MOTIVATION: The reconstruction of signaling pathways from gene knockdown data is a novel research field enabled by developments in RNAi screening technology. However, while RNA interference is a powerful technique to identify genes related to a phenotype of interest, their placement in the corresponding pathways remains a challenging problem. Difficulties are aggravated if not all pathway components can be observed after each knockdown, but readouts are only available for a small subset. We are then facing the problem of reconstructing a network from incomplete data. RESULTS: We infer pathway topologies from gene knockdown data using Bayesian networks with probabilistic Boolean threshold functions. To deal with the problem of underdetermined network parameters, we employ a Bayesian learning approach, in which we can integrate arbitrary prior information on the network under consideration. Missing observations are integrated out. We compute the exact likelihood function for smaller networks, and use an approximation to evaluate the likelihood for larger networks. The posterior distribution is evaluated using mode hopping Markov chain Monte Carlo. Distributions over topologies and parameters can then be used to design additional experiments. We evaluate our approach on a small artificial dataset, and present inference results on RNAi data from the Jak/Stat pathway in a human hepatoma cell line.

When trying to learn a complex task, people can use different strategies. They can use systematic exploration in which they take on an active approach to discover the computer functions and make use of problem solving steps such as planning, evaluation of feedback, and control of emotion and motivation. Alternatively, they can use non-systematic strategies like trial-and-error, rigid exploration, and encapsulation in information seeking. This study examined whether the exploration strategies were related to error consequences and performance when people learned a new computer program. Strategies were assessed by means of coding. Analysis showed strong correlations between strategies, error consequences, and task performance. These results can have implications for training design and human reliability in dealing with complex devices.

This article reports an empirical study on effectiveness of software development teams. It was predicted that both the extent to which design methods are used and team leaders' goal orientation is related to team effectiveness. Follow-up data (6-12 months after first assessment) were available from 25 German and Swiss software development projects. Hierarchical regression analysis showed that use of design methods and team leaders' goal orientation predicted team efficiency and changeability of the software product. In addition, goal orientation predicted maintaining schedule and budget. Practical implications of this study refer to the necessity to consider both the use of design methods and leadership issues when aiming at effective software development projects.

On the basis of a longitudinal field study of 29 commercial software development projects, the pros and cons of user centredness in software development were analysed. We looked at two concepts: user participation-an organizational device-involving a user representative in the team, and user orientation-a cognitive-emotional concept-which pertains to positive attitudes towards users. Both were found to be associated with project difficulties relating to process and product quality as well as overall project success. We suggest that the issue is no longer whether or not to involve users, but instead to develop a realistic understanding of the difficulties associated with user centredness.

A study of 30 psychology students replicated and extended the findings of Frese et al. (1991) on the superiority of error training, a procedure in which trainees are exposed to many errors. The hypothesis was that error training would lead to exploration and this would increase performance. The error training group performed better than the group that received training in which they were not allowed to make any errors. Students in both the error training and the error avoidant groups performed better if they explored. Exploration was done in contrast to the instruction in the error avoidant group. The data suggest that it is necessary to look into the question of whether errors instigate exploration.

An action‐oriented taxonomy of errors in human‐computer interaction in the office differentiated four classes: functionality problems, usability problems, interaction problems, and inefficient behavior. Functionality problems were differentiated in how they affect the action process. Usability problems were differentiated according to levels of action regulation and steps in the action process. For example, conscious strategies were differentiated from automatic ones. To examine the taxonomy's construct validation, several hypotheses regarding error‐handling time, need for external support, complexity at work, and novices versus experts were tested in a field study of 198 clerical workers at 11 German companies and 7 small firms. A total of 1,749 errors were observed within a 2‐hour period, 1,306 were rated concordantly by two re‐raters. As expected, errors resulting from conscious regulation and functionality problems needed more error‐handling time than errors resulting from more automatic actions. There were more thought and memory errors at workplaces with high complexity. The most external support was needed for knowledge errors. Novices committed more knowledge errors and experts more habit errors. Practical implications are discussed both for software development and training.

Traditionally, errors are avoided in training. In contrast to this approach, it is argued that errors can also have a positive function and that one has to learn to deal efficiently with errors on a strategic and an emotional level (error management). An experiment tested these assumptions. One group (n = 9) received guidance for error-free performance; another group (n = 15) received error training. In the latter group, errors were produced by assigning problems that were too difficult to deal with. The error-training group showed higher scores in the nonspeed performance tests. Error training seems to be positive for people with high scores on the cognitive failure questionnaire (Broadbent, Cooper, FitzGerald, & Parkes, 1982).

In a field study (29 engineers), the transfer from expertise acquired in training to software use at work was shown to be mediated by task application knowledge (i.e. knowledge used to connect skills learned in training with tasks at work). Moreover, person variables like setting long range goals and developing detailed plans and an organizational variable like job decision latitude (i.e. how much freedom do workers have to do their work) influenced the transfer process. People with high goal orientation and planfulness and with high job decision latitude showed a higher transfer.