The present study tested the hypothesis that distractor-based facilitation of visual search occurs because familiar distractors are processed and rejected in groups. We recorded participants’ eye movements during a visual search task to determine if familiar distractors were associated with an increased average distance between fixations and distractors. The study provided convergent evidence of a strong relation between search efficiency and distractor familiarity, wherein the distance between fixations and distractors increases with the efficiency of search. Further examination of eye movements suggested that the grouping of familiar distractors resulted in an efficient scanning of the search display by increasing the area of the display effectively processed during each fixation and therefore reducing the need to fixate individual distractors.

The present study tested the hypothesis that distractor-based facilitation of visual search occurs because familiar distractors are processed and rejected in groups. We recorded participants’ eye movements during a visual search task to determine if familiar distractors were associated with an increased average distance between fixations and distractors. The study provided convergent evidence of a strong relation between search efficiency and distractor familiarity, wherein the distance between fixations and distractors increases with the efficiency of search. Further examination of eye movements suggested that the grouping of familiar distractors resulted in an efficient scanning of the search display by increasing the area of the display effectively processed during each fixation and therefore reducing the need to fixate individual distractors.

This work aims to integrate Byrne’s theory of visual salience computation (2006) with Salvucci’s model of eye movements (2001) by testing participants on a visual search task similar to Findlay (1997). By manipulating the number, salience, and spacing of targets, participants exhibited the global effect averaging phenomena during the first recorded saccade, whereby short‐latency saccades land in between adjacent objects. Previous work has argued that the saccadic targeting system causing the averaging is influenced both by the salience and arrangement of objects displayed (Rao, Zelinsky, Hayho, & Ballard, 2002). However, to accurately account for these results, we did not have to couple the salience system with the saccadic targeting system. Instead, the systems work sequentially and in isolation, whereby the salience system simply hands off the next object to examine to the targeting system, whose accuracy depends only on saccadic latency and the location of the targeted and non‐targeted items.

Fill-in-the-blank multiple choice questions (MCQs) play an important role in the educational field, but the manual generation of them is quite resource-consuming, so it has gradually turned into an attractive NLP task. Thereinto, question creation itself has become a mainstream NLP research topic, while distractor (wrong alternative) generation (DG) still remains out of the spotlight. Although several studies on distractor generation have been conducted in recent years, there is little previous work on languages other than English. The goal of this thesis is to generate multilingual distractors in Chinese, Arabic, German, and English across domains. The initial step is to construct small-sized multilingual scientific datasets (En, Zh, Ar, and De) and general datasets (Zh and Ar) from scratch. Considering that there are limited multilingual labelled datasets, unsupervised experiments based on WordNet, Word Embedding, transformer-based models, translation methods, and domain adaptation are conducted to generate their corresponding candidate distractors. Finally, the performance of methods is evaluated against our newly-created datasets, where three metrics are applied. Lastly, statistical results show that monolingual transformer-based together with translation-based methods outperform the rest of the approaches for multilingual datasets, except for German, which reaches its highest score only through the translation-based means, and distractor generation in English datasets is the simplest to implement, whereas it is the most difficult in Arabic datasets.

The capacity of the visual system to process information about multiple objects at any given moment in time is limited. This is because not all information can be processed equally or in parallel and subsequently reach consciousness. Previous research has utilized behavioural experiments to explore visual attention. More recently research, however, has used electroencephalography (EEG) to measuring the electrical brain activity in the posterior scalp. By time locking visual stimulus events to fluctuations in scalp activity researchers have been able to estimate the time course of attentional changes by measuring changes in these event-related potentials (ERP). One component in particular (N2pc) has been a reliable tool in measuring either the suppression of, or the shift of attentional to, both ignored and attended items in the visual scene. The N2pc is measured by comparing the ERP activity contralateral and ipsilateral to the visual field of interest. More recently, evidence has been presented that the mechanisms of attention thought to be represented by the N2pc (suppression and attentional selection) could be separated into different ERP components (Pd: indexing attentional suppression of an ignored item; and Nt: indexing attentional selection of the target) and measured independently. In six experiments, using ERPs, this thesis employs these components to explore the mechanisms and strategies of the human attentional system. Additionally, this thesis focuses on the impact of different types of simultaneous processing load on the attentional system and how the mechanisms of this system are influenced. Experiment 1 explores the idea that the type or valence of information to be ignored may influence the ability to suppress it. Results of this experiment 4 show that neither the type nor valence of the irrelevant information modulated the amplitude of the distractor positivity (Pd), indicating suppression of the irrelevant distractor was not altered. Noted in experiment 1 was also the presence of an early negativity (Ne) that appeared to represent attentional capture of the ignored lateral stimulus. Experiment 2 demonstrated that the valence of the lateral target did not alter the target negativity (Nt), indicating a different pattern of results between the Nt and the N2pc reported in previous studies (e.g. Eimer & Kiss, 2007; Feldmann-Wüstefeld et al., 2010). Experiment 2 also showed a similarity of the target negativity (Nt) to the early negativity (Ne; the N2pc like component observed in exp 1) toward face and non-face stimuli. This comparison supported the idea that the early negativity (Ne) reflected attentional capture of the ignored lateral distractor and as a result was relabelled the distractor negativity (Nd) in subsequent experiments. Experiment 3 showed that the salience of the lateral image did not modulate the Pd as should be the case if the Pd reflected sensory-level processing. An early contralateral negativity (similar to the Nd observed in exp 1) was altered by the salience of the distractor which added support to the hypothesis that this reflects attentional capture of the lateral ignored image. Experiment 4 attempted to manipulate working memory (WM) to assess the effect of WM load on attentional capture and suppression. While the results did indicate modulation of suppression under WM load, the limitations of the design of experiment 4 made any definitive interpretation of the results unreliable. The results of experiment 5 showed that suppression, as indexed by the Pd, was not altered by cognitive load. However, reductions in attentional capture under high cognitive load, as indexed by the distractor negativity (Nd), were observed and contradict the results of previous experiments (c.f. Lavie & De Fockert, 2005) 5 where cognitive load resulted in an increase in attentional capture. Although, there appears to be some issue in the authors interpretation of the results of these experiments (see chapter 6 for discussion). The results of Experiment 6 show the opposite effect with a significant increase in the laterality of the Pd under high perceptual load. A similar increase in the laterality of the Pd was not reflected in terms of valence though, where suppression of threat related distractors was not altered under high perceptual load. The hypothesis that an increase in perceptual load will result in a decrease in attentional capture was generally supported by the results of experiment 6. Under high perceptual load angry face distractors captured attention, as indexed by the laterality of the Nd, with neutral face distractors showing a reduction in attentional capture. While under low perceptual load, both angry and neutral face distractors resulted in a significant (and similar) laterality of the Nd. The thesis concludes by discussing issues concerning Lavie’s Load Theory of attention and outlines some potential misinterpretations of previous data that have led to the proposal that cognitive load results in a decrease in attentional resources and therefore a decrease in attentional capture of ignored stimuli. It is argued in this thesis that the results of Lavie and de Fockert (2005), which concluded that the increase in cognitive load resulted in a decrease in attentional capture, are more likely to be due to changes in attentional capture (i.e. a reduction) and changes in RT (i.e. an increase), under cognitive load being separate responses to the availability of resources, one that focusses attention on the goal directed task and the other that results in extended processing time to carry out the more difficult task. In this case both ‘changes’ appear to work to prioritise resources in favour of the goal directed task.

The Remote Distractor Effect (RDE) is a robust phenomenon where a saccade to a lateralised target is delayed by the appearance of a distractor in the contralateral hemifield (Walker, Kentridge, & Findlay, 1995). The main aim of this thesis was to test whether the RDE generalises to response modalities other then the eyes. In Chapter 2, the RDE was tested on saccadic and simple manual keypress responses, and on a choice discrimination task requiring a covert shift of attention. The RDE was observed for saccades, but not simple manual responses, suggesting that spatially oriented responses may be necessary for the phenomenon. However, it was unclear whether distractor interference occurred in the covert task. Chapter 4 compared the effects of distractors between spatially equivalent tasks requiring saccadic and manual aiming responses respectively. Again, the RDE was observed for the eyes but not for the hands. This dissociation was also replicated in a more naturalistic task in which participants were free to move their eyes during manual aiming. In order to examine the time-course of distractor effects for the eyes and the hands, a third experiment investigated distractor effects across a wider range of target-distractor delays, finding no RDE for manual aiming responses at distractor delays of 0, 100, or 150 ms. The failure of the RDE to generalise to manual aiming suggests that target selection mechanisms are not shared between hand and eye movements. Chapter 5 further investigated the role of distractors during covert discrimination. The first experiment showed that distractor appearance did not interfere with discrimination performance. A second experiment, in which participants were also asked to saccade toward the target, confirmed the lack of RDE for covert discrimination while saccades were slower in distractor trials. The dissociation between covert and overt orienting suggests important differences between shifts of covert attention and preparation of eye movements. Finally, Chapter 6 investigated the mechanism driving the RDE. In particular it was assessed whether saccadic inhibition (Reingold & Stampe, 2002) is responsible for the increase in saccadic latency induced by remote distractors. Examination of the distributions of saccadic latencies at different distractor delays showed that each distractor produced a discrete dip in saccadic frequency, time-locked to distractor onset, conforming closely to the character of saccadic inhibition. It is concluded that saccadic inhibition underlies the remote distractor effect.

One means by which inhibitory control in selective attention may be studied is with the negative priming (NP) procedure. It is widely assumed that children are characterised by reduced capacity for inhibition (Diamond, 2002) and that inhibitory dysfunction is a key characteristic of children and adolescents with ADHD (Barkley, 1997). This should translate into reduced NP effects for these populations. In this dissertation, four studies using the NP procedure find no evidence for reduced inhibitory function in typical children or in adolescents with ADHD. Study 1 examined the magnitude of NP in children compared with adults. An important line of support for the idea that children suffer an inhibitory decrement has been based an empirical report suggesting that conceptual (identity or semantic) NP effects, assumed to reflect the by-product of distractor inhibition, while consistently found in adults are lacking in children (Tipper, Bourque, Anderson, & Brehaut, 1989). In Study 1, the opposite result was found. Study 2 compared NP effects between 7-year-old children and adults while replicating the respective methodologies of the only two studies to explore conceptual NP effects in developmental populations to date (Pritchard & Neumann, 2004, vs. Tipper et al., 1989) to determine the nature of the divergent results between these studies. In Study 2, it was found that distractor inhibition effects are comparable between children and adults when a NP task contains trials in which the distractor stimulus is consistently incongruent with the target stimulus, but that children may be more susceptible than adults to divide attention between target and distractor when a NP task contains a number of trials in which target selection difficulty is reduced. These are critical new findings, highlighting that reduced NP may often relate to methodological artifacts, and when considered in the light of current theories of NP, are also problematic for anti-inhibitory accounts of NP. Having distinguished more definitively the role of inhibition in developmental NP effects, Studies 3 and 4 explored whether the inhibitory process underpinning NP was implicated in young persons with ADHD. To date, evidence for NP in ADHD populations is equivocal. Study 3 found no evidence for a reduced NP effect in ADHD devoid of a corresponding diagnosis. Study 4 found that conduct and oppositional defiant disorders had the potential to confound the evaluation of NP in ADHD. Taken together, results in Studies 1 - 4 parallel very recent results in the literature on NP in older adults and adult psychopathology where presumed reductions of NP in these populations may also be accounted for by methodological artifacts (Buchner & Mayr, in press). It is concluded that NP may reflect a primitive and robust form of inhibitory processing, one that develops early and one that is often the last to deteriorate.

La théorie dominante suggérant une séparation entre une vision consciente pour la perception et une vision non consciente pour l'action au sein du système visuel fut particulièrement discutée au cours des dernières décennies. La thèse défendue ici se positionne dans le cadre d'une approche alternative reposant sur une évaluation conjointe des réponses motrice et perceptive en présence de stimuli au seuil perceptif. Les travaux réalisés dans ce contexte ont initialement porté sur les temps de réaction et ont contribué au développement d'un modèle psychophysique postulant que les réponses perceptives et motrices dépendraient d'un signal sensoriel unique mais de seuils de décision spécifiques. Les trois études réalisées au cours de cette thèse ont permis de tester ces propositions à partir de travaux portant sur le contrôle moteur manuel. Tout en confirmant une forte association entre les traitements visuels pour la perception et pour l'action, ces travaux ont permis de mettre en évidence un rôle important des conditions expérimentales (par exemple le contraste des stimuli et la présence ou non d'un masque visuel) dans les résultats obtenus. Les résultats se sont révélés compatibles avec les modèles neurophysiologiques du masquage visuel qui postulent que la réponse neurale à une stimulation visuelle est constituée d'une vague d'activation feedforward associée à la présence physique de la stimulation et de boucles de rétroaction liées à la perception consciente de celle-ci. Nos travaux ont également permis d'adapter au contrôle moteur manuel le modèle de décision à signal unique initialement développé dans le cadre des études portant sur les temps de réaction. / The dominant position of a separation between a conscious vision for perception and an unconscious vision for action within the visual system has been particularly discussed in the last decades. The present dissertation is to be placed in the context of an alternative approach consisting in evaluating jointly both perceptual and motor responses in the presence of near-treshold visual stimuli. Previous work within this framework, which have mainly concerned reaction times, have contributed to develop a psychophysical model in which perceptual and motor decision are taken relatively to the same single incoming signal but are based on different tresholds. The three studies conducting during this PhD aimed to testing these proposals in experiments involving manual motor control. While confirming the strong link between perceptual and motor processing within the visual system, the results obtained in these studies underlined the importance of experimental parameters such as the contrast of the stimuli and the presence of visual masks. These results are congruent with neurophysiological models of visual masking, which postulate that the neural response to a visual stimulus is composed of a transient feedforward sweep of activation related to the presence of as stimulus and recurrent feedback loops linked to the conscious perception of this stimulus. Our work also led to the adaptation to manual motor control of the single signal decisional model initially developed on the basis of reaction time studies.

Complex behaviors require selectively attending to task-relevant items, and ignoring conspicuous, irrelevant items. For example, driving requires selectively attending to other cars on the road while ignoring flashing billboards. Dominant models of attentional control posit that we avoid distraction by biasing attention towards task-relevant items, and our ability to avoid distraction depends on the strength and specificity of this bias. I find that a strong, specific bias towards task-relevant items is insufficient for preventing distraction. Instead, preventing distraction also requires past experience ignoring distractors. I also find that long-term memory systems, rather than visual short-term memory or priming memory systems, maintain this experience. Based upon these findings, I propose that effective attentional control not only demands a strong, specific bias towards task-relevant items, but also requires that observers learn to ignore conspicuous, irrelevant items.

Der Distraktoreffekt beschreibt eine Verlängerung der Fixationsdauer als Folge der Darbietung irrelevanter Störreize (Distraktoren). In der Dissertation wurde der Frage nachgegangen, ob sich der Distraktoreffekt funktionell auch im Rahmen des Konzepts der Orientierungsreaktion (OR – Sokolov, 1963) betrachten lässt. Reizeigenschaften wie Neuheit und Relevanz wurden in der Vergangenheit als wesentliche Auslösebedingungen einer OR diskutiert. Im Rahmen dieser Arbeit wurden beide Merkmale untersucht, um die Plausibilität des Zusammenhangs von Distraktoreffekt und OR empirisch zu testen. Hierfür kam in allen Studien ein experimentelles Paradigma zum Einsatz, bei dem Distraktoren blickkontingent während einer freien Bildbetrachtungsaufgabe dargeboten wurden. Der Einfluss von Neuheit wurde durch Untersuchungen zur Habituation des Distraktoreffekts geprüft. Die Ergebnisse der Studien ergaben deutliche Anzeichen einer Habituation im okulomotorischen Verhalten (Abnahme der Fixationsdauerverlängerung) sowie in elektrophysiologischen Parametern der kortikalen Verarbeitung des Distraktors, im Sinne einer Abnahme der N1 Komponente des EKPs. Somit konnte also eine Modulation durch Neuheit eines Distraktors nachgewiesen werden. Die Frage nach dem Einfluss von Relevanz wurde experimentell durch Manipulation aufgabenbezogener Relevanz von Distraktoren, durch Verwendung emotionaler Distraktorinhalte (unterschiedliche emotionale Gesichtsausdrücke) und durch Darbietung neutraler Distraktoren in einem emotionalen Kontext untersucht. In keiner der drei Studien konnten eindeutige Befunde, die für eine Modulation des Distraktoreffekts durch Relevanz sprechen, aufgezeigt werden. Zusammengenommen konnte der postulierte Zusammenhang von Distraktoreffekt und OR nicht bestätigt werden. Im letzten Teil der Arbeit wurde das Distraktorparadigma genutzt, um zu prüfen, ob Unterschiede in der Art der Verarbeitung und Aufmerksamkeitsausrichtung innerhalb einer Fixation die Verarbeitung eines Distraktors beeinflussen. Frühere Überlegungen aufgreifend (Pannasch & Velichkovsky, 2009) wurden Verarbeitungsunterschiede anhand des Musters von Sakkaden, die eine Fixation umgeben, operationalisiert. Die distraktorinduzierte Verlängerung der Fixationsdauer war am größten, wenn die betroffene Fixation in kurze vorhergehende und nachfolgende Sakkaden eingebettet war und am kleinsten im Kontext langer Sakkaden. In parallel aufgezeichneten elektrophysiologischen Daten zeigten sich Unterschiede zwischen den sakkadischen Kontextbedingungen vor allem als Variation der distraktorbezogenen P2 Komponente. Diese Ergebnisse bestätigten die Annahme, dass anhand des sakkadischen Kontexts Unterschiede in der Art bzw. dem Modus der Verarbeitung innerhalb einer Fixation identifiziert werden können und sich in Veränderungen von Verhaltensparametern und kortikalen Aktivitätsmustern der Distraktorverarbeitung widerspiegeln. Diese Ergebnisse unterstützen einerseits frühere Annahmen bezüglich funktioneller Unterschiede von Blickbewegungsmustern (Velichkovsky, Joos, Helmert, & Pannasch, 2005) und legen außerdem eine Beteiligung kortikaler Areale an distraktorinduzierten Prozessen der sakkadischen Hemmung nahe.