L. Helle, M. Nivala, P. Kronqvist, Andreas Gegenfurtner, P. Björk, R. Säljö
Traditional microscopy instruction versus process-oriented virtual microscopy instruction: A naturalistic experiment with control group
Diagnostic Pathology, vol. 6, no. S1
Virtual microscopy is being introduced in medical education as an approach for learning how to interpret information in microscopic specimens. It is, however, far from evident how to incorporate its use into existing teaching practice. The aim of the study was to explore the consequences of introducing virtual microscopy tasks into an undergraduate pathology course in an attempt to render the instruction more process-oriented. The research questions were: 1) How is virtual microscopy perceived by students?) Does work on virtual microscopy tasks contribute to improvement in performance in microscopic pathology in comparison with attending assistant-led demonstrations only?
During a one-week period, an experimental group completed three sets of virtual microscopy homework assignments in addition to attending demonstrations. A control group attended the demonstrations only. Performance in microscopic pathology was measured by a pre-test and a post-test. Student perceptions of regular instruction and virtual microscopy were collected one month later by administering the Inventory of Intrinsic Motivation and open-ended questions.
The students voiced an appreciation for virtual microscopy for the purposes of the course and for self-study. As for learning gains, the results indicated that learning was speeded up in a subgroup of students consisting of conscientious high achievers.
The enriched instruction model may be suited as such for elective courses following the basic course. However, the instructional model needs further development to be suited for basic courses.
Motivation and transfer in professional training: A meta-analysis of the moderating effects of knowledge type, instruction, and assessment conditions
Educational Research Review, vol. 6, no. 3, pp. 153-168
This meta-analysis (148 studies, k = 197, N = 31,718) examined the relationship between motivation and transfer in professional training. For this purpose, motivation was conceptualized in the following nine dimensions: motivation to learn, motivation to transfer, pre- and post-training self-efficacy, mastery orientation, performance orientation, avoidance orientation, expectancy, and instrumentality. Population correlation estimates ranged between −0.11 and 0.52. Three moderator effects were estimated. First, correlations were higher when the training focused on declarative and self-regulatory, rather than on procedural, knowledge. Second, learner-centered environments tended to show greater numbers of positive correlations than did knowledge-centered environments. Third, when compared with external, supervisory, or peer assessment, self-assessment of transfer produced upwardly biased population estimates irrespective of the transfer criterion. These findings are discussed in terms of their implications for theories of training effectiveness and their significance for the practice of training evaluation.
Andreas Gegenfurtner, E. Lehtinen, R. Säljö
Expertise differences in the comprehension of visualizations: A meta-analysis of eye-tracking research in professional domains
Educational Psychology Review, vol. 23, no. 4, pp. 523-552
This meta-analysis integrates 296 effect sizes reported in eye-tracking research on expertise differences in the comprehension of visualizations. Three theories were evaluated: Ericsson and Kintsch’s (Psychol Rev 102:211–245, 1995) theory of long-term working memory, Haider and Frensch’s (J Exp Psychol Learn Mem Cognit 25:172–190, 1999) information-reduction hypothesis, and the holistic model of image perception of Kundel et al. (Radiology 242:396–402, 2007). Eye movement and performance data were cumulated from 819 experts, 187 intermediates, and 893 novices. In support of the evaluated theories, experts, when compared with non-experts, had shorter fixation durations, more fixations on task-relevant areas, and fewer fixations on task-redundant areas; experts also had longer saccades and shorter times to first fixate relevant information, owing to superiority in parafoveal processing and selective attention allocation. Eye movements, reaction time, and performance accuracy were moderated by characteristics of visualization (dynamics, realism, dimensionality, modality, and text annotation), task (complexity, time-on-task, and task control), and domain (sports, medicine, transportation, other). These findings are discussed in terms of their implications for theories of visual expertise in professional domains and their significance for the design of learning environments.
Robert Hable, A. Christmann
On Qualitative Robustness of Support Vector Machines
Journal of Multivariate Analysis, vol. 102, no. 6, pp. 993-1007
Support vector machines (SVMs) have attracted much attention in theoretical and in applied statistics. The main topics of recent interest are consistency, learning rates and robustness. We address the open problem whether SVMs are qualitatively robust. Our results show that SVMs are qualitatively robust for any fixed regularization parameter λλ. However, under extremely mild conditions on the SVM, it turns out that SVMs are not qualitatively robust any more for any null sequence λnλn, which are the classical sequences needed to obtain universal consistency. This lack of qualitative robustness is of a rather theoretical nature because we show that, in any case, SVMs fulfill a finite sample qualitative robustness property.
For a fixed regularization parameter, SVMs can be represented by a functional on the set of all probability measures. Qualitative robustness is proven by showing that this functional is continuous with respect to the topology generated by weak convergence of probability measures. Combined with the existence and uniqueness of SVMs, our results show that SVMs are the solutions of a well-posed mathematical problem in Hadamard’s sense.
Elektrotechnik und Medientechnik
László Juhász, J. Maas, B. Borovac
Parameter identifications and hysteresis compensation of embedded piezoelectric stack actuators
Mechatronics, vol. 21, no. Februar (1), pp. 329-338
A novel method for the identification of embedded piezoelectric stack actuator parameters in combination with a real-time capable hysteresis compensation measure is presented. The presented algorithms are based on the Maxwell resistive capacitor model and are particularly useful for the identification of piezoelectric actuators embedded in a high-precision micropositioning system where the disassembly of the complete system for separate actuator identification is not recommended or not possible. The parameter identification can be performed in a fully automated way and enables the adaptation of the compensation routine to the changed circumstances (temperature difference, wearing of actuators) as well. The hysteresis compensation method proposed here does not require significant CPU or memory resources. It can be implemented as an additional task on the already existing controller or a low-budget FPGA. As an example, the proposed method was validated experimentally by the parameter identification and hysteresis compensation of the piezoelectric actuators incorporated in a commercially available hybrid micropositioning system. The achieved experimental results are in very good agreement with the theoretical ones.