hz sampling rate).
Analysis: Preprocessing was done by the HCP. We applied ICA to the middle 2 minutes of preprocessed data. The top components contributing to 10-13Hz were selected for further analysis. Using SIFT, Vieira-Morph multivariate autoregressive modelling (MVAR) with a model order of 8 was applied to the selected components, and GGC was computed.
Results
The top components contributing to alpha activity in the range of 10-13Hz showed topographies that included frontal, central and occipital regions. GGC showed possible functional connectivity between and within these areas.
Discussion
The spatial pattern of alpha activity in these subjects showed some consistency with the resting state literature, suggesting this form of analysis is worth further exploration.

Each time we make an eye movement the retinal image shifts, yet we typically perceive the external world as stable. Predictive remapping mechanisms contribute to this visual stability by anticipating the future position of the eye. Here we used focal transcranial magnetic stimulation (TMS) to test the role of distinct areas of the human posterior parietal cortex (PPC) in predictive visual remapping. In an initial experiment, 20 participants judged the orientation of a target in one of two vertically aligned locations in the left visual field. Targets flashed briefly and were followed by a mask at the same location as the target or at the other location. In separate runs participants either maintained fixation or performed a vertical saccade. We observed a reliable predictive remapping effect for saccade trials only, which manifested as impaired orientation judgements when the target and mask were in different locations, and improved judgements when target and mask were in the same (upper) location. Crucially, these remapping effects were evident in the 100 ms immediately prior to the saccade. To investigate the role of the PPC in predictive remapping, we designed a second experiment in which high-frequency TMS pulses were delivered to the right superior parietal lobule (SPL) or the primary somatosensory cortex (S1; control) 150 ms before saccade onset. Results are pending, but we predict selective disruption of visual remapping for the SPL stimulation site relative to the S1 control site, consistent with the hypothesized role of the PPC in predictive remapping.

How different is colour perception among people, and can the differences in colour perception be explained by individual variations in the ability to discriminate colours? To answer these questions, we asked observers to judge contrast between colours belonging to different colour categories and measured threshold for colour discrimination. Experiments were performed using Mitsubishi Diamond Pro 2070SB monitor controlled by VSG visage (Cambridge system). The stimuli (coloured disks, 2.4° in diameter, presented on achromatic background, 24.5cd/m²) were characterised as vectors in the three-dimensional cone-contrast space. To quantify the ability of individual observers to discriminate colours, we measured colour thresholds at 17 points in colour space located along the “yellow-blue” and “red-green” cardinal directions. For each observer, we used a second order polynomial to describe how thresholds depend on saturation, which allowed us to calculate the separation between points in the colour space in terms of the number of just noticeable distances (JND). We then presented two colour pairs with strong colour contrast, well above the discrimination thresholds, and asked the observers which colour pair contained colours that were more different from each other. Each observer was consistent in his/her judgements. However the perception of strong contrast between colours could not be explained by their difference in terms of JND. Moreover, the analysis of performance of individual observers indicates that observers use observer-specific rules to evaluate strong colour contrast. Our results demonstrate substantial variations in colour perception that cannot be explained by individual variations in the ability to discriminate colours.

The debate whether the human visual area V4 (hV4) is located entirely ventrally or partially split across the dorsal and ventral surfaces, still remains. Winawer et al. (2010) posited that artefact from the Transverse Sinus (TS) distorts the BOLD response in the lower visual field region of hV4 in some subjects, where it appears as though there is a ‘missing’ part, which might be on the dorsal surface. However, no dorsal component of hV4 has yet been found. Puckett and DeYoe (2014) suggest that the TS artefact causes an inversion of the BOLD responses in affected regions, which can be corrected. Here, we tested the hypothesis that the TS artefact can be corrected and hV4 identified as a continuous ventral hemifield in affected subjects. We collected fMRI data from 8 subjects, using rotating bowties, expanding rings, drifting bars and full-field flash stimuli. pRF analyses were conducted on the data using the procedure outlined in Dumoulin and Wandell (2008). A correlation analysis was performed on the full-field data to identify inverted voxels, which were then corrected. Clusters of inverted voxels were found to be present in regions near the TS. Incomplete hemifield hV4 maps were identified on the ventral surface in 7/16 hemispheres prior to correction. Following the correction, a full hemifield hV4 map was identified on the ventral surface in 1/7 hemispheres. We conclude that the venous eclipse may not be sufficient to explain the persisting difficulties in identifying complete and continuous V4 hemifields in all hemispheres.

With over 285 million visually impaired people worldwide there is growing interest in sensory substitution -- a non-invasive technology substituting information from one sensory modality (e.g., vision) with another sensory modality (e.g., touch). Previous work focused primarily on how blind or vision-impaired people discriminate between different types of objects using sensory substitution devices (SSDs). A fraction of work explored whether and to what extent SSDs support precise localisation of objects in space. The few localisation studies found distance errors of around 8-14 cm from the target. We custom built a visual to tactile SSD to investigate the object localisation ability of visually impaired participants. In three separate conditions participants had to point to a white disk presented against a black background on a touchscreen. In the first task the SSD conveyed information only about the disk, in the second task the participant's hand was displayed in addition to the disk, and in the third task a white reference border marking the monitor frames was also added to the display. We found participants were slightly more accurate overall than in previous studies (<6 cm error) but localisation accuracy did not differ across conditions. Participants' responses were slower in the “hand” and “reference” conditions, indicating that the additional information acted like a distractor, rendering the task more difficult. This result suggests that processing of tactile information is more limited, especially when multiple objects are presented in parallel, indicating that filtering of relevant information is critical to enhancing performance of future SSDs.

Background
The gray matter of human cortex is marked by depth-dependent differences in neuronal activity, connections and vasculature. Advances in imaging hardware have pushed the resolution limits of functional magnetic resonance imaging (fMRI) below a millimeter, permitting these differences to be measured in awake and behaving human subjects. We present a prerequisite for such endeavors, a detailed, three-dimensional hemodynamic response function (HRF) reconstructed through the use of submillimeter fMRI.

Methods
High-resolution fMRI data (0.8 mm isotropic) were acquired using a 3T scanner, T1 data were recorded at 0.5 mm. Six parallel surfaces were constructed at the gray/white boundary 0.8mm apart. Subjects viewed a visual stimulus consisting of a thin ring stimulus (1 pixel = 0.03 deg, flickering 250 ms) 1 degree (‘ON’ for 4 seconds) alternated with a uniform gray field (‘OFF’ for 16.5 s). Stimulus onset was varied with respect to the MRI timing (500ms). Spatiotemporal HRF responses were measured by averaging time-course responses based on the distance from the centreline.

Results
A fully three-dimensional HRF is presented describing the response vs. time and two spatial dimensions: tangential and perpendicular to the cortical surface. The response depends significantly on depth. At the gray-white boundary the FWHM of the response is 4.8mm, increasing linearly by 0.6mm in width per mm in depth reaching 6.7mm FWHM at 3.2mm at the surface of the gray matter. Moreover, at all depths part of the response propagates as a traveling wave evident as increasingly delayed BOLD away from the stimulus location.

Although visual information from the retina to the brain is primarily transmitted along the magno-cellular (M) and parvo-cellular (P) pathways, comparatively little is known about the differential levels of M and P functional activation of early visual areas V1, V2d, V2v, V3d, and V3v. Thus an FMRI study (n=9) was conducted. Two experiments were performed: one that used standard retinotopic procedures to independently define the boundaries of each visual area and the other that presented rotating semi-circle wedges that exploited the distinct spatial, temporal, and chromatic characteristics of M and P cells. The M stimulus consisted of a 30% contrast grating of 0.2 cpd flickering at 15 Hz while the P stimulus consisted of a psychophysically isoluminant red/green grating of 2 cpd flickering at 2 Hz. Activation for the M and P biased stimuli were calculated for each ROI. Our results revealed a complex three-way interaction between Stimulus Type (M versus P) × Eccentricity (Fovea versus Periphery) × Stream (Dorsal versus Ventral), strong in V3 and evident but weaker in V2. Namely, M stimulation most strongly activated peripheral subdivision of dorsal visual field subdivisions, while P stimulation most strongly activated the foveal subdivisions of the ventral visual field subdivisions. Such functionally defined spatial biases of fMRI responses in the human seem consistent with the known spatial arrangement of the M and P representations in subdivisions of the early visual areas in the monkey brain.

Multisensory signals can enhance information processing even with task irrelevant sensory signals and cues, suggesting that some aspects of multisensory processing may be automatic. However, the role of awareness in multisensory processing remains elusive. We aimed to investigate multisensory processing in a patient with posterior cortical atrophy (65 years at the time of testing) who presented with severe visuospatial problems and strong aspects of simultanagnosia. We also tested 12 age-matched controls using a simple audiovisual detection task. Participants had to respond to auditory, visual and semantically congruent and incongruent audiovisual presentations of a ‘bird’ and a ‘dog’ with varying stimulus onset asynchronies (SOAs): -300 to 300 ms. Healthy control participants responded to all signals with very high accuracy (M accuracy > 95% for all stimuli). Surprisingly, they also showed large multisensory gains for both congruent and incongruent presentations of multisensory stimuli in close temporal proximity (i.e., SOA < 200 ms). The participant with PCA failed to respond to over 95% of visual only stimuli, and the hit rate when visual signals were present did not differ from the false alarm rate when they were absent. Nevertheless he showed multisensory enhancements in reaction time when the auditory signal was presented first, and, overall, the reliability of motor responses improved. The data show for the first time that awareness of sensory signals is not a pre-requisite for multisensory enhancements. Patients with severe perceptual deficits can benefit from multisensory signals even at an unconscious level. This finding broadens the possibility of new multisensory-based intervention strategies for patients with severe perceptual deficits.

Evidence from neuroimaging and transcranial magnetic stimulation (TMS) studies suggests that foveal cortex plays an important role in peripheral perception of objects. Multi-voxel pattern analysis of fMRI data show that visual information about objects in the periphery is represented in the foveal confluence, a region entirely separate from the feedforward input. Furthermore, when the foveal confluence is disrupted with TMS, participants’ discrimination of such peripheral objects is impaired. In a series of experiments, we developed a psychophysical paradigm to test the effect of this phenomenon on behaviour. Participants performed a difficult discrimination task on target objects presented briefly (100ms duration) in the periphery while fixating centrally. A visual distractor presented at the fovea 100ms after target onset disrupted discrimination accuracy more than a distractor presented at other stimulus onset asynchronies (SOAs), including simultaneous presentation with the targets. We further demonstrated that this effect is location-specific: at 100ms SOA, a foveal distractor impaired discrimination accuracy more than a peripheral distractor. These results are consistent with the claim that foveal confluence is recruited for extra-foveal perception of objects. Follow-up experiments suggest that foveal confluence is also involved in the feedback of colour: a visual distractor presented at the fovea significantly impairs discrimination accuracy when participants performed a similar colour-discrimination task. Taken together, these results suggest that feedback to the foveal confluence is not simply an artifact of peripheral perception, but a component of visual processing essential for peripheral perception.

The appearance of illusory contours, as in the Kanizsa square, is thought to result from the perceptual inference of unseen occluding objects. While the neural correlates of the formation of illusory contours have been previously described, little is known about how new sensory evidence affects the associated perceptual inference. Here, we investigated the event-related potentials (ERPs) evoked by the breaking of an illusory contour, representing the violation of a perceptual hypothesis. We recorded ERPs while participants observed arrays of four inducers which either formed a Kanizsa square or were perceptually incomplete. After one second of static presentation, inducers rotated dynamically so as to either support the percept of an occluding surface (completion-supporting) or break the Kanizsa illusion (completion-violating). Consistent with previously observed correlates of perceptual completion, the initial static presentation of completed Kanizsa squares evoked more negativity than incomplete inducer arrays in lateral occipital electrodes, in the N1 component and in the 250-350 ms time window. Completion-violating inducer motion elicited greater positivity within 200ms in frontal electrodes compared to completion-supporting motion. Pulses in lateral occipital negativity were also noted for completion-violating motion, time-locked to momentary re-completion of illusory contours. Critically, no significant differences between inducer motion types were seen in perceptually incomplete arrays. The scalp distributions of violation-related ERP effects suggest that feedback signals associated with reassessment of perceptual hypotheses and the visual environment may originate in frontal areas and propagate back to visual cortical areas.

Area MT plays a key role in the perception of motion, and one of the proposed functions of neurons in MT is to compute direction of motion in a manner that is invariant to other stimulus features such as contrast, brightness and contour orientation. Single and multi-unit electrophysiology studies have explored response properties of single neurons within MT. This work suggests that the distinct motion sensitivity of area MT takes times to develop. Here, we explored how the population encoding of direction evolves over time. Using multivariate classification analysis, we determined the amount of information about the orientation and direction of a simple moving grating stimulus that was present in the spiking responses of populations of neurons in area MT of marmoset monkey, measured using a multi electrode array. We found evidence for an early, contour-orientation dependent response to grating direction, along with a later contour-orientation invariant response, with distinct tuning for stimulus spatiotemporal frequency.

The mechanism of action of anaesthetic-induced analgesia is poorly understood. Increasing this understanding could aid in the search for a reliable objective pain measure. Such a measure could decrease our reliance on the subjective self-reports of pain. There is currently no published research that involves analysing oscillatory magnetoencephalography (MEG)/electroencephalography (EEG) activity in sedated patients who are exposed to painful stimuli. In the current study, participants were administered either propofol, dexmedetomidine, or a placebo, in a cross-over design, and a cold pack was used to implement a cold pressor test (thought to resemble postoperative pain). The cold pack was left on for one minute past the participant’s individual pain threshold in five trials, and was alternated with a warm pack as a recovery period. Oscillatory MEG activity was compared between the cold pain periods and the warm pack periods, using a nonlinear beamformer analysis to localise oscillatory power changes in the brain in pre-defined frequency bands. It was found that both propofol and dexmedetomidine administration resulted in a significant lessening of the decrease in alpha (8-13 Hz) and beta (13-30Hz) oscillatory activity seen during pain, in occipital, temporal, and parietal areas, as compared to placebo. There were no significant differences found between the two drugs. Further analysis will map the temporal evolution of the changes in regions of interest. This study confirms the usefulness of MEG in the study of pain and anaesthesia.

Perceptual adaptation studies show that prolonged exposure to a face (adaptor) typically results in a contrastive face aftereffect (FAE) where a subsequently presented face appears less like the adaptor. As a process recruiting perceptual brain areas, visual imagery is expected to generate similar FAEs. Recent studies of imagery adaptation to facial gender have however yielded inconsistent results. While some experiments report contrastive FAEs, other experiments report no effect or atypical (i.e. non-contrastive) FAEs. In a recent study, D’Ascenzo et al. (2014) observed atypical FAEs, in which androgynous faces appeared more feminine after imagining female faces of recently familiarised strangers than after imagining male faces. Our study aims to replicate this observation and to investigate the effect of familiarity on adaptation to perceived and imagined facial gender. We found stronger adaptation for celebrities than unfamiliar faces in both perception and imagery tasks, and no evidence of atypical FAEs for imagery. These findings suggest that familiarity may have a modulatory effect on adaptation to perceived and imagined facial gender.