Revisiting confidence intervals for repeated measures designs

Loftus and Masson (1994) proposed a method for computing confidence intervals (CIs) in repeated measures (RM) designs and later proposed that RM CIs for factorial designs should be based on number of observations rather than number of participants (Masson & Loftus, 2003). However, determining the correct number of observations for a particular effect can be complicated, given that its value depends on the relation between the effect and the overall design. To address this, we recently defined a general number-of-observations principle, explained why it obtains, and provided step-by-step instructions for constructing CIs for various effect types (Jarmasz & Hollands, 2009). In this note, we provide a brief summary of our approach.     

On two types of conditional reflex: General laws of association

Both types of conditioning are based on the general laws of associations-connections between the centers involved. Whereas the experimental procedures of classical conditioning expose mainly the conditioned stimulus (CS)-unconditioned stimulus (US) connection, those of instrumental conditioning expose the conditioned stimulus (CS)-response (R) connection. Thus, the main differences between the two types of conditioning are those associated with the different centers involved in each, not the associative-connective laws themselves.     

Data compression in machine learning applied to natural language

In this paper, we investigate the possibility of applying machine learning methods to data derived from the area of natural language and show how rules, induced by machine learning, are changed after the original data are compressed by grouping together entries, attributes, and attribute values. Also shown is how excessive compression of input data may affect the accuracy of induced rules.