Undergraduate honors paper--Mount Holyoke College, 2007. Dept. of Psychology and Education. / Includes bibliographical references (leaves 78-82).

Thesis (Ph. D.)--University of California, Riverside, 2009. / Includes abstract. Includes bibliographical references. Issued in print and online. Available via ProQuest Digital Dissertations.

Ant algorithms were first written about in 1991 and since then they have been applied to many problems with great success. During these years the algorithms themselves have been modified for improved performance and also been influenced by research in other fields. Since the earliest Ant algorithms, heuristics and local search have been the primary knowledge sources. This thesis asks the question "how is knowledge used in Ant algorithms?" To answer this question three Ant algorithms are implemented. The first is the Graph based Ant System (GBAS), a theoretical model not yet implemented, and the others are two influential algorithms, the Ant System and Max-Min Ant System. A comparison is undertaken to show that the theoretical model empirically models what happens in the other two algorithms. Therefore, this chapter explores whether different pheromone matrices (representing the internal knowledge) have a significant effect on the behaviour of the algorithm. It is shown that only under extreme parameter settings does the behaviour of Ant System and Max-Min Ant System differ from that of GBAS. The thesis continues by investigating how inaccurate knowledge is used when it is the heuristic that is at fault. This study reveals that Ant algorithms are not good at dealing with this information, and if they do use a heuristic they must rely on it relating valid guidance. An additional benefit of this study is that it shows heuristics may offer more control over the exploration-exploitation trade-off than is afforded by other parameters. The second point where knowledge enters the algorithm is through the local search. The thesis looks at what happens to the performance of the Ant algorithms when a local search is used and how this affects the parameters of the algorithm. It is shown that the addition of a local search method does change the behaviour of the algorithm and that the strength of the method has a strong influence on how the parameters are chosen. The final study focuses on whether Ant algorithms are effective for driving a local search method. The thesis demonstrates that these algorithms are not as effective as some simpler fixed and variable neighbourhood search methods.

No description available.

No description available.

Previous research on the relationship between emotion and attention has focused primarily on attention to emotionally valenced stimuli; trait anxiety and attentional biases for threat; or the relationship between emotion and attention in clinical contexts. Few studies have investigated the effect of emotion on attentional processing irrespective of the valence of the stimuli that is being attended. However, such studies are important as they shed light on issues central to emotions theory such as whether the experience of discrete emotions is associated with distinct patterns of attentional processing. In this thesis six experiments and one correlational study are described. The experimental studies investigate whether the experience of discrete emotions - specifically amusement, happiness, sadness and fear - influence attentional processing in comparison to a neutral condition. Film clips, emotional images and music were used to elicit a target emotional state. A modified version of the Attention Network Test (ANT) was used to assess three forms of attention – phasic alerting, covert exogenous orienting and executive attention. The correlational study required participants to complete a set of emotion-related questionnaires including the Basic Emotion Scale (BES) and to perform the ANT. The results suggest that: i) fear reduces executive attention costs, ii) sadness reduces intrinsic alerting, but does not influence alerting, orienting or executive attention, iii) amusement and happiness do not differentially influence alerting, orienting or executive attention, iv) individual differences in the tendency to experience high arousal negative emotions are associated with phasic alerting, i.e. faster mobilisation of attentional resources in response to an impending stimulus and v) exogenous orienting of attention may be impervious to the influence of emotion, at least in context of neutrally valenced stimuli. Results relating to anxiety, emotion regulation and attention network performance are also discussed. Taken together these findings provide only limited support for the broaden-and-build theory (Fredrickson, 1998) of positive emotions. Amusement and happiness did not result in broadening (as assessed by executive attention costs) in the present studies. An attentional narrowing effect was found for fear but not for sadness. It is proposed that fear, but not sadness, facilitates inhibition and reduces executive attention costs, indicative of more focused attention. The results here also suggest a relationship between negative emotions characterised by high arousal and phasic alerting – an aspect of attention which has received little coverage in emotions research to date. Implications relating to the use of the ANT as a measure of attentional performance, and the challenges associated with manipulating emotion in a lab setting are discussed.

No description available.

Thesis (M.S.)--Florida State University, 2003. / Title from PDF t.p. (viewed Feb. 19, 2006). Advisor: Dr. Walter R. Tschinkel, Florida State University, College of Arts and Sciences, Dept. of Biological Science. Includes bibliographical references (p. 53-58).

Not available / text

Gli scavi effettuati a Classe, a sud di Ravenna, presso i siti archeologici dell'area portuale e della Basilica di San Severo, hanno portato alla luce un numero abbondante di moneta, 2564 dall'area portuale e 224 dalla basilica, un totale di 2788 reperti monetali, di cui solo 863 sono leggibili e databili. La datazione dei materiali dell’area portuale, fondata agli inizi del V secolo, parte dal II secolo a.C. fino all’VIII secolo d.C.. La maggior parte dei reperti è relativa al periodo tra il IV e il VII secolo, il momento di massima importanza del porto commerciale, con testimonianza di scambi con altri porti del bacino mediterraneo, in particolare con l’Africa del Nord e il Vicino Oriente. La documentazione proveniente dalla Basilica di San Severo, fondata alla fine del VI secolo per la custodia delle reliquie del santo, mostra un trend diverso dal precedente, con monetazione che copre un arco cronologico dal I secolo a.C. fino al XIV secolo d.C.. La continuità dell’insediamento è dimostrato dall’evidenza numismatica, seppur scarsa, fino alla costruzione del monastero a sud della basilica, l’area dalla quale provengono la maggior parte delle monete. I quantitativi importanti di monetazione tardoantica, ostrogota e bizantina, in particolare di tipi specifici come il Felix Ravenna, ipoteticamente coniato a Roma, oppure il ½ e il 1/4 di follis di produzione saloniana emesso da Giustiniano I, hanno concesso uno studio dettagliato per quello che riguarda il peso, le dimensioni e lo stile di produzione di queste emissioni. Questi dati e la loro distribuizione sul territorio ha suggerito nuove ipotesi per quello che riguarda la produzione di questi due tipi presso la zecca di Ravenna. Un altro dato importante è il rinvenimento di emissioni di Costantino VIII, alcune rare e altre sconosciute, rinvenute solo nel territorio limitrofo a Classe e Ravenna. / The excavations carried out in Classe, South of Ravenna, at the sites of the harbour area and the nearby Basilica of S. Severo, have brought to light a large amount of coins, 2564 from the commercial harbour and 224 from the Basilica, a total of 2788 specimens of which only 863 are legible and can be dated, however still providing some important information. The dating of the numismatic evidence from the harbour area, that was founded in the early 5th century, ranges from 2nd century BC to late 8th century AD. Most of the finds date from the 4th to the 7th century AD, the most important period for the harbour, with evidence of trade with other areas of the Mediterranean basin, particularly with North Africa and the Near East. The evidence of the Basilica of San Severo, founded in the late 6th century for the safekeeping of the saint relics, shows a quite different pattern, with a chronological span that ranges from the 1st century BC to the 14th century AD. The occupation of the site is continuous but the numismatic evidence is scarce, until the building of the monastery South of the Basilica, the area from which most of the coins were retrieved. The large amounts of late antique, Ostrogothic and Byzantine coinage, particularly certain specific types, like the Felix Ravenna, hypothetically minted in Rome, or ½ and ¼ follis Salona emissions of Justinian I, have allowed some in-depth analysis, specifically related to weight, size and style. This type of data, as well as the pattern of distribution, have allowed to suggest new hypothesis about their production at the mint of Ravenna. Also important is the finding of coinage of Constantine VIII, rare and unknown emissions found only in the neighbourhood of Classe and Ravenna.