Prever eventos ambientais, com base em memórias sobre regularidades passadas, é uma das funções fundamentais de sistemas nervosos complexos. Eventos ordenados serialmente ou sequências estruturadas de estímulos permitem extrair informação passível de descrição formal que define seu padrão serial. Esse padrão inclui informações temporais e espaciais que facultam prever os próximos eventos da sequência, possibilitando a preparação prévia do organismo para lidar com sua ocorrência. Não surpreende que animais, incluindo o ser humano, aprendam, de maneira relativamente rápida, sobre regras e estruturas de padrões sequenciais de estímulos. O uso de tarefas de tempo de reação serial (TRS) é recorrente em estudos envolvendo a formação de associações, antecipação, atenção, as bases da memória e aprendizagem de relações complexas. Resumidamente, voluntários devem responder a estímulos apresentados em sequências repetitivas ou aleatórias. Com o treino, há redução no tempo de reação a cada estímulo, refletindo a aprendizagem de relações percepto-motoras. Essa redução, porém, é maior na sequência repetitiva em relação à sequência aleatória, indicando um aprendizado também sobre a sequência repetitiva, mesmo quando o voluntário não a percebe (conscientemente) e seja incapaz de relatar sua existência. Trata-se, portanto, de uma aquisição (inicialmente) implícita. A complexidade de uma sequência de estímulos pode ser expressa quantitativamente por meio de uma ferramenta matemática proposta por Shannon (1948), a entropia informacional (EI), que considera, entre outras coisas, a probabilidade de ocorrência dos estímulos em diferentes níveis. No presente trabalho, avaliamos em que extensão o conceito de EI permite prever o desempenho de ratos na tarefa de TRS envolvendo sequências com diferentes níveis de complexidade. Ratos foram treinados a reagir (1) a uma sequência repetitiva de estímulos, cuja quantidade de EI no nível 1 (que relaciona os estímulos da sequência 2 a 2) era 2,75. Após atingirem um nível assintótico de desempenho, os animais foram expostos (2) a sequências variáveis de estímulos com a mesma quantidade de EI no nível 1, porém, com maior quantidade de EI no nível 2 (que relaciona os estímulos da sequência 3 a 3). Numa etapa posterior os animais foram expostos (3) a uma nova sequência repetitiva de estímulos, cuja quantidade de EI no nível 1 era 3,00; por fim, os animais foram submetidos (4) a sequências variáveis com a mesma quantidade de EI no nível 1em relação à sequência anterior, porém, com maior quantidade de EI no nível 2 . Os resultados mostraram que os ratos aprenderam sobre os padrões seriais e, mais interessante, que seu desempenho esteve fortemente correlacionado à quantidade de EI no nível 2. Em outras palavras, quanto maior a EI, pior o desempenho dos animais tanto em termos do tempo de reação como em termos da percentagem de respostas corretas. Portanto, o conceito de EI permite não apenas quantificar a complexidade de sequências empregadas em estudos envolvendo aprendizagem serial, mas também prever o desempenho dos animais. / Prediction of environmental events, relying on memories of past regularities, is one of the fundamental functions of complex nervous systems. Sequences of serially ordered stimuli allow extracting information that defines its serial pattern. These patterns allow prediction of the next item in a sequence of events, facultating previous preparation to deal with its occurrence. Not surprisingly, animals, including humans, can identify rules present in serial structures of stimuli. Serial reaction time tasks (SRTT) have been extensively used in studies involving association, anticipation, attention, and learning and memory. Typically, subjects have to react to stimuli presented either in random or in repetitive sequences. As training proceeds, reaction time to each stimulus decreases, reflecting acquisition of this perceptual-motor skill. However, reaction time reduction is greater for repetitive sequences relative to the random sequences, indicating acquisition about the repetitive structure of the sequence. In human beings, this may occur even when the subject in uncapable of reporting the existence of a sequence, indicating that the acquisition was (at least initially) implicit rather than explicit. The complexity of a sequence of stimuli, at different levels, may be quantifyed by means of a mathematical tool proposed by Shannon (1948), the information entropy (IE). In this study we evaluated to which extent IE can predict performance of rats in SRTT involving sequences of stimuli organized at different levels of complexity. Rats were trained to react (1) a repeated sequence of stimuli which IE at the level \"1\" (i.e., expressing to which extent a given item allow prediction of the next) was 2.75. After reaching an asymptotic level of performance, the animals were exposed (2) a variable sequence of stimuli with the same amount of IE in the level \"1\", but with more IE in the level \"2\" (i.e., expressing to which extent two given items allow prediction of the next). Later the animals were exposed to (3) a new repeated sequence of stimuli, which IE at the level \"1\" was 3.00. Finally, the animals were submitted to (4) a random sequence of stimuli with the same amount of IE at the level \"1\", i.e., 3.00, but with greater IE in level 2. Results showed that rats learned about the serial patterns and, more interestingly, their performance strongly correlated to the amount of IE at the level \"2 \", both in terms of reaction times and in terms of percentage of correct responses. Therefore, IE allows not only to quantify complexity of sequences in studies involving serial learning, but also to predict performance of the subjects.

Experiências armazenadas acerca de regularidades passadas permitem a previsão do ambiente e, consequentemente, a possibilidade de ações antecipadas. Esta capacidade cognitiva é expressa em modelos de aprendizagem de sequências, que são capazes de acessar a previsibilidade das sequências de eventos e gerar descrições do desempenho em protocolos experimentais como a tarefa de tempo de reação serial. Nos experimentos 1, 2 e 3 deste trabalho, a abordagem informacional foi aplicada à descrição do desempenho na tarefa de tempo de reação serial. A relação entre medidas de entropia e desempenho na tarefa de tempo de reação serial envolvendo diferentes tipos de sequência foi investigada nos Experimentos 1a e 1b. As medidas de entropia foram feitas pelo processamento das frequências de eventos das sequências (i.e., pares, trios, quadras etc). Os resultados revelaram que a entropia informacional das sequências é um bom descritor do desempenho: (1) sequências de baixa entropia são realizadas mais rapidamente e são mais frequentemente reconhecidas ao final da sessão do que as de alta entropia; (2) uma curva sigmóide relaciona valores de entropia aos de tempo de reação: parâmetros \"min\" (tempo de reação com a previsão total), \"max\" (tempo de reação sem previsão) e \"x50\" (valor de entropia relacionada ao limiar de previsão); (3) o treinamento torna previsíveis sequências de alta entropia (o \"x50\" aumenta com o treinamento); e (4) com o treinamento, mais elementos prévios da sequência passam a ser utilizados para a previsão do próximo elemento. A relação entre desempenho e expectativas probabilísticas geradas durante o treinamento foi investigada no Experimento 2. Esse experimento envolveu múltiplas combinações de sequências de treino e teste, aplicadas a voluntários em sessões únicas. A diferença entre as previsibilidades das sequências de teste e treino foi quantificada pela distância de Kullback-Leibler: pequenas distâncias indicam que o treino proporciona boa previsão sobre o teste. Desconsiderando os efeitos de entropia (descrito no Experimento 1), a distância de Kullback-Leibler entre as sequências de teste e treino está relacionada ao desempenho: (1) distâncias pequenas levam à manutenção das expectativas (prévias) e tempos de reação curtos; (2) distâncias grandes levam à negligência das previsões e tempos de reação intermediários; e (3) distâncias intermediárias estão relacionadas a um conflito entre as estratégias de manutenção e negligência das expectativas, e geram tempos de reação elevados. Portanto, a flexibilidade das previsões ocorre em distâncias pequenas; uma estratégia alternativa, de negligência das previsões, é adotada em distâncias grandes. A estratégia desenvolvida nos Experimentos 1a e 1b foi útil para avaliar, no Experimento 3, a equivalência funcional entre treinamento imaginativo e real na aprendizagem de sequências. Este experimento envolveu voluntários testados na tarefa de tempo de reação serial ao longo de várias sessões de treinamento imaginativo e real. Os desempenhos durante o treinamento imaginativo e real foram descritos e comparados; o experimento mostrou também que a previsibilidade da sequência acessada por meio do treinamento imaginativo pode ser expressa posteriormente no desempenho real da tarefa. No entanto, o limite de previsibilidade das sequências acessado pelo treinamento imaginativo é inferior ao limite acessado por treinamento real, descrita pelo menor \"x50\" do (1) treinamento imaginativo em relação ao treinamento real e (2) desempenho real após o treinamento imaginativo em relação ao desempenho real após o treinamento real. Em conclusão, é possível afirmar que o modelo de entropia informacional é capaz de descrever a variabilidade do desempenho na tarefa de tempo de reação serial. Estes achados apóiam a existência de um princípio geral de acesso à previsibilidade para explicação da aprendizagem e memória. / Stored experiences of past regularities allow the prediction of the environment and, consequently, the possibility of anticipatory actions. This cognitive capacity is expressed in models of sequence learning, which are able to access the predictability of sequences of events and to generate descriptions of performance on experimental protocols as serial reaction time task. In Experiments 1, 2 and 3 of this work the informational framework was applied to the description of performance in serial reaction time task. The relationship between entropy measures and performance on serial reaction time task involving multiple sequence types was investigated on Experiments 1a and 1b. The entropy measures were done by processing the frequencies of events of the sequences (i.e. pairs, triads, quads etc). The results revealed that information entropy of the sequences is an impressively good descriptor of performance: (1) low entropy sequences were performed more rapidly and were more frequently recognized in the end of the session than the high entropy ones; (2) a sigmoid curve relates entropy to reaction time: parameters \"min\" (reaction time with total prediction), \"max\" (reaction time with no prediction) and \"x50\" (entropy value related to threshold of prediction); (3) training makes high entropy sequences predictable (the \"x50\" increases with training); and (4) with training, more previous elements of sequence are used for prediction of the next one. The relationship between performance and probabilistic expectancies generated during training was investigated on Experiment 2. This experiment involved multiple arrangements of training and testing sequences, applied to volunteers on single sessions. The difference between the predictabilities of testing and training sequences was quantified by the Kullback-Leibler divergence: small divergence indicates that training provides good prediction on testing. Disregarding the entropy effects (described on Experiment 1), Kullback-Leibler divergence between training and testing sequences is related to performance: (1) short divergences lead to (previous) predictions maintenance and low reaction times; (2) large divergences lead to predictions negligence and intermediate reaction times; and (3) intermediate divergences are related to conflict between the strategies of maintenance and negligence of predictions, and generate high reaction times. Therefore, the flexibility of predictions occurs on short divergences; an alternative strategy, of predictions negligence, is adopted on large divergences. The strategy developed on Experiments 1a and 1b was useful to evaluate, on Experiment 3, the functional equivalence between imagery and actual training on sequence learning. This experiment involved volunteers tested on serial reaction time task along multiple imagery and actual training sessions. Performances on both imagery and actual training were described and compared; the experiment also showed that the sequence predictability accessed on imagery training can be expressed on posterior actual performance. However, the limit of sequence predictability accessed by imagery training is lower than the limit accessed by actual training, described by the lower \"x50\" of (1) imagery training compared to actual training and (2) actual performance after imagery training compared to actual performance after actual training. In conclusion, it is possible to state that the entropy model is able to describe the variability of performance on serial reaction time task. These findings support the existence of a general principle of accessing the predictability to explain learning and memory.

This thesis consisted of experiments designed to explore the construct of attention and investigate the disruptive effects of psychotomimetics, with a specific focus on NMDA antagonists. Phencyclidine (PCP) was administered through a variety of treatment regimens in order to to determine the ability of inducing cognitive-specific disruptions in attentional functioning. The hypothesis that sub-chronic exposure to PCP would result in persistent attentional impairment was tested, using the 5-choice serial reaction time task (5-CSRTT). The 5-CSRTT assesses not only visuospatial attention, but also components of impulsivity, compulsivity, speed of processing and motivation. It was determined that an additional task-related intervention that increased the attentional load was required to elucidate attentional impairment following sub-chronic PCP treatment. The ability of rats to perform the modified version of the 5-CSRTT, known as the 5-choice continuous performance test (5C-CPT), was investigated. The 5C-CPT was implemented to provide a task that may have greater analogy to the human CPT, than the original 5-CSRTT. The consequence of dopaminergic D1 system activation was investigated. It was revealed that D1 partial agonism improved attentional performance in a baseline-dependent manner. Following successful acquisition of the task, it was shown that repeated PCP treatment induced cognitive disruption that was cognitive-specific, and not confounded by generalised response disruption. Furthermore, a partial attenuation of the PCP-induced performance disruption was achieved following administration of the D1 partial agonist, SKF 38393. Moreover, sub-chronic PCP treatment was shown to impair 5C-CPT performance in the drug-free state. However, an additional challenge that further increased the attentional load was needed to elucidate a performance deficit. This highlighted that sustained attention/vigilance is sensitive to persistent impairment following sub-chronic PCP administration in a manner consistent with deficits observed in schizophrenia patients. This prompted the investigation that tested the hypothesis that sub-chronic PCP treatment could induce enduring structural deficits in regions associated with attentional performance. Magnetic resonance imaging (MRI) was conducted, in conjunction with 5-CSRTT and pre-pulse inhibition (PPI). It was revealed that sub-chronic PCP treatment resulted in morphological brain abnormalities in brain regions associated with 5-CSRTT performance. This was coupled with deficits in sustained attentional performance following an increase in attentional load, yet PPI was unaffected. Taken together, these findings suggested sub-chronic PCP treatment impairs attentional functionality, an effect that dissociates between effortful and passive attentional processes.

Das aktuelle Modell zur Pathophysiologie der Zwangserkrankung (OCD) geht von fronto-striatalen Dysfunktionen aus. Damit werden Beeinträchtigungen im prozeduralen Lernen, das anhand der Serial Reaction Time Task (SRTT) erfasst werden kann, in Verbindung gebracht. Die Befunde zu Defiziten von OCD-Patienten in der SRTT sind widersprüchlich, was auf Unterschiede im methodischen Vorgehen sowie die zusätzliche Auslastung des Arbeitsgedächtnisses durch die Vorgabe einer Gedächtnisaufgabe zurückgeführt wird. Weiterhin ist unklar, ob die vermutete fronto-striatale Dysfunktion und die Defizite im prozeduralen Lernen kennzeichnend für OCD sind oder auch bei anderen Störungsbildern auftreten. Die prozedurale Lernleistung von OCD-Patienten wurde mit der Lernleistung von gesunden Probanden, Patienten mit einer Angststörung und Patienten mit einer Depression verglichen. Weiterhin wurde die prozedurale Lernleistung unter der Vorgabe der SRTT alleine (single-task) und bei gleichzeitiger Vorgabe einer Gedächtnisaufgabe (dual-task) untersucht. OCD-Patienten zeigten im Vergleich zu gesunden Probanden Defizite im prozeduralen Lernen – sowohl unter der single-task als auch unter der dual-task Bedingung. Im Vergleich zu Patienten mit einer Angststörung zeigte sich lediglich in der dual-task Bedingung eine tendenzielle Beeinträchtigung der OCD-Patienten. Keine Unterschiede ergaben sich in der prozeduralen Lernleistung zwischen OCD-Patienten und Patienten mit einer Depression. Die stärkste Beeinträchtigung der prozeduralen Lernleistung in den verschiedenen Symptomdimensionen (Kontrolle, Waschen, Symmetrie, Horten) der OCD zeigte sich in der Dimension Horten. Außerdem waren die Defizite im prozeduralen Lernen stärker bei Patienten mit einem früheren Beginn der Zwangsstörung ausgeprägt. Die Befunde stehen im Einklang mit bisherigen Ergebnissen, wonach bei OCD eine Beeinträchtigung im prozeduralen Lernen vorliegt, was für die Annahme einer fronto-striatalen Dysfunktion bei OCD spricht. / Obsessive-compulsive disorder (OCD) is seen as a disease that implicates fronto-striatal dysfunctions. These dysfunctions are hypothesized to be related to neuropsychological deficits. One of the putative deficits regards procedural learning, which can be assessed by using the serial reaction time task (SRTT). So far, the results regarding procedural learning in OCD patients are inconsistent, which is attributed to differences in methods, like the implementation of a secondary task (dual-task condition) using a concurrent working memory load. Moreover, it is still uncertain whether the fronto-striatal dysfunctions and the deficits in procedural learning are specific to OCD or if they can also be found in other mental illnesses. Procedural learning performance as measured with the SRTT was compared in OCD patients, healthy subjects, patients with anxiety disorder and depression. Additionally, procedural learning was examined under single- (SRTT) and dual-task (SRTT plus a secondary task) conditions. In comparison with healthy subjects, procedural learning was impaired in OCD patients – both in the single- and in the dual-task condition. Yet in the dual task condition, procedural learning was diminished in both groups. Compared to patients with anxiety disorder, OCD patients were impaired in procedural learning only in the dual-task condition, while there was no difference between OCD and depression in procedural learning. Regarding the different symptom dimensions of OCD (checking, washing, symmetry, hoarding), patients high on the dimension hoarding exhibited the most distinct impairment. Moreover, deficits in procedural learning were more pronounced in OCD patients with an earlier age at illness onset. The results confirm previous findings and add supportive evidence for performance deficits in procedural learning and the fronto-striatal dysfunction model of OCD.