Knowledge level learning and the power law: A soar model of skill acquisition in scheduling

Zusammenfassung. Das hier vorgestellte Modell lernt graduell, Planungsaufgaben aus der Klasse der Maschinenbelegungsprobleme (job-shop-scheduling problems) zu lösen. Mit Hilfe des Chunking-Mechanismus von Soar wird episodisches Wissen über die Belegungsreihenfolge von Aufträgen auf Maschinen memoriert. Bei der Entwicklung des Modells wurden zahlreiche qualitative (z. B. Transfereffekte) und quantitative Befunde (z. B. Bearbeitungszeiten) aus einer früheren empirischen Untersuchung berücksichtigt. In einer Validierungsstudie wurden dieselben Aufgaben von 14 Probanden und dem Modell bearbeitet. Die Passung von Simulationsdaten und empirischen Ergebnissen fiel insgesamt gut aus. Allerdings löst das Modell die Aufgaben schneller und zeigt auch einen etwas besseren Lernverlauf als die Probanden. Das Modell liefert eine Erklärung für das Rauschen, das typischerweise bei Bearbeitungszeiten zu beobachten ist: es handelt sich um erworbenes Wissen, das mehr oder weniger gut und auch unterschiedlich häufig auf neue Situationen übertragen wird. Der Lernverlauf der Probanden entspricht nur für aggregierte Daten einer Potenzfunktion (power law). Der vorgestellte Mechanismus zeigt, wie ein symbolisches Modell der Informationsverarbeitung graduelle Verhaltensänderungen generiert und wie der offensichtliche Erwerb allgemeiner Prozeduren ohne explizites Lernen von deklarativen Regeln erfolgen kann. Es wird nahegelegt, daß es sich hier um die Modellierung einer Form impliziten Lernens handelt. Summary. The model presented here gradually learns how to perform a job-shop scheduling task. It uses Soar's chunking mechanism to acquire episodic memories about the order to schedule jobs. The model was based on many qualitative (e.g., transfer effects) and quantitative (e.g., solution time) regularities found in previously collected data. The model was tested with new data where scheduling tasks were given to the model and to 14 subjects. The model generally fit these data with the restrictions that the model performs the task (in simulated time) faster than the subjects, and its performance improves somewhat more quickly than the subjects' performance. The model provides an explanation of the noise typically found in problem solving times - it is the result of learning actual pieces of knowledge that transfer more or less to new situations but rarely by an average amount. Only when the data are averaged (i.e., over subjects) does the smooth power law appear. This mechanism demonstrates how symbolic models can exhibit a gradual change in behavior and how the apparent acquisition of general procedures can be performed without resorting to explicit declarative rule generation. We suggest that this may represent a type of implicit learning.     

An automated version of the operation span task

We present an easy-to-administer and automated version of a popular working memory (WM) capacity task (operation span; Ospan) that is mouse driven, scores itself, and requires little intervention on the part of the experimenter. It is shown that this version of Ospan correlates well with other measures of WM capacity and has both good internal consistency (alpha=.78) and test-retest reliability (.83). In addition, the automated version of Ospan (Aospan) was shown to load on the same factor as two other WM measures. This WM capacity factor correlated with a factor composed of fluid abilities measures. The utility of the Aospan was further demonstrated by analyzing response times (RTs) that indicated that RT measures obtained in the task accounted for additional variance in predicting fluid abilities. Our results suggest that Aospan is a reliable and valid indicator of WM capacity that can be applied to a wide array of research domains.     

Simulation à la Goldman: pretend and collapse

Theories of mind draw on processes that represent mental states and their computational connections; simulation, in addition, draws on processes that replicate (Heal 1986) a sequence of mental states. Moreover, mental simulation can be triggered by input from imagination instead of real perceptions. To avoid confusion between mental states concerning reality and those created in simulation, imagined contents must be quarantined. Goldman bypasses this problem by giving pretend states a special role to play in simulation (Goldman 2006). We argue that this path leads to the resurgence of the threat of collapse (Davies 1994), diluting the principled distinction between simulation and theory use. Exploration of a related method of real-mental states operating in a pretend mode leads to a factually untenable model. Our main goal here is to raise this problem as a challenge for Goldman’s reconfigured simulation theory. Only at the end we will briefly sketch a possible alternative way of quarantine that preserves the replicative element of simulation and avoids collapse. Figure 1 provides a guide to our argument. Fig. 1 Structure of argument

---

Decision-making impairments in patients with Parkinson's disease

A high percentage of Parkinson's disease (PD) patients show cognitive impairments in addition to the cardinal motor symptoms. These deficits primarily concern executive functions most probably linked to dysfunctions in prefrontal regions due to decreased dopaminergic transmission in fronto-striatal loops. To investigate possible associations between decision-making and executive functions in PD, we examined 20 non-demented PD patients and 20 healthy control subjects with a neuropsychological test battery and the Game of Dice Task. In this computerised decision-making task, the rules for gains and losses and the winning probabilities are obvious and stable. Thus, strategic components besides feedback processing might influence decision-making in this task. We found that PD patients were impaired in the Game of Dice task performance and that the frequency of disadvantageous choices correlated with both executive functions and feedback processing. We suggest that decision-making deficits of PD patients in explicit gambling situations might be associated with dysfunctions in two different fronto-striatal loops: the limbic-orbitofrontal-striatal loop, involved in feedback processing, and the dorsolateral prefrontal-striatal loop, involved in executive functions.     

A Bizarre Property Equivalent to the -Fan Theorem

It is shown, with intuitionistic logic, that if every locallyconstant function from to has a property akinto constancy, then the fan theorem for -bars holds, and conversely.     

Adaptation in the constancy of visual direction measured by a one-trial method

Adaptation to field displacement during head movements in the direction with the head rotation and in the direction against it was produced under otherwise identical conditions and compared; the field displacement rate was also varied. A rapid training procedure was used, and a novel one-trial test was employed that could measure the adaptation well enough to compare the effects of various training conditions. The one-trial test measured the magnitude of one of the manifestations of adaptation, the apparent displacement of a stationary target during head movements. This apparent horizontal target displacement was transformed into an oblique one by having the head movements that brought forth the apparent target displacement simultaneously cause an objective vertical target displacement. The slant of the resultant apparent motion path varied with the magnitude of the apparent horizontal target displacement. It was measured by having S reproduce its slant angle. It was found that adaptation to field displacement in the direction with the head rotation was consistently greater than adaptation to the opposite displacement conditions. An explanation for this result is offered.     

Adaptation in distance perception based on oculomotor cues

Accommodation and convergence primarily serve to adjust the eyes to the distance of the object viewed, but, once made, these oculomotor adjustments serve as cues for the object’s distance. Experiments are reported that show that the relation between oculomotor adjustments and the distances they signify can be changed by adaptation to glasses that cause alteration in the oculomotor adjustments with which objects are viewed. This changed relation manifested itself in marked alterations of size perception. Wearing, for 30 min, glasses that caused a change in accommodation and convergence corresponding to a smaller object distance and equivalent to 1.5 lens diopters caused subsequent mean size increases that ranged from 50% to 65%. Adaptation to glasses that changed oculomotor adjustments in the same amount but in the opposite direction resulted in decreases in perceived sizes that varied from 18% to 40%, dependent on the distance of the test object. These were the results of size estimates obtained before and after the adaptation period under conditions where only accommodation and convergence served as cues for distance. A newly developed test of size perception was also used, in which S adjusted the size of the projected image of an array of familiar objects on a screen until the size of the objects appeared normal. Again, such adjustments were made before and after the adaptation period, and size differences were obtained that were in the direction to be expected of adaptation and varied in amount between 12% and 33%, dependent on the distance of the screen. The reason for the different amounts of size change measured by the two kinds of tests was investigated.