When asked to indicate the midpoint of a horizontal line, most people perceive the midpoint slightly to the left of true centre. This mis-bisection is thought to reflect a bias of attention towards the left hemispace, also known as pseudoneglect. While pseudoneglect is commonly explored with tasks such as line bisection, it has not been applied to some other common attentional tasks. The current study explored whether targets on the left side of an array are detected better during a search task. Participants (n=24) searched a 60-item array for an inverted triangle. To investigate the effect of overt eye movements, the arrays were presented for either 200 or 4000ms. Targets on the left side were detected more accurately and faster than targets on the right for long presentations, but not for short presentations. The possible contribution of asymmetries in attention or reading habits is discussed.

In object individuation, spatiotemporal episodic cues are used to register an object as a distinct perceptual event, relative to other stimuli. The present study examined the time course of individuation to assess the information processing stage at which it arises and the extent to which it draws on early sensory cortices. Previous research has failed to provide definitive evidence on these issues, as individuation manipulations have co-varied with other low-level physical differences (e.g., hue, luminance). Similarly, no previous study has had both high spatial and temporal resolution, and consequently has been unable to speak directly to both the neural and cognitive substrates of individuation. Here we used a multimodal EEG/fMRI approach and a novel enumeration paradigm that equated physical stimulus properties across conditions. We could therefore pinpoint the time window and visual sensory regions associated with individuation for items in attended and unattended locations, as indexed by set-size-dependent changes in activity. P1 (100-140 ms) and N2 (185- 250 ms) event-related potentials both increased in amplitude with the number of attended targets, but unattended non-targets only modulated the N2. Using fMRI, we found that V2 and other extrastriate visual areas were the likely source of the early P1 effect observed for attended targets. Thus, object individuation appears to arise at a perceptual stage of processing and can be detected in early sensory brain regions. In addition, individuation of unattended items occurs at a later stage of processing than attended targets, suggesting a role of selective attention in this operation.

Healthy individuals direct more attention to the left side, mirroring the rightward attentional biases seen amongst hemispatial neglect patients. Interestingly, upper space distractors accentuate leftward attentional asymmetries. Prior work has made use of constant visual stimulation when examining distractibility and the current work examined what occurred in the presence of brief distractors. In Experiment 1, a stronger leftward bias occurred when a flash of light occurred in upper or lower space, compared to a baseline condition where flashes occurred across both visual fields. Interestingly, a significant number of participants consistently demonstrated rightward biases. A second experiment included a baseline block of trials without distractors, which allowed participants to be separated based on their baseline direction. Upper space distractors, compared to lower space distractors, were found to exacerbate attentional asymmetries in the direction of the baseline bias. These findings illustrate the importance of considering individual differences in pseudoneglect, as upper space distractors led attention to be directed more strongly toward either the left or the right side, depending on the direction at baseline.

What goes wrong when participants make mistakes? Multivoxel patterns of fMRI activity in frontoparietal cortex code a range of task features when participants successfully perform tasks, but can we use this method to discover the reasons for errors? Participants performed a challenging stimulus-response task which resulted in moderate error rates. We used participant responses to categorise these as ‘rule’ errors, where participants used the wrong stimulus-response mapping rule, or ‘other’. We trained a pattern classifier on correct trials to identify the neural signature of correctly encoded stimulus and rule information and then used this classifier to examine the representational content of frontoparietal and visual cortex on error trials. We found a striking double dissociation: When participants made rule errors, stimulus information was correctly encoded but patterns in frontoparietal cortex consistently represented the incorrect rule. On ‘other’ errors, frontoparietal activity patterns represented the correct rule information, but the incorrect stimulus information. The brain data therefore suggest that the ‘other’ errors were due to incorrect perception of the visual information. Interestingly, visual stimulus information was missing from the visual cortex on these trials, despite being strongly encoded on rule error trials. Neither the correct nor the incorrect stimulus was represented here, suggesting that the incorrect stimulus information in frontoparietal cortex may have been internally generated. The data validate a method for decoding neural representations during errors. Rather than reducing to noise, frontoparietal patterns of activation systematically encode the incorrect stimulus or rule information, predicting the particular error the person will make.

Mechanisms of attention allow us to focus on task-relevant events under conditions in which multiple stimuli compete for selection simultaneously. Such selection mechanisms are subserved by a network of brain areas, including the prefrontal and parietal cortices. When these regions are damaged unilaterally due to stroke, patients often fail to detect contralesional stimuli in the presence of competing ipsilesional distractors, a condition known as extinction. The extinction phenomenon is thought to reflect a pathological bias in interhemispheric competition and mutual inhibition. Here we investigated the role of the inferior parietal cortex – specifically the supramarginal gyrus (SMG) – in resolving competition in a novel auditory attention task, using repetitive transcranial magnetic stimulation (rTMS) in healthy participants. Before and after rTMS, participants listened to concurrent streams of phonemes presented to each ear, and monitored for occasional targets in the left-ear only, right-ear only, or in both ears (divided attention). In separate sessions, participants underwent 20 minutes of 1 Hz (inhibitory) rTMS over one of three right hemisphere sites: the SMG, the superior temporal gyrus (STG) or the somatosensory cortex (S1). In the critical divided attention condition, inhibitory stimulation of the right SMG significantly reduced detection of contralateral targets, and marginally increased detection of ipsilateral stimuli. There was no such effect for the STG or S1 stimulation sites. The findings point to a causal role for the right SMG in auditory attention, and suggest that this region is crucial for selecting between competing auditory events under conditions of divided attention.

The evolutionary significance of reptiles and fire-arms may moderate attention to images that feature these stimuli. Late-occurring event-related potentials (ERPs), such as the late positive potential (LPP), are shown to index motivational salience, whereas earlier ERPs are more sensitive to the perceptual qualities of an image. To examine the contribution of evolutionary significance to picture processing we compared central N1 (90-150ms) and parietal LPP (90-150ms) activity elicited by a modified Flanker task. Congruent and incongruent arrays were constructed using high threat or neutral images of reptiles (attacking snakes, turtles) and fire-arms (armed hand-guns, water-pistols). EEG was recorded from 64 electrodes for 42 participants (21 male). N1 activity for high threat distractors with semantically-similar targets (i.e., turtles or water-pistols) were more negative than for flanker arrays in the same congruency condition with neutral distractors and high threat targets (i.e., snakes with turtles, hand-guns with water-pistols). Congruent arrays with snakes or hand-guns also evoked more positive LPP activity than those flanker arrays with neutral distractors and high threat targets of the same stimulus type. This effect did not occur for snake distractors paired with turtle targets, or hand-gun distractors shown with water-pistol targets. These results indicate that snake and hand-gun distractors could be prioritised during early perception, and require less elaborative processing as targets when shown with semantically-similar neutral distractors. The threat level of these stimuli may mediate perceptual processing and motivational salience when featured in unpleasant images, but not as a function of their stimulus type.

During visual search, observers can prioritize items on the basis of elementary features such as color and shape. Such feature-based selection is assumed to involve coordinated interactions between frontoparietal areas and early visual cortex, but it remains unclear how this selection evolves across time and cortical regions. Here, we combined a rapid event-related design with high-density functional near-infrared spectroscopy (fNIRS) to examine hemodynamic responses elicited by task-irrelevant target- and distractor-colored probe stimuli during visual search. Compared with target-colored probes, distractor-colored probes increased oxygenated hemoglobin within right hemisphere frontoparietal regions previously linked with feature-based selection. Feature-based selection within early visual cortex showed the opposite pattern: decreases in oxygenated hemoglobin for distractor-colored probes compared with target-colored probes. Feature-based selection occurred earlier in frontoparietal cortex compared with early visual cortex, and functional connectivity between frontoparietal and visual cortices predicted perceptual sensitivity. Together, the results indicate a delayed suppressive influence of right hemisphere frontoparietal regions over visual cortical responses to distractor features during visual search.

In the Simon task, participants respond faster when the spatial location of a target and response are compatible as opposed to incompatible, even though stimulus location is irrelevant to the task. This ‘Simon effect’ is typically attributed to response priming processes that are evoked by the task-irrelevant (spatial) information, which either facilitates or competes with the task-relevant (colour) information. The primary objective of the present study was, therefore, to ascertain the temporal dynamics of motor cortex excitability in a Simon task for both task-relevant and task-irrelevant information. We also explored whether the bilingual experience will affect the time point at which the task-relevant and task-irrelevant information are activated in the Simon task. To investigate these aims, we recorded motor evoked potentials (MEPs) from the responding effector (i.e., correct finger) and non-responding effector (i.e., incorrect finger) as participants completed a Simon task. Single-pulse TMS was delivered over the left motor cortex at 4 different stimulus onset asynchronies (50 ms, 100 ms, 150 ms, 200 ms) relative to target onset. By recording MEPs simultaneously from the responding and non-responding effectors, we have shown that task-irrelevant spatial information is activated early (150 ms) and is followed by the activation of task-relevant stimulus attributes after 50 ms. Further, the lack of a difference between bilingual and monolingual participants challenges the notion of bilingual advantage in the early motor preparation stage (~200 ms).

Motor adaptation can affect spatial attention, as shown by prism adaptation in healthy adults and patients with spatial neglect. This study investigated whether visual spatial attention is modulated by adaptation to force fields that push movements to one side.

Forty right-handed, healthy participants were divided into two subsamples based on force field direction; with movements perturbed either left or right. Spatial attention was assessed by two tasks (temporal order judgement (TOJ); visual target detection (VTD)). The following procedure was applied: 1.) Pre-tests of both attention tasks, 2.) Adaptation to a velocity dependent force field that was gradually introduced, 3.) Post-test of one attention task; counterbalanced order, 4.) Adaptation to the same field at full strength, 5.) Post-test of the other attention task. Dependent measures were the perceived point of simultaneity for the TOJ task, and a reaction time asymmetry index for the VTD task.

Preliminary results from 24 participants show a Force Field Direction (left, right) x Time (pre, post) interaction, F(1,22)=9.058, p=.006, ηp²=.29) for asymmetry indices in the VTD. Posthoc tests indicate a rightward shift of spatial attention for the group that adapted to a rightward force field (p = .009, pre: 8%, post: 0% left asymmetry), and a non-significant trend toward a leftward attention shift after leftward force field adaptation (p = .180, pre: 1%, post: 5% left asymmetry). No effects were found for the TOJ task.

These preliminary data raise the possibility that force field adaptation might lead to directionally specific changes in visual spatial attention.

During reading, the reader’s eyes make several saccades, separated by fixational pauses where visual information is acquired. The regions of focal attention generally sweep more to the right of each fixation (right-visual-hemifield; RVF) when reading left-to-right scripts such as English and more to the left (LVF) when reading right-to-left scripts, such as Farsi. Since visual attention is mandatory for reading, we hypothesized that visual attention may be allocated asymmetrically depending on the direction of reading.

Three groups of participants – left-to-right (English), right-to-left (Farsi) and bidirectional readers (fluent in both languages) were tested on a serial search task, where the stimuli were presented on the right or the left half of the monitor.

Left-to-right readers performed significantly better for targets presented in the RVF, while right-to-left readers performed better for targets presented in the LVF. Bidirectional readers showed no significant asymmetry.

To investigate short-term effect of reading on attention, bidirectional readers were recruited for a pre-reading conjunction search task (stimuli randomly presented either to the left or right of monitor) and a 30-min silent reading of English or Farsi followed by the conjunction task again. Based on the result of the pre-reading conjunction task, Farsi or English was chosen for those who showed poorer performance in LVF or RVF respectively.

Visual attention improved significantly in both groups in both the reading (LVF for Farsi readers and RVF for English readers) and the non-reading hemifields.
This finding has important implications for improving reading performances in populations with attention related impairments.

Mechanisms of visual attention allow us to select behaviourally relevant stimuli while filtering out irrelevant distractions. Despite an extensive literature on the cognitive and neural processes responsible for selective attention, little is known about how people attend when competing stimuli must be approached or avoided, as occurs in many real life situations. We used electroencephalography (EEG) and frequency tagging to measure neural activity while participants manoeuvred a cursor amongst a large set of moving objects. In Experiment 1, participants were rewarded for pursuing objects in one colour and simultaneously punished for failing to avoid objects in a different colour, with a third set of coloured objects serving as neutral distractors. In Experiment 2, approach and avoidance were assessed in separate blocks of trials with competing low- and high-value rewards or punishments. Across both experiments, approached, avoided and neutral objects were tagged at unique flicker frequencies, producing steady-state visual evoked potentials (SSVEPs) that were recovered using frequency-based analyses. In Experiment 1, approached objects elicited larger SSVEP amplitudes than neutral and avoided objects, which produced similar amplitudes. In Experiment 2, SSVEP amplitudes for approached objects scaled positively with value. High-value reward objects produced larger amplitudes than low-value reward objects, which in turn produced larger amplitudes than neutral objects. In contrast, SSVEP amplitudes for avoided objects did not scale with value. Instead, relative to neutral objects, both high- and low-value avoided objects were similarly suppressed. These results link approach behaviour with value-weighted neural enhancement and avoidance behaviour with value-independent neural suppression.

Attention, once considered the ‘gateway’ to conscious access, has recently been re-conceptualised as a distinct process to that of awareness. Attention is not a unitary construct, however, and evidence that endogenous spatial attention can dissociate from awareness is relatively limited. Here we present data from a recent experiment that investigated the effect of spatial attention on neural responses to visible and invisible stimuli. Participants (N = 23) were presented with two flickering image streams on either side of fixation, each of which contained a visible or invisible signal embedded in dynamic noise. The task was to count the number of brief contrast decrements in one of the two image streams across each 10s trial. Visibility of the signal was titrated and checked using a two-interval forced-choice detection task. Steady-state visual-evoked potentials (SSVEPs) were computed from electroencephalography (EEG) data at the signal and noise frequencies of interest. A significantly larger neural response to noise stimuli at the attended location relative to the ignored location confirmed that the attention manipulation was successful. Cluster-based permutation analyses revealed a significant neural response to both visible and invisible signals over occipital electrodes. In line with previous findings, spatial attention increased neural responses to visible signals embedded within noise. Crucially, spatial attention also increased neural responses to invisible signals. As such, our findings support theories suggesting that attention and conscious awareness are distinct and dissociable neural processes.

Attentional bias towards threat can be demonstrated by enhanced processing of threat-related targets and/or greater interference when threat-related distractors are present. These effects are argued to reflect processing within the orienting and executive control networks of the brain respectively. This study investigated behavioural (RT) and electrophysiological correlates of early selective attention and top-down attentional control among high (n=15) and low (n=13) spider fear participants. Participants completed a novel flanker go/nogo task in which a central schematic flower or spider stimulus was flanked by either congruent or incongruent distractors. Participants responded to green stimuli (go trials) and withheld response to yellow stimuli (nogo trials). High fear participants demonstrated significantly shorter reaction times and greater P1 amplitude to spider targets, suggesting specific hypervigilance towards threat-relevant stimuli. In contrast to predictions, there was little evidence for behavioural interference effects when distractor stimuli were threat-relevant. However, there was some preliminary evidence for differences in attentional control as indexed by Nogo N2 amplitude.

Faces capture attention automatically, even when completely irrelevant for one’s current goal (Devue et al., 2009, 2012). Threatening faces have a high socio-biological value and can also bias attention. However, it has not been clearly established that they draw attention in a bottom-up fashion. We used a visual search task to assess whether irrelevant angry faces attract and/or retain the eyes more than neutral faces. Participants made a saccade towards a uniquely coloured circle among six circles and six irrelevant objects. One of the six objects was always a butterfly, a neutral or an angry face. They were located next to the target circle, allowing to assess guidance, or at another location, to examine capture. Neutral and angry faces captured the eyes more than butterflies, although neutral faces more so than angry ones. After capturing the eyes, both faces were fixated as long as a butterfly. Neutral, and to some extent, angry faces presented next to the target guided the eyes to their location: they decreased the number of saccades and time necessary to reach the target and helped to initiate the saccade towards the target faster compared to butterflies. In that case, faces were both fixated more briefly than the butterfly. These effects were decreased or abolished with inverted faces. Therefore, threat in itself does not capture attention but upright faces are selected automatically because of the potentially important information they contain (e.g. threat, identity). Emotion is decoded once the face is selected and brought into foveal vision.

Emotional stimuli compete more effectively for our attention than neutral stimuli. Prioritised emotional information usually guides behaviour in adaptive ways. However, when emotional stimuli are irrelevant to our current goals, they can be distracting. To inhibit distractors, people flexibly shift between more and less effective cognitive control strategies. Because more effective strategies require more effort, motivational factors may influence the strategy that is used. In non-emotional contexts, rewarding successful task performance encourages effective control by guiding attention towards goal-relevant and away from goal-irrelevant stimuli. However, it is unclear whether rewards can similarly increase control over task-irrelevant emotional stimuli. In the current study, task-irrelevant neutral and emotional distractor images were presented peripherally while participants completed a simple letter discrimination task near fixation. In a between-subjects manipulation, one group was rewarded with money for fast and accurate task performance, while the other received no reward. Distraction was measured as the difference between distractor-present and distractor-absent trial RTs. Both groups showed more distraction by emotional images (positive and negative) than neutral images. Importantly, reward reduced distraction by neutral and emotional images equally. Thus, motivation towards the task goal enhanced sustained cognitive control of both neutral and emotional distraction. Results will be compared to an ongoing study in which the possibility of reward is cued from trial to trial. This manipulation will show us whether people can dynamically shift cognitive control over emotional and neutral distraction. Motivational manipulations may be a fruitful way of enhancing goal-relevant processing while inhibiting distraction from irrelevant emotional information.

Background: Cognitive control deficits, as captured by inhibitory control measures, are indicative of increased impulsivity and are considered a marker of substance use disorder vulnerability. While individuals with alcohol use disorder (AUD) typically exhibit inhibitory control dysfunction, evidence of impaired inhibitory control among harmful drinkers, who are at increased risk of developing an AUD, is mixed. This study examined response inhibition in binge drinkers using a task that employed neutral, as well as immediate and delayed rewards, to determine whether reward induced heightened impulsivity in this population.

Methods: Binge alcohol users (n = 42) and controls (n = 42) were administered a Monetary Incentive Control Task that required participants to successfully inhibit a prepotent motor response to both neutral and immediately rewarding stimuli in order to secure a large delayed reward.

Results: Binge drinkers had significantly worse response inhibition than controls irrespective of trial condition. Though binge and control participants exhibited significantly worse inhibitory control in the presence of immediate reward, the control group showed a greater reduction in inhibition accuracy compared to the binge group in reward relative to neutral conditions. Both groups demonstrated significantly enhanced control when forewarned there was an increased chance response inhibition would be required. Controls secured the delayed reward more often than binge participants.

Conclusions: Despite the variability in the literature, this study demonstrated generalised impulse control deficits among binge drinking individuals that were unrelated to reward manipulations. These findings point to mechanisms that may confer vulnerability for transition from binge drinking to AUD.

Cognitive control is a vital executive process that is involved in selecting, generating, and maintaining appropriate, goal-directed behaviour. One such operation that draws heavily on this resource is the mapping of sensory information to appropriate motor responses (i.e., response selection/decision making). Recently, a transcranial direct current stimulation (tDCS) study demonstrated that the left posterior lateral prefrontal cortex (pLPFC) is casually involved in response selection and response selection training. Correlational brain imaging evidence has also implicated the pre-supplementary motor area (pre-SMA) in response selection (pre-SMA), and there is causal evidence that this brain region is involved in the proactive modulation of response tendencies when occasional stopping of a response is required. However, to date there is no causal evidence implicating the pre-SMA in response selection. Here, we investigated the causal role of pre-SMA in response selection, response selection training (Experiment 1) and inhibitory response selection control (Experiments 2 and 3), by employing anodal (excitatory), cathodal (inhibitory), and sham tDCS. Cathodal stimulation of the pre-SMA modulated response selection by increasing reaction times, but this effect was only observed in the context of proactive response inhibition, where participants had an executive set to stop. Our results suggest a context dependent role of the pre-SMA in response selection and, more broadly, show how task set can influence the interaction between the brain and behaviour.

An innate non-symbolic magnitude (NSM) system is claimed to scaffold children’s emerging symbolic magnitude (SM) system and ipso facto their numerical cognition. Two hypotheses have been proposed for differences in NSM/SM relationships: (1) a deficit NSM system; and (2) an intact NSM system but deficient SM system. Even though similar error, RT and IPS brain activation patterns in NSM/SM judgment abilities are cited in support of both hypotheses, variability in the data provide only limited support for either one; and the nature of NSM/SM associations, and their relationship with emerging numerical cognition, is yet to be clarified. We suggest (1) given the variability in NSM/SM response data, it is likely that different NSM/SM response patterns are embedded within the overall distribution, and (2) different cognitive factors may mediate different NSM/SM relationships. To test these claims, we used latent class analysis to classify the RT/error NSM/SM response patterns of 160, 5- to 8-year-olds. We assessed children’s visuospatial working memory (VSWM), Arabic digit (symbolic access), non-verbal IQ and basic RT abilities since these factors have been linked with children’s math abilities. We also assessed reading numbers (transcoding) and single-digit addition (SDA) math abilities. Four NSM/SM profiles were identified, three of which provide support for the two hypotheses. Only number symbol access and VSWM abilities were associated with profile membership. And profile membership predicted transcoding and SDA abilities. The findings clarify reasons for differences in NSM/SM relationships observed previously, and suggest a framework for studying neurophysiological correlates of NSM and SM representations.

Background: Cognitive reserve (CR) attempts to explain why some individuals can maintain cognitive functioning despite a burden of age- and dementia-related brain changes. Current theory states that cognitive domains are equally affected by cognitive reserve, however this is unlikely the case given that the pattern of decline is not uniform across cognitive domains in late adulthood.
Methods: 551 participants aged 60 years and older (M = 68.2, SD = 6.79, 300 female) were taken from the Brain Resource International Database (BRAINnet; www.BRAINnet.net). Participants underwent cognitive testing as part of the IntegNeuro battery of computerised and standardised tests. Cognitive reserve was measured by years of education.
Results: In a series of regression analyses controlling for age and sex, it was found that cognitive reserve significantly predicted performance on tasks of verbal memory, word generation, time estimation, span of visual memory, digit span, verbal interference, switching of attention (complex), and executive maze. However, cognitive reserve was not significantly associated with choice reaction time, motor tapping, switching of attention (simple), sustained attention, Go-NoGo performance, and emotion recognition.
Conclusions: This research suggests that complex tasks of memory, attention, and executive function more affected by cognitive reserve; conversely, the advantage of high cognitive reserve does not extend to simpler reaction time, motor response, and emotion recognition tasks. These results inform cognitive reserve theory, and provide avenues for the development of cognitive screening measures less biased by cognitive reserve (as current diagnostic tools focus on domains affected by cognitive reserve).

Classical decision theory suggests that choices result from cognitive representations of stable preferences. This study examined whether choices not only reveal preferences, but also shape them, a phenomenon referred to as ‘choice-induced preference change’. Previous studies have failed to provide reliable evidence for such an effect. We argue that in these studies, potential effects were masked by noisy measurements of preferences. In this study, we eliminated these measurement problems by using incentive-compatible elicitation of preferences, employing a Becker-DeGroot-Marschak auction.

Two groups completed a valuation-choice-valuation paradigm, in which participants valued foods before making binary choices between equally valued items, and subsequently valuing the foods again. Additionally, participants completed a control experimental sequence (valuation-valuation-choice), which assessed unsystematic changes in their valuations. One participant group completed a version of the task in which choices and valuations were incentive-compatible. The other group completed a version that was not incentive-compatible, choosing and evaluating snacks based on their liking rating, similar to previous studies. The effect of choices on preferences was assessed by comparing post-choice value change (in the valuation-choice-valuation sequence) to unsystematic value change (in the control valuation-valuation-choice sequence).

We found that choice history systematically affected preferences: the value of chosen items increased and decreased for rejected items. These results were only observed, however, in the incentive-compatible condition. Our findings imply that, contrary to dominant accounts of decision-making, preferences evolve as a function of past decisions. Additionally, our study suggests that studies aiming to investigate choice-induced preference change effects should adopt incentive-compatible preference-elicitation mechanisms.

In most risk-taking paradigms, instances of loss and reward are fully determined by a single choice. However some negative consequences result from many repetitions of a given behaviour (e.g. cancer from smoking). In these situations, rewards (e.g. pleasure from smoking) are typically highly probable, immediate and fully determined by a single choice, whereas losses (e.g. cancer) are uncertain (temporally and probabilistically) and only partially (and minimally) determined by a single choice. Conceptualising risk-taking in this way may provide important insights into the processes underlying people’s willingness to engage in repetitive health-damaging behaviours. We present a novel task designed to capture the essential features of this repetitive risk-taking environment, with preliminary data on a between-subjects manipulation designed to decrease risk-taking by increasing the salience of loss between choices.

Even glimpses of biological motion facilitate an array of social judgments. Remarkably, this ability may arise from processing occurring prior to the mediating effects of attention: Evidence to that effect has arisen in tasks involving to-be-ignored information in the form of flanker walkers. This Flanker-Interference Paradigm requires observers to extract information about a central Point-Light Walker (PLW) target accompanied by adjacent PLW flankers that observers are instructed to ignore. Using this paradigm in conjunction with the task of discriminating walking direction, Thornton and Vuong (2004) found that flanker-presence selectively impeded performance reaction time. More specifically, performance was most affected when the flankers were incongruent to the target. Here, the paradigm was adapted to a different yet equally socially-relevant task; Biological sex discrimination. Across a suite of experiments in which the qualities of the flanker walkers were manipulated, flanker presence was shown to elicit a different response in comparison to flanker-absent conditions in all experiments. Interestingly though, the impact was sex-specific to the target walker rather than dependent on flanker-type as per Thornton and Vuong. That is, observer performance was impaired when the flanked target was female, but facilitated when the flanked target was male. Drawing from the boarder literature, these effects are inline with the theory of a male bias; as the task difficulty increases with flanker-presence, so does the likelihood of a ‘male’ response.

Thornton, I. M., & Vuong, Q. C. (2004). Incidental processing of biological motion. Current Biology, 14(12), 1084-1089.

Background: Social decision-making in the Ultimatum Game (UG) is underlain by affective and cognitive monitoring mechanisms held in Insula and Medial Frontal regions respectively. We aimed to test this notion via confirmatory factor analyses (CFA) and structural equation modelling (SEM) of functional Magnetic Resonance Imaging (fMRI) data of the UG.

Methods: 216 healthy participants underwent fMRI during the UG, involving one proposer and one responder (i.e. the participant). In each trial, participants were prompted with the picture of a proposer, which made an offer to share monetary stakes. Offers varied in fairness: Fair offers (~50% of the money is shared); and Unfair offers (less than 25% of the money is shared). If participants accepted the offer, both players were paid; if they rejected, none of them was paid. We analysed BOLD-indexed activation during Unfair versus Fair offers in 5 regions of interest: Insula, Precentral Cortex, Anterior Cingulate Cortex, Supplementary Motor Area and Cerebellum. We applied CFA to test the two-factor structure: Affect and Cognitive Monitoring, and SEM to test the association between the resulting model and behavioural choices (i.e., percentage of Unfair offers accepted).

Results: CFA endorsed the two-factor model: Factor 1 – Affect: Insula, Precentral Cortex and Cerebellum; Factor 2 – Cognitive Monitoring: Anterior Cingulate Cortex and Supplementary Motor Area. The reliability of the factors was 0.758 and 0.673 respectively. SEM showed significant associations between the model and UG choices.

Conclusion: CFA and SEM are useful tools to test cognitive models of decision-making based on fMRI data.

In a dynamic world, an accurate model of the environment is vital for survival. To facilitate adaptive decision-making under uncertainty, agents ought constantly to seek out new information with which to update their world models. This aspect of behaviour is not well-captured by rational theories of decision-making, and the neural mechanisms of information seeking are poorly understood. However, one recent theory posits that canonical neural reward-processing circuits assign an intrinsic value to information, even when that information lacks any practical use. We investigated this question by recording EEG from twenty-two healthy participants performing a novel information-seeking task. In this task, participants could choose to pay a cost to receive information in advance about the likelihood of receiving monetary reward in a lottery following each trial. Importantly, acquiring information did not alter reward probabilities, which were known to participants. Behavioural results showed that participants were willing to incur considerable monetary costs to acquire early but payoff-irrelevant information, which indicates a preference for early resolution of uncertainty. Analysis of the event-related potential revealed that early information about the lottery outcome was encoded in the feedback-related negativity (FRN), a component associated with reward processing and specifically a reward prediction error, but not in the visual N1, which indexes selective attention. Crucially, information encoding in the FRN was orthogonal to reward encoding, suggesting that information may represent a distinct dimension of valuation in decision-making under uncertainty. These results suggest information seeking might result from an intrinsic valuation of information within canonical neural reward circuits.

Mindfulness is being used to both treat and prevent a wide variety of mental illnesses. However, the exact neural mechanisms of action by which mindfulness improves mental health are uncertain. Previous research has hypothesized that improvements may be related to enhanced attentional network activity, and improved direction of attention. Neural inhibition of non-relevant information is implicit in attention direction (such as neural activity related to a pre-potent response in a Go/Nogo task). In order to examine whether mindfulness enhances neural inhibition of non-relevant information, our study examined whether neural activity in a Go/Nogo task differed in individuals who practice mindfulness compared to healthy controls. We recorded neural activity with EEG while we presented a Go/Nogo task to 18 participants with significant experience in mindfulness, and 16 participants with no experience with mindfulness. Topographical distributions of neural activity and overall neural response strength comparisons were made across stimulus locked epochs between the two groups. The results indicated that the group that practiced mindfulness showed increased neural response strength during response inhibition compared to healthy controls who had no mindfulness experience. This finding provides support for the idea that the mechanism of action by which mindfulness treats and prevents mental illness is strengthened attentional network function, resulting in an improved ability to direct attention. The finding also suggests that if you wish to improve your ability to focus on grant application writing, ten minutes of mindfulness a day may be a worthwhile investment.

The vagus nerve is crucial for balancing energy expenditure in response to metabolic signals. As well as integrating peripheral homeostatic signals and influencing cardiovascular function, it also affects behaviour via the prefrontal cortex (e.g. cephalic phase responses). As a result, cardiac signals can reflect prefrontal neural activity, as well as behaviours that are often associated with prefrontal areas (e.g. emotion regulation, decision making, and time perception). Cardiac measurements have thus been used as a non-invasively measured index of behaviourally-related basal metabolic activity. As basal metabolic activity determines the temporal organisation of energy intake, it is possible that it is related to delay discounting (the devaluation of rewards as a function of delay). We hypothesised that cardiac signals – as a proxy for basal metabolic activity – would be associated with both time perception and delay discounting, within individuals. We recruited 145 participants and assessed delay discounting (choices between monetary rewards delivered at different delays), temporal reproduction, and heart rate. We found a significant negative correlation between discount rates and both mean heart rate and heart rate variability. We also found a positive correlation between the slope of the psychophysical function for time reproduction (i.e. the rate of time perception) and sympathetic tone as indexed by time-frequency cardiac measurements. Our results indicate that individuals with higher heart rates, and higher heart rate variability, have a relatively faster sense of time and lower discount rates. This suggests that basal metabolic activity affects both time perception, and intertemporal decision making.

A major goal of health research is to develop interventions that can prevent or delay age-related declines in cognitive abilities. Recently, exercise and cognitive training have been extensively investigated for their possible protective effects on cognition, with mixed results. This study aimed to determine whether combining exercise with cognitive challenge in an immersive virtual reality exer-game on cognitive performance in sedentary young adults. This population has been identified as being at high risk for developing cognitive decline in later adult life, and some studies have suggested that the neural changes associated with the development of age-related dementias may begin in young adulthood. Participants were randomly allocated to one of two acute exercise conditions, (1) an exercycle-only condition or (2) an immersive exer-gaming condition, which involved wearing an Oculus Rift ™ so participants were immersed in a 3D virtual reality gaming environment. The exercise protocols in each condition were matched for time, duration and intensity. A battery of cognitive tasks and self-efficacy questionnaires were administered before and after participation in exercise interventions. While both exercise conditions were expected to result in acute improvements in cognitive performance, we predicted that performance gains would be greater in the exer-game condition than in the exercycle-only condition. In accordance with these hypotheses, post-test data show the exer-game condition significantly improved self-efficacy and cognitive function compared to the exercycle-only condition. This study suggests that combining the effects of exercise and cognitive challenge in immersive virtual environments is a promising intervention for improving or maintaining cognitive performance.

Cognitive control refers to a group of processes associated with performance of specific tasks through appropriate adjustments in selective attention and response selection, whilst minimising interference from irrelevant information. Recent event-related potential (ERP) research has shown that task switching (a process that requires high levels of cognitive control) is associated with a sustained positive component, occurring approximately 500-800 ms post stimulus onset, that has been shown to have a different pattern of neural generators to the domain general P3 ERP component. The current study tested 22 young adults on a switch task whilst EEG data were continuously recorded and calculated difference waveforms by subtracting the grand average ERP of the third trial after a cue, where reactive control of rule execution is required, from trials immediately following the cues, where reactive control of rule execution and task switching processes are required. We investigated single-trial alignment of the sustained positive ERP component to response time, and found that the elicited positivity was perfectly response aligned, but potentially as early as 300 ms post stimulus onset. This is, to our knowledge, the first single-trial analysis of task switching data. The results provide further support for the sustained positivity being related to task switching, but suggest this cognitive process begins significantly earlier than previously thought.

Motivation is underpinned by cost/benefit-based decisions. Previous studies have shown that rewards (benefits) are devalued by the physical effort (costs) of actions. Yet we also commonly make choices between options that differ in their cognitive demands, but have equitable physical costs. Here, using computational modelling techniques in conjunction with event-related functional magnetic resonance imaging (fMRI), we examined how healthy human adults subjectively value rewards when faced with varying cognitive and physical demands.

Participants trained on two tasks that varied parametrically in either cognitive or physical effort. After training, participants made choices between a fixed low effort option for low reward, and a variable higher effort offer for higher reward. Choices were made on separate trials for the cognitive and physical tasks while they were scanned with fMRI.

Our results showed that both physical and cognitive effort devalued rewards, with participants choosing the fixed low-effort/low-reward option more frequently as the effort associated with variable high-effort/high-reward offer increased. Computational modelling of participants’ choices revealed the distinct discounting functions that described the devaluation of rewards by cognitive and physical effort, and fMRI analyses demonstrated the shared and separate neural circuits that underpinned effort discounting across the two domains.

These data characterise the mechanisms by which cognitive and physical effort discount rewards and highlight the influence of each domain on motivation and incentivised decision-making. These findings highlight the neural circuits underlying cognitive and physical effort discounting, and pave the way for understanding disorders of motivation such as apathy.

Cued task-switching paradigms have been used to investigate proactive and reactive control processes, by examining rapid alternation between two or more task sets using cues that provide information about the upcoming task. Proactive control processes refer to preparatory adjustments that optimise performance in anticipation of repeat or change in context. Reactive control processes refer to target-driven interference control mechanisms. In evidence accumulation models, proactive control processes can be modelled by response caution – an adjustment of response threshold based on cue information (e.g., Karayanidis et al. CABN 2009:9:202-215). In this study, we examine how individual variability in response caution affects behavioural and ERP measures of task-switching performance. Participants (N=219, 21±5y, 97 male) completed a cued task-switching paradigm that included fully and partially informative cues (Karayanidis et al., 2009). Response caution and other decision parameters were derived using a drift diffusion model. Response caution adjustment across different cue types revealed two groups of participants: ‘adjusters’ showed the predicted pattern of response caution adjustment as a function of cue information (N=135), whereas ‘non-adjusters’ showed a fixed level of response caution across all cue types (N=84). Groups did not differ on key demographic variables, such as age, gender and IQ. However, non-adjusters were faster and just as accurate as adjusters. Preliminary examination of ERP data show differences in proactive control processes associated with anticipatory attention between adjusters and non-adjusters. The implications for models of task-switching will be discussed.

Background
Evidence from tobacco cessation studies suggests an association between sleep disturbance during withdrawal and an increased risk of relapse. A potential mechanism for this relationship is the dysregulation of inhibitory control (IC) function as a result of an interactive effect of sleep loss and tobacco withdrawal.
Method
Cigarette smokers (n=19, Mage±SD = 23.2±4.3 years) and non-smokers (n=10, 21.5±1.9 years) were recruited. Smokers were randomly assigned to an active smoking or acute (2-14 hour) abstinence group. Participants spent one night sleeping in a polysomnography lab and a second night deprived of sleep separated by 1-week (order counterbalanced). Participants completed a Stop Signal Task and Brief Questionnaire of Smoking Urges both pre- and post-sleep, and every 3 hours across a 12 hour sleep deprivation period. Stop Signal Reaction Time (SSRT) was calculated as a measure of IC.
Results
Following sleep deprivation, active smokers demonstrated slower SSRT (325.3±119.9 ms) compared to both abstaining smokers (225.0±37.7ms) and non-smokers (227.9±51.4 ms). While a significant main effect of sleep deprivation on SSRT was found (p=.02), no effect of normal sleep, group, or group x condition interaction was found. Higher craving was reported by abstaining (M±SD; 44.8±19.1) than active smokers (22.8±8.0) following sleep deprivation (p=.04).
Discussion
Extended wakefulness involving sleep restriction appears to be associated with impaired IC, however there appears to be no interactive effect of sleep deprivation and acute withdrawal on IC. Higher variability for active smokers SSRT may be suggestive of increased impulsive responding arising due to hyper-stimulation.

Background Products containing 1-benzylpiperazine (BZP) and 1-(3-trifluoromethylphenyl)piperazine (TFMPP) were widely available for recreational use around the world. They were sold as the active ingredient of what were collectively known as ‘Party Pills’ or ‘Herbal/Natural Highs’ and marketed as safe alternatives to 3,4-methylenedioxymethamphetamine (MDMA) and the other amphetamines. Evidence suggests that BZP selectively elevates dialysate dopamine (DA) levels as well as modulating serotonergic (5-HTergic) and noradrenergic (NAergic) systems in vivo, while TFMPP mainly affects the serotonergic pathways and exhibits varying affinity for most 5-HT receptor subtypes. Despite reported similarities between BZP, TFMPP and psychostimulants such as amphetamine, the acute effects of these drugs on human information processing are unclear.
Objective This study aimed to compare the acute effects of BZP and TFMPP on the auditory P300 in comparison to dexamphetamine in healthy young male adults.
Methods A double-blind, randomised, placebo-controlled study was used. Eighty healthy, right-handed males were given a single oral dose of either BZP (200mg), TFMPP (60 mg), dexamphetamine (20 mg) or placebo (methylcellulose) and tested both before and 120 minutes after drug administration while undertaking the auditory oddball task.
Results Contrast analysis revealed that BZP (t(12)=4.48, p=0.01), TFMPP (t(12)=3.03, p=0.01) and dexamphetamine(t(15)=2.53, p=0.02) significantly reduced the P300 amplitude. Neither P300 latency (F(4,82)=0.339, p=0.093) nor the mean reaction time (F(4,82)=1.274, p=0.960) was affected by any of the drug treatments.
Conclusion A single oral dose of BZP or TFMPP affected the auditory evoked P300 waveform in a manner similar to dexamphetamine.

Background Alcohol relapse after detoxification is a principal challenge for alcohol-dependent individuals (ADI). Identifying predictors of relapse after alcohol detoxification is critical to improve clinical outcomes. Traditional measures including baseline alcohol use do not consistently predict outcomes; therefore cognitive functioning was added to a prediction model. We investigated whether decision-making skills predicted early alcohol relapse and craving and whether cognitive functioning compared to traditional measures, better predicted relapse and craving.
Methods We applied a prospective cohort naturalistic design with follow-up assessments two weeks after detoxification in a public inpatient facility in Melbourne. Forty-two ADI were assessed at baseline, and 36 followed-up at two weeks. During this time, 52.8% relapsed. Baseline measures included impulsive decision-making (Iowa Gambling Task) and general cognitive functioning (Montreal Cognitive Assessment) as cognitive measures, and an alcohol use self-report (Severity of Alcohol Dependence Questionnaire) as a traditional measure. Outcome measures were relapse status (Timeline Followback Interview) and alcohol craving (Alcohol Craving-Short-Form-Revised).
Results Our primary analyses did not support hypotheses. However, we found a significant correlation between general cognitive functioning and craving at follow-up. Post-hoc analyses determined that IGT predicted days to relapse; ADI who had increased impulsive decision-making were likely to relapse sooner.
Conclusion Poorer general cognitive functioning is associated with craving in ADI early after detoxification and impulsive decision-making predicts immediate relapse after detoxification, suggesting that cognitive profiles contribute to alcohol treatment outcomes. Implementing cognitive enhancement may reduce craving levels early after treatment and testing decision-making during detoxification can identify ADI vulnerable to relapse quicker.

Deficits in emotion regulation have been implicated in mood and anxiety disorders which are more prevalent in women than men. Few electrophysiological studies have investigated sex differences in early emotional reactivity and later emotional regulation controlling for the influence of menstrual phase. Event-related potentials (ERPs) were recorded from 28 early follicular women, 29 midluteal women, and 27 men who were instructed to ‘reappraise’/‘maintain’ or ‘suppress’/‘maintain’ their emotional responses while viewing unpleasant images. Early ERP components (P1 and N1) were increased in midluteal women compared to men regardless of instructional set, reflecting greater early emotional processing. A novel finding of increased N2 amplitude during suppression was found for midluteal women compared to men, suggesting that midluteal women are less able to suppress emotional responses to negative stimuli compared to men. No sex or menstrual phase differences were apparent in P3 or LPP although P3 amplitude was lower during suppression of emotional response. This study underscores the importance of considering menstrual phase when examining sex differences in the cortical processing of emotion regulation and demonstrates that midluteal women have greater deficits in down-regulating their emotional response compared with men and early follicular women.

Background: Recent neuroimaging studies in body dysmorphic disorder (BDD) have identified abnormalities in the structure, function and network connectivity of brains in BDD patients. This data indicates structural and functional abnormalities of the visual system and the fronto-limbic circuit. This study aimed to investigate structural brain abnormalities further by examining cortical thickness in BDD.

Methods: Twenty participants with BDD and 20 healthy controls matched on age, gender and handedness underwent T1-weighted magnetic resonance imaging. Gray matter cortical thickness was compared across groups. Relationships between cortical thickness changes and symptom severity, insight, social anxiety and depression were explored in the BDD group.

Results: BDD patients showed significantly reduced cortical thickness in the middle temporal gyrus, which was evident after controlling for total brain volume. Further, within the BDD group, lack of insight was associated with thinner cortical thickness in the left fusiform gyrus.

Conclusions: This study demonstrated that cortical thinning in middle temporal and fusiform regions is associated with common symptoms seen in BDD.

Background: A hyper-glutamatergic commonality has been as proposed as the basis for both schizophrenia and autism, disorders that exhibit similar social and interpersonal dysfunction that exists between them. This study investigates the neurotransmitter status in those with high levels of this shared phenotype - Social Disorganisation, compared with a low scoring group. It was predicted that the level of 1H-MRS quantified glutamate would be increased in the auditory cortex of the High group, compared to the Low group. Glutamine levels were also investigated. Method: An 80ms TE PRESS 1H-MRS sequence in a 3T MR scanner was delivered to the left and right auditory cortices of 30 adults (Low: 9 female, 4 male, High: 10 female, 7 male). Results: The data revealed a significantly higher glutamate level in the right auditory voxel for the High group as predicted (t(18.29)=2.11, p=0.048) but not in the left voxel (t(26.88)=1.77, p=0.088). Glutamine level did not differ between the groups (left: t(27.62)= 1.50, p= 0.145, right: t(25.07)= 1.67, p= 0.1076). Conclusions: These data support the hyper-glutamatergic hypothesis of autism and schizophrenia, and suggest that excess of glutamate may link the two spectra. The data also highlights the spectrum nature of autism and schizophrenia with the trait phenotype explained in part by metabolite level differences. Finally, the data provides a strong argument for a focus on the identification of phenotype specific neural markers for more effective diagnosis and targeting of symptom treatment, rather than whole disorders.

Theory of Mind (ToM), our ability to attribute mental states to others, is vital for everyday social interactions. However, the prevalence and severity of ToM deficits following stroke remain unknown. Here we investigated whether impairments on a measure of ToM, the Reading the Mind in the Eyes Test (RMET), were associated with specific lesion sites. The task consisted of 36 trials, each displaying a black and white photograph of an eye region. On each trial, the patient was asked to indicate which word he or she felt best described what the person in the photograph was thinking or feeling. The study included 75 unifocal lesion stroke patients (65% male), aged 22–88 years (M=62.12). Patients with severe neglect and dysphasia were excluded. Lesion sites were identified from clinical diagnostic scans and mapped onto a standard brain template using MRIcron. A voxel-based lesion symptom mapping (VLSM) analysis was then conducted to determine which lesion sites were associated with poorer performance. RMET scores ranged from 10-33 (M=19.95) out of a possible 36, with 44 patients (58.67%) scoring within the abnormal range (≤21). Lesions in the right hemisphere were reliably associated with lower scores on the RMET. Key sites of damage included the precentral gyrus, superior temporal gyrus, temporal parietal junction, inferior frontal gyrus, insula, temporal pole, caudate nucleus, and external capsule. These VLSM findings suggest that a variety of right hemisphere lesions can be associated with impaired performance on the RMET, and highlight the relatively high incidence of ToM impairments following stroke.

Spoken language production theories and lesion studies highlight several important prelinguistic conceptual preparation processes involved in the production of cohesive and coherent connected speech. Broader cognitive functions such as attention may mediate these processes. Sustained and selective attention are two attentional processes recently highlighted as potential candidates integral for connected speech. We aimed to investigate 1) whether stroke patients without significant aphasia exhibit impairments in cohesion and coherence in connected speech, and 2) the role of sustained and selective attention in the production of connected speech. Stroke patients (N=18) and matched healthy controls (N=21) completed two narrative tasks designed to elicit connected speech. The tasks comprised the retelling of a fairy tale and discussion of a topic chosen by participants. A multi-level analysis of within and between-sentence processing ability was conducted on speech samples. The results revealed impairments in cohesion and coherence of connected speech for the stroke group relative to controls. For stroke patients, coherence at the global level (i.e. fewer propositional repetitions) was related to better performance on selective and sustained attention tasks. Selective attention performance also negatively correlated with global coherence errors. Sub-group analyses showed that for right hemisphere stroke patients, cohesive speech was associated with increased selective attention while coherent speech was related to increased sustained attention. Thus, connected speech deficits were revealed in a heterogeneous stroke group without prominent aphasia. Furthermore, broader cognitive processes, such as sustained and selective attention, appear to play a role in producing connected speech at the early conceptual preparation stage.

In the literature, the representation of the meaning of a word (semantics) has been modeled either as a network of holistic (nondecomposed) nodes or as a decomposed set of features. Despite experimental studies showing that semantic relatedness and association have different effects on word retrieval, these models primarily focus on semantic coordination (within category relationships), and neither model clearly implements a wider range of semantic or associative relationships.
The current study aimed to contribute to this discussion by examining the role of part-whole relations (e.g., ‘tyre’ is a part of ‘car’) and semantic coordination (e.g., ‘bus’ and ‘car’) in lexical access in spoken word retrieval while controlling for association and its direction (e.g., ‘car’ is produced as associate given ‘seatbelt’ but ‘seatbelt’ is not produced as an associate given ‘car’).
Two picture naming experiments were conducted to compare the effects of prior naming of parts and semantic coordinates on spoken word retrieval. Results of the first experiment demonstrated an inhibitory effect of semantic coordinates on naming latency but only when they were also associated. The second experiment replicated the inhibitory effect produced by associate-coordinates in experiment 1, but failed to detect any significant effect of parts on word retrieval.
These results demonstrate that both the overlap of semantic features (coordination) and association between prime and target are needed to produce an inhibitory effect on word retrieval, with implications for theories of word retrieval.

Although left-hemispheric damage can impair both the production and comprehension of language, it has been claimed that comprehension is more bilaterally represented than is production. A variant of this theme is based on the theory that different aspects of language are processed by a dorsal stream, responsible for mapping words to articulation, and a ventral stream for processing input for meaning. Some have claimed that the dorsal stream is left-hemispheric, while the ventral stream is bilaterally organized. We used fMRI to record activation while left- and right-handed participants performed covert word-generation task and judged whether word pairs were synonyms. Regions of interest were Broca’s area as part of the dorsal stream and the superior and middle temporal gyrus as part of the ventral stream. Laterality indices showed equal left-hemispheric lateralization in Broca’s area for word generation and both Broca’s area and temporal lobe for the synonym judgments. Handedness influenced laterality equally in each area and task, with right-handers showing stronger left-hemispheric dominance than left-handers. Although our findings provide no evidence that asymmetry is more pronounced for production than for comprehension, correlations between the tasks and regions of interest support the view that lateralization in the temporal lobe depends on feedback influences from frontal regions.

Background: To determine whether individual variability in conscious error awareness might influence the discrepancy between subjective cognitive concerns and objective cognitive function that commonly manifests in cognitively-normal older adults.
Methods: Thirty cognitively-normal older adults (Mage=71.1yrs,SD=5.3) were administered a subjective cognitive concern questionnaire, neuropsychological memory measures, and the Error Awareness Task. This task was designed to elicit errors, which were made with or without awareness.
Results: Poorer error awareness was related to greater concerns of executive dysfunction, and was also related to a tendency of older adults to underestimate their performance (i.e. greater subjective concerns compared to objective cognitive function), after accounting for variance in the model explained by age, education level and depressive symptoms.
Conclusions: These findings contextualise the subjective concern-objective cognitive performance disparity within the context of error awareness dysfunction; that is, poorer awareness of errors relates to greater concerns of organisational dysfunction, which may well be driven by an inability to keep track of the errors being produced. An individual's level of awareness of everyday errors should be considered when assessing the clinical implications of SCD in cognitively-normal older adults.

Background : Neural activity in sleep is important for daytime cognitive function but the exact process and function of different sleep stages remain unclear. Sleep fragmentation quantified as brief periods of awake like EEG called arousals is associated with neurocognitive deficits. Children with sleep-disordered breathing (SDB) have fragmented sleep as well as an altered arousal response that serves to protect their sleep integrity but may come at the expense of some of the normal sleeping neural processes. This change in arousal response may be reflected in EEG spectral power associated with arousal, potentially providing a neural correlate of the deficits reported.
Hypothesis: Children with SDB will have an altered EEG arousal spectral profile compared to controls which is correlated with cognitive performance.
Methods: 42 children with SDB and 48 healthy controls underwent overnight continuous EEG and physiological monitoring and daytime cognitive assessment. Subjects were divided into two groups based on results from a sleep test: Mild sleep disturbance (n = 24) and severe sleep disturbance (n = 18). Mean relative power was calculated for various frequency band over the 10 seconds prior to and following arousals in sleep.
Results: Children with SDB have an altered EEG arousal spectral profile compared to controls with reduced high frequency power. Many cognitive measures significantly correlated with the various power frequency bands for certain sleep stages.
Conclusion: Results indicate the types of cognitive processes disrupted by sleep disturbance and point to potential neurocognitive functions of the different sleep stages.

The human parahippocampal cortex has been ascribed central roles in both visuospatial and mnemonic processes. More specifically, evidence suggests that the parahippocampal cortex subserves both the perceptual analysis of scene layouts as well the retrieval of associative contextual memories. It remains unclear, however, whether these two functional roles can be dissociated within the parahippocampal cortex anatomically. Here we examined whether neural activation patterns associated with visuospatial analysis of scenes and contextual mnemonic processing can be dissociated along the parahippocampal longitudinal axis. We used functional magnetic resonance imaging (fMRI) to measure parahippocampal responses while participants engaged in a task that required them to judge the contextual relatedness of scene and object pairs, which were presented either as words or pictures. Results from a combined factorial and conjunction analyses indicated that the posterior section of parahippocampal cortex is driven predominantly by judgments associated with pictorial-scene analysis, whereas its anterior section is more active during contextual judgments, irrespective of stimulus category (scenes versus objects) or modality (word versus picture). Activation maxima associated with visuospatial and mnemonic processes were spatially segregated, providing support for the existence of functionally distinct subregions along the parahippocampal longitudinal axis, and suggesting that in humans the parahippocampal cortex serves as a functional interface between perception and memory systems.

Evidence from recent neuroimaging research examining brain changes associated with Post Traumatic Stress Disorder (PTSD) suggests the possibility of a ‘neural signature,’ predominantly characterised by abnormalities within the temporal and parietal lobes of the right hemisphere (e.g., Engdahl et al., 2010). This possibility fits with existing data demonstrating the role of the right hemisphere in mediating visuospatial and emotional processing, particularly of negative emotions. The current study begins to test whether these neurological associations elicit behavioural changes in a non-clinical population exposed to analogue trauma in the laboratory. Specifically, we wondered: is there a right hemisphere memory bias for remembering traumatic, compared to neutral, images? We used a strongly right-handed sample (N = 72) and manipulated image valence (negative, neutral) and visual field (left, right) within-subjects. In a combined encoding-recognition phase, participants viewed matched-valence image pairs, in the left and right visual fields, for 400ms. We tested participants’ memory for target images (one from each pair) using a forced-choice testing paradigm. We also recorded confidence judgments and reaction times. Participants had significantly better recognition memory for images presented in the left versus the right visual field, a pattern consistent with pseudoneglect. This right-hemisphere benefit was primarily driven by a difference in recognition memory for negative images. This study provides preliminary evidence that a right hemisphere memory bias for processing traumatic stimuli may exist. The findings are consistent with neurological data suggesting that impairments in right hemisphere processing may be contributing to the development of PTSD.

Counter-intuitively, we are far more likely to remember a mistake if we were confident that our initial response was correct relative to those incorrect responses in which we have little confidence. This is the hypercorrection effect. Most explanations of hypercorrection hinge on the metacognitive mismatch between the expected outcome (our response is correct) and the actual outcome (we made an error). That is, we remember these errors because they are surprising. However, these errors have one other defining feature that might make them more memorable; they are rare. Hence hypercorrection might just be a special case of the Von Restorff effect. To distinguish between these possibilities we questioned participants on 190 common misconceptions to ensure a large number of high confidence errors but varied the feedback to manipulate the believed rate of high confidence errors. Behavioural data indicated a strong hypercorrection effect regardless of the frequency of high confidence errors. A significantly higher amplitude P3 ERP waveform was observed over central sites for high confidence errors regardless of frequency. Overall levels of autonomic arousal did not differ between conditions. We interpret these results as consistent with the hypercorrection effect resulting from the metacognitive mismatch between expectations and actual outcomes.

Past research has shown that stimuli presented in more than one modality (multi-modal) can facilitate recognition memory compared to stimuli presented in a single modality. Furthermore, the semantic congruency of the multi-modal stimuli plays a crucial role in facilitating recognition memory. However, it is unclear whether this effect is still observed after multiple repetitions. Thus, the aim of this study was to determine the effects of repetition in four different modality conditions, i.e. congruent (when auditory and visual pairs were semantically congruent), incongruent (when auditory and visual pairs were not semantically congruent), tones (visual stimuli presented with a meaningless tone) and visual (visual stimuli presented alone) in older and younger participants. Participants were required to make old/new discriminations in a continuous recognition test. Stimuli were presented once, or repeated after a short delay (after 6 intervening items) and then repeated again after a long delay (after 37 intervening items). Results show that both older and younger participants showed significantly fewer errors for the congruent condition compared to the other conditions in the 1st presentation and short delay trials, but not in the long delay trials. This suggests that semantic congruency only facilitates recognition memory for the first repetition, and has no effect in subsequent repetitions. Findings also suggest that problems in recognition lie in retrieval rather than in encoding.

It has recently become apparent that common genetic single nucleotide polymorphisms (SNPs) may account for some of the variation in cognitive performance seen in neurotypical populations. With regard to memory, two SNPs have emerged as potential mediators of this variation: the BDNF Val66Met polymorphism, and the COMT Val158Met polymorphism. Both SNPs influence the expression of known regulators of Long-Term Potentiation (LTP) -Brain-Derived Neurotrophic Factor (BDNF) and catecholamines respectively- however, until recently, the relationship between these SNPs and LTP has been difficult to assess in human subjects. Using a newly developed EEG paradigm for non-invasively measuring LTP-like changes in visually evoked potentials (VEPs), the aim of the current study was to assess the impact of these two SNPs on the neural basis of memory formation in human subjects in vivo. Following a ‘photic tetanus’, BDNF Val homozygotes demonstrated VEP enhancements indicative of LTP while Met carriers did not, suggesting an association between the reduced BDNF secretion of the Met allele and reduced LTP. Conversely, while COMT Met homozygotes demonstrated the expected LTP-like changes in VEPs, Val carriers did not, which may be due to the reduced availability of dopamine and noradrenaline associated with the Val allele. The current results provide a neural mechanism through which these two SNPs influence mnemonic performance, and support the role of genetics in understanding natural variation in human cognitive performance.

Low frequency theta band oscillations are thought to provide a timing mechanism for hippocampal place cell firing and to mediate the formation of spatial memory. In rodents, hippocampal theta has been shown to play an important role in encoding the configural layout of a new environment during spatial navigation, but it is unclear if this is also the case for humans. To address this question, we recorded healthy participants’ brain responses with a 160-channel whole-head MEG system as they performed two training sets of a virtual Morris water maze task. Environment layouts of the two sets were kept constant to measure theta activity (4-8 Hz) during spatial learning in novel and familiar environments. We found that right hippocampal theta was stronger during the first training set compared to the second one, whereas left hippocampal theta showed a comparable increase in both sets (relative to a sensorimotor control condition). At the individual level, right hippocampal theta in the first training set correlated with spatial navigation performance in both training sets, which strongly argues for the functional importance of right hippocampal theta in initial environment layout encoding during spatial navigation. Our findings provide the first evidence that right hippocampal theta activity is associated with encoding environment layout in the human brain.

Long-term potentiation (LTP) is considered to be the basic mechanism of synaptic plasticity underlying learning and memory. LTP is induced non-invasively in the human visual cortex through rapid, repetitive presentation of sine grating stimuli. With EEG recording, this is observed as post-stimulation enhancement of the N1b component of the visual evoked potential (VEP). The role of inhibitory processes in the phenomenon is currently unclear but is likely to involve GABA and GABAA receptors, the activity of which is enhanced by the benzodiazepine lorazepam. Twelve healthy adult males participated in a placebo controlled double blind crossover study. VEPs elicited by horizontal and vertical sine grating stimuli presented at 1Hz were assessed at baseline, and at 17 minutes following a two-minute photic tetanus (either the horizontal or vertical stimulus presented at 9Hz). LTP was measured as amplitude change of the VEP recorded from occipital-parietal electrode sites. Subjective and objective measures confirmed the sedative effect of lorazepam on participants. Lower overall amplitude of the VEP was observed with lorazepam compared to placebo, consistent with the increased cortical inhibition produced by benzodiazepines. While the evoked responses demonstrated the expected component morphology, the photic tetanus failed to produce significant potentiation of the N1b component in either condition. Minor modifications to the LTP protocol from previous studies, including compressed timing and performance of an unrelated visuomotor task during the wait period, may have been responsible. The findings do not fit with the current understanding of LTP, suggesting that fundamental elements of the LTP protocol require revisiting.

Experience with visual stimuli can improve their perceptual performance, a phenomenon referred to as visual perceptual learning (VPL). It remains unclear whether VPL is accompanied with conscious (explicit) or unconscious (implicit) knowledge. In line with the former, previous VPL studies have reported concomitant increases in subjective measures of conscious awareness. However, these studies have largely failed to control for task accuracy, which typically correlates with measures of consciousness. Here, using a staircase method to control objective task accuracy, we sought to investigate whether the VPL of high- and low-level face properties can improve participants’ metacognitive accuracy in their perceptual decisions. Across three consecutive days, two groups of participants were trained to discriminate faces based on either their identity (n = 10) or contrast (n = 8). Using the staircase, task accuracy remained constant while identity and contrast thresholds decreased with training. Importantly, metacognitive accuracy decreased in the identity condition, but remained constant in the contrast condition. This suggests that while metacognition was not improved with the VPL of face identity, it was improved with the VPL of face contrast. One explanation for this counter-intuitive finding may be the Reverse Hierarchy Theory proposed by Hochstein and Ahissar (2002, Neuron), according to which, the VPL of high-level object properties is accomplished implicitly, while the VPL of low-level features, requiring focused attention, is accompanied with explicit knowledge.

The Laboratory for Animation Technologies (LAT) is developing a computational framework for the integration of realistic interactive computer graphics with neural systems modelling. This system, known as Brain Language (BL), is a modular, “Lego-like” framework with a particular focus on real-time interactive animation and visualisation/modification of the underlying neural network models. It is designed to allow researchers to integrate a wide range of models from diverse fields such as psychology, physiology, neuroscience and even animation. Unlike many other neural simulation software packages, BL also incorporates high quality 3D computer graphics as integral parts of models.

We are currently designing user interface tools to facilitate real-time modifications of models in BL and to enhance users' conceptual understanding of model structure and behaviour. Real-time modification is particularly beneficial when dealing with larger and more complex systems as the effects of modifying certain parameters can be difficult to predict. Using BL’s visual interface tools, users can instantly see the effect of changes to a model both numerically and behaviourally, allowing “soft exploration” of parameter spaces in real time.

We will present a small selection of models implemented in BL, ranging from simple neural circuits to larger, more complex networks. These are intended to showcase a number of the features of BL’s user interface, including model visualisation, model-driven 3D animation and interactive parameter space exploration.

Relaxation is a core component of psychological therapies such as CBT and Exposure Therapy yet achieving an appropriate state of relaxation can be time-consuming and often requires pharmacological assistance. Recently, however, a closed loop EEG feedback (CLEF) system has been designed and trialled which appears to induce deep relaxation within just a few seconds. This system monitors EEG using a standard EPOC/Emotiv® EEG headset then feeds back selectively filtered EEG in real time via two flickering LED lights placed at the visual periphery. In response to this feedback, an individual’s underlying neural activity is automatically modulated according to the feedback bandwidth (e.g. theta, alpha). Pilot studies into the effects of CLEF clearly demonstrate that participants experience a sense of deep calm and relaxation on both subjective and objective measures. In addition, CLEF appears to improve information flow between brain regions as measured by transfer entropy, as well as increasing functional brain network activity as measured by connectivity density and clustering coefficients. Taken together, these results suggest that CLEF may provide a new, non-invasive method for effecting relaxation which could potentially be used as an adjunct to traditional therapies such as CBT. Furthermore, CLEF appears to enhance cognitive processing such as working memory through reduced error rates and improved reaction times as well as hinting that the human brain possesses opti-responsive network properties that can be externally manipulated.

In neurological disorders such as Parkinson’s disease, correct functioning of the basal ganglia circuitry is compromised, resulting in patients suffering from movement-related and cognitive impairments. Structural mapping could give valuable insight into the functional organisation of the basal ganglia and provide useful information to help understand the pathophysiological processes associated with diseases involving basal ganglia functional impairment.
Here, we use diffusion-weighted magnetic resonance imaging (dMRI) tractography in 20 healthy subjects to examine the connectivity of the main functional circuits of the basal ganglia. Connectivity between caudate, putamen, globus pallidus internal and external segment (GPi/e), substantia nigra and subthalamic nucleus, and, motor cortex, frontal cortex, parietal cortex, occipital cortex and temporal cortex was examined. Basal ganglia structures were parcellated based on their connectivity to cortical target regions in all subjects and the resulting connectivity maps were averaged in a study specific template space. Preliminary results suggest, that the functional organisation of the basal ganglia is consistent across individuals and follows a general rostro-caudal/ventro-lateral pattern. While some regions are predominantly connected to a single cortical target, other areas within the basal ganglia show overlapping connections with multiple cortical areas. Our preliminary connectivity based parcellation maps presented here are in line with previous dMRI studies using similar techniques and in accordance with anatomical tracer studies in animals. We believe that our method is useful for studying disorders affecting basal ganglia networks and could facilitate treatments that directly target basal ganglia structures, such as deep brain stimulation.

Background:
The basal ganglia are a group of subcortical nuclei involved in the control of motor movements, cognition, and emotion. These nuclei have been notoriously difficult to image using fMRI due to their small size and high iron content. Here we sought to circumvent these issues using ultra-high field MRI. We investigated the application of a novel multi-echo (ME) EPI sequence that should improve signal in the iron-rich nuclei of the basal ganglia, such as the globus pallidus, by allowing an optimized weighting of data from different echo times. We also examined the effects of increasing spatial resolution in a more conventional, single-echo EPI sequence.

Methods:
Data were acquired using a MAGNETOM 7T scanner (10 subjects). We acquired single-echo data at three resolutions (2.5mm³, 1.3mm³, 1.0mm³) and ME at a single resolution (2.5mm³) while subjects performed four-finger movements. Resting scans were acquired to measure the temporal signal-to-noise ratio (tSNR) associated with each sequence. Anatomical regions of interest (ROIs) were defined using a high-resolution atlas and guidelines for sub-dividing the striatum and included the caudate, putamen, globus pallidus, and ventral striatum.

Results:
When comparing the ME data to the single-echo data (2.5mm³) we find an increase in the tSNR throughout the basal ganglia. Critically, we also find that the functional contrast-to-noise ratio (CNR) increases in most ROIs, particularly in the globus pallidus with CNR increasing up to four-fold.

Conclusions:
Our results indicate that advanced fMRI techniques at ultra-high field (7T) such as ME-EPI can improve functional imaging of the basal ganglia.

When participants view a stimulus twice, the blood flow recorded by functional magnetic resonance imaging (fMRI) is reduced relative to the first presentation. This “adaptation” is thought to indicate that the same neuronal subpopulation is involved in processing the stimulus. Areas that are more selective for a given stimulus type tend to show greater adaptation. Adaptation is measured within regions of interest (ROIs) identified using a structural or functional localiser or based on prior research. As with all ROI analyses, the benefit of greater power and less comparisons carries a cost of potentially missing effects in other areas. We are developing a new method for measuring adaptation across the whole brain using a searchlight technique. We analysed the data using our Whole Brain Adaptation Searchlight (WBAS) analysis as well as conducting a standard Statistical Parametric Mapping (SPM) whole-brain analysis. Participants viewed faces and objects, some of which repeated sequentially. Applying the same cluster level correction for multiple comparisons in both methods, the resulting adaptation maps for WBAS show more areas of activation than the SPM method. These results suggest the WBAS could be a sensitive method for identifying regions that show adaptation without needing a priori localisation. Adaptation has the potential to tell us about the functional properties of neuronal subpopulations within a voxel. Our method could provide a powerful new way to use this technique without requiring prior assumptions about functional localisation.

Background: Tripping over obstacles is one of the major causes of falls in symptomatic Huntington’s disease (symp-HD) individuals due to motor deficits. In particular, postural instability and gait impairment may possibly maximize the risk of falls when stepping over obstacles during daily walking. Objectives: The purpose of this study was to examine the adaptive gait dynamics and balance recovery steps which are related to falls during step over an obstacle. Method: A total of thirty two individuals participated in this study, including 16 symp-HD and 16 age- and sex-matched healthy controls. The obstacle crossing task required participants to walk along a pathway and step-over one ground based bar obstacle. GAITRite electronic walkway was used to record spatiotemporal gait variables (e.g., step length, step velocity, step width). Parallel group comparisons were employed in which participants completed 6 trails of baseline walking (no obstacles) and 6 trails of stepping over the obstacle. Results: Repeated measures MANOVA revealed fundamental differences in adaptive gait patterns between the two groups. For example, compared to controls, symp-HD individuals showed reduced approach step velocity and step length, as well as distance of foot placement before and after the step-over-obstacle. The findings also showed that increased step width during step-over-obstacle could be a possible compensatory gait adaptive pattern to accommodate the reduced step length in symp-HD individuals. Conclusion: Decrements in gait performance suggests that symp-HD chose a more conservative strategy for stepping over obstacle compared to controls.

Recent neuro-computational models indicate that developmental improvements in motor planning may be subserved, at least in part, by an increased capacity to represent action internally. Presently, no study has directly examined this important relationship across a wide developmental spectrum. Accordingly, we investigated the purported association between developmental changes in grip selection planning and improvements in an individual’s capacity to internally represent action (via motor imagery). The sample comprised healthy children aged 6-7 (n = 12) and 8-12 years (n = 19). A group of adolescents aged 13-17 (n = 19) and adults aged 18-34 years (n = 19) allowed for consideration of childhood development in the broader context of neuro-motor maturation. A group of children aged 8-12 years (n = 19) with probable Developmental Coordination Disorder (pDCD) was included as a reference for atypical motor development. Participants’ proficiency in generating and/or engaging action representations was inferred from performance on the hand rotation task, a well-validated motor imagery measure. A grip-selection task designed to elicit the end-state comfort effect provided a window into motor planning efficacy. Consistent with earlier accounts, grip selection planning efficiency followed a non-linear developmental progression in neurotypical individuals. As expected, analysis confirmed that these improvements were predicted by an increased capacity to engage the internal action representation. These findings are consistent with recent computational accounts of action planning. Interestingly, no such association was found for our pDCD sample, suggesting that individuals with atypical motor skill may adopt an alternative, sub-optimal strategy to plan their grip selection.

Feeling ownership over one's thoughts and actions is a central feature of everyday life, but for people who suffer from schizophrenia, hearing intrusive voices and feeling as if strangers control their body are common, debilitating experiences. Evidence suggests that these bizarre psychotic symptoms could be related to a cognitive mechanism that attenuates sensations produced by our own actions, distinguishing them from sensations produced by the external world. Consequently, patients with deficits in sensory attenuation could misinterpret self-produced sensations as external in origin. However, though psychotic symptoms occur across all sensory domains, attenuation studies are mostly limited to stimuli from one sensory domain (typically auditory) produced by one motor area (typically the hand, during button-pressing). We measured the electroencephalogram (EEG) of 34 participants exposed to both auditory (pure tone) and visual (unstructured flash) stimuli. In addition, using eye-tracking technology, participants were able to produce these stimuli by either pressing a button or moving their eyes. We found that attenuation of self-produced sensations, indexed by auditory and visual N1-component amplitudes, significantly differed by sensory domain and motor area, and was strongest when there was a natural link between action and sensation (i.e. hand–auditory and eye–visual). This is consistent with the fact that hands can produce auditory sensations (for instance, by clapping), but eyes normally only evoke visual sensations. Our research reveals important mediators of sensory attenuation, which, along with future clinical studies, could improve our understanding of characteristic psychotic symptoms.

Eating disorders (ED) are characterized by extreme body image distortion and disturbed eating behaviours. Although it is the patients’ subjective experience there is limited objective evidence that ED involves changed perceptual mechanisms that underlie representing one’s own body. Body representations comprise the integration of visual, tactile and proprioceptive information. Key factors for this integration are 1) temporal synchrony registration (e.g., for viewing and feeling touch) and 2) viewing information plausible for one’s own body (e.g., plausible hand shape). These factors can be investigated with the rubber hand illusion (RHI) paradigm. Previous ED RHI studies provide evidence for altered multisensory integration, but imply different key factors (changed temporal synchrony versus visual hand information processing). These studies relied on subjective body-judgements that are prone to emotional and motivational bias.
To obtain an objective measure we combined the RHI with a reaching paradigm. Participants viewed an artificial hand next to their hidden hand. Both hands were touched either synchronously or asynchronously prior to ballistic reaching movements to visual targets. We tested 23 Anorexia Nervosa patients and 23 controls. Replicating our previous research, we found that temporal synchrony modulates hand location estimates and leads to significant shifts of reach trajectories (F(1,44)=21.27, p<.001). However, group does not modulate this effect of temporal synchrony (group * synchrony interaction F(1,44)=.18, p=.677). In contrast, we found a main effect of group (F(1,44)=5.24, p=.027). This suggests that in ED, estimation of hand position for reaching may be generally more sensitive to the influence of visual hand information.

We examined adult’s (N = 17) illusory susceptibility to the vertical–horizontal illusion under different orientations, sizes, and eye gazing strategies. In this illusion, a vertical line bisects and rises from the center of the horizontal line of the same length – participants typically have a tendency to perceive the length of the vertical line as being longer than the horizontal line. Our results reveal a number of interesting effects. First, susceptibility to this illusion is greater when the configuration is presented in its original position (or rotated 180 degrees) relative to when its rotated either 90 or 270 degrees, and that this difference amplifies as a function of size (Orientation × Size interaction: F(15,65) = 3.06, p < .001). Second, we also found that people were more susceptible when they fixated their gaze at the bisection point of the two lines relative to when they were permitted to free gaze but only for the smallest configuration of the illusion (Size × Gaze interaction: F(5,55) = 7.36, p < .001). Eye-tracking confirmed that the participants could maintain fixation when they were required too. A secondary experiment (same participants) revealed further that the reason why our participants were more susceptible to the illusion when it was presented either in its original or mirror position relative to when it was rotated either 90 or 270 degrees had to with their elliptical shape of their visual fields.

Colour is a fundamentally visual attribute – it has no existence in any other sensory modality; no feeling, no sound, and no taste. Rather than being a property of objects or surfaces, colour arises in the interaction of light with retinal photoreceptors, and in the post-receptoral processing in the eye and brain. How, then, do people who are blind conceive of colours? Here we address two questions which hitherto have had no substantive answer: (1) what concepts of colour develop in the minds of those who have been deprived of this visual experience all their lives, and (2) how are they constructed? We used multidimensional scaling (MDS) to generate and compare colour similarity matrices in samples of congenitally blind (CB), adventitiously blind (AB) and trichromatic adults. Participants rated the similarities of pairs of colours drawn from the set of red, oragnge, yellow, green, blue, purple, brown, white, and black. The trichromatic and AB samples produced two-dimensional MDS solutions closely resembling one another, with dimensions corresponding to red-green and light-dark axes. The CB sample produced a three-dimensional MDS solution with axes corresponding to warm-cool and light-dark dimensions, as well as a third dimension that was difficult to characterise. These data suggest that congenitally blind people form coherent concepts of colour by way of exposure to semantic associations with colour terms. People who become blind after being exposed to normal visual input – the adventitiously blind – appear to have very similar colour concepts to sighted trichromats, despite decades of blindness.

Under natural viewing conditions, visual stimuli are often obscured by occluding surfaces. To aid object recognition, the visual system actively reconstructs the missing information, as exemplified in the classic Kanizsa illusion, a phenomenon termed “modal completion”. Single-cell recordings in monkeys have shown that neurons in early visual cortex respond to illusory contours, but it has proven difficult to measure the neural correlates of modal completion in humans. Here we used electroencephalography (EEG) to measure steady-state visual evoked potentials (SSVEPs) from ‘pacman’-shaped disks arranged to induce an illusory shape (or rotated to veto the illusory shape in control trials). Diagonally opposing pairs of disks within the shape configuration were tagged with one of two flicker frequencies (5 or 8 Hz). During stimulus presentations, participants judged the orientation of a briefly flashed bar at fixation, while ignoring congruent (same orientation) or incongruent (different orientation) flanker bars that appeared on or off the illusory surface. Frequency-based analyses revealed that SSVEP amplitudes were reliably enhanced for trials in which an illusory square appeared, relative to control trials, both at the fundamental flicker frequencies (5 and 8 Hz) and at an intermodulation frequency of 13 Hz. Participants’ reaction times in the flanker task were significantly slower for incongruent versus congruent trials, and this effect was larger when an illusory surface was present than when there was no illusory surface. Our results reveal a robust neural correlate of modal completion in the human visual system, and demonstrate the influence of filling-in processes on visual attention.

High-frequency visual stimulation produces a stimulus-specific potentiation of the N1 component of the visual evoked potential (VEP). This phenomenon – “visual LTP” – has been implicated as a manifestation of synaptic plasticity, and as a possible mechanism for visual perceptual learning. Here, we assessed whether visual LTP results in perceptual learning for the inducing stimulus. We induced visual LTP using an obliquely oriented (45 or 135 degrees) Gabor stimulus presented at high frequency (8.6Hz) for five two-minute blocks (total of 10 minutes). The reflected orientation was not stimulated and served as a control stimulus. Following the induction of visual LTP, detection thresholds for Gabor stimuli at both the stimulated and non-stimulated orientations were assessed using a two-interval forced-choice (2-IFC) contrast sensitivity task. In accordance with previous research in visual LTP, the VEP data confirm that prolonged high-frequency visual stimulation produces a stimulus-specific modulation of the N1. The 2-IFC task revealed an increase in contrast threshold for detection of the Gabor stimulus that was used to induce visual LTP compared to the control orientation. That is, there does appear to be a perceptual consequence of visual LTP, but that it manifests as decreased sensitivity to the stimulus used to induce LTP.

The animate/inanimate dichotomy of visual object categorisation is well established as an organisational principle in the brain representation of objects. Recent research suggests that a representational continuum, grouping objects based on similarities between biological classes, may account for fine-grained distinctions between categories (Connolly et al., 2012; Sha et al., 2014). However, it is difficult to apply this theory to inanimate subcategories, which vary widely on numerous dimensions and have no clear basis for a continuum. Factors such as the level of familiarity or object movement may contribute to how we categorise objects, but are yet to be explored. Using magnetoencephalography, we aimed to replicate the animate/inanimate distinction with multivariate pattern analysis techniques, and subsequently evaluate the evidence for subcategory differences. Fifteen adults viewed 120 object exemplars from 12 object categories (six animate, six inanimate). Representational similarity analysis (Kriegeskorte et al., 2008) was used to test the hypothesised categorical models. Our results replicated the animate/inanimate distinction, with a significant sustained correlation between the model and the neural data from 155ms post stimulus onset. Further evaluation using a subcategory model showed sustained decoding of subcategory groupings slightly earlier, from 125ms post stimulus onset. This suggests that these activation patterns may contain important information about object subcategory membership, which goes beyond the animate/inanimate distinction.

Does the visual awareness of an object's size influence how we perceive the size of other objects? In two experiments, we manipulated visual awareness via continuous flash suppression (CFS) to investigate how the presentation of a circle, influences the apparent size of a subsequently presented circle. Stimuli consisted of Craik–O'Brien–Cornsweet illusion circles. In experiment 1, we used an adaptation paradigm to see how an adapter influenced the perceived size of a target. We found that the presentation of a small (4° in diameter) adapter increased the apparent size of a larger (8-10°) target when the adapter was visible (p < .01) but not when the adapter was suppressed from awareness (p = .94). Interestingly, the perceived size of a small (3-5°) target was not altered when preceded by a large (9°) adaptor that was visible (p = .24) or suppressed from awareness (p = .92). In experiment 2, we used a priming paradigm to see how a prime circle influenced performance on a size discrimination task. We found a main effect of congruency (F(1,15) = 6.93; p = .02) with faster reaction times when the prime was the same size as the target (relative to a different size as the target) but no interactions with object size and CFS were observed (p > .13). Taken together, the experiments highlight how the processing of object size can invoke independent mechanisms, some of which occur outside of consciousness and some which might be dependent on awareness.