Task Decomposition Through Competition in a Modular Connectionist Architecture: The What and Where Vision Tasks

A novel modular connectionist architecture is presented in which the networks composing the architecture compete to learn the training patterns. An outcome of the competition is that different networks learn different training patterns and, thus, learn to compute different functions. The architecture performs task decomposition in the sense that it learns to partition a task into two or more functionally independent tasks and allocates distinct networks to learn each task. In addition, the architecture tends to allocate to each task the network whose topology is most appropriate to that task. The architecture's performance on “what” and “where” vision tasks is presented and compared with the performance of two multilayer networks. Finally, it is noted that function decomposition is an underconstrained problem, and, thus, different modular architectures may decompose a function in different ways. A desirable decomposition can be achieved if the architecture is suitably restricted in the types of functions that it can compute. Finding appropriate restrictions is possible through the application of domain knowledge. A strength of the modular architecture is that its structure is well suited for incorporating domain knowledge.     

Pain,emotion, and the situational specificity of catastrophising

Two studies were designed to explore the cross-situational nature of catastrophising and the emotions associated with pain and catastrophising. The crosssituational consistency of catastrophising in response to a finger-pressure procedure and during an episode of headache pain was examined in the first study. The second study examined differences between catastrophisers and noncatastrophisers with respect to state and trait measures of positive and negative emotions. Results of study one indicated that almost half of the subjects remained consistent in their classification as catastrophiser or noncatastrophiser in both pain situations. The majority of subjects that switched classification changed from being classified as catastrophisers during the headache experience to noncatastrophisers during the finger-pressure procedure. Results of the second study indicated that catastrophisers experienced significantly greater fear, sadness, anger, hostility, guilt, disgust, and shame during the finger-pressure procedure as compared to noncatastrophisers. Unexpectedly, catastrophisers were not a homogeneous group in regard to the pattern of negative emotions reported. Catastrophisers with headaches experienced greater sadness in response to finger-pressure pain than catastrophisers without headaches. Theoretical and clinical implications of these findings are discussed.     

Return of the Candy Witch: Individual differences in acceptance and stability of belief in a novel fantastical being

Recent research indicates that preschoolers make sophisticated choices in accepting testimony as a source of knowledge. Nonetheless, many children accept fantastical beings as real based on misleading testimony. The present study probes factors associated with belief in a novel fantastical figure, the Candy Witch, that 3‐ to 7‐year‐olds heard about at school. Short‐term belief was predicted by an interaction of age, existing beliefs in fantastical figures, and whether the child was ‘visited’ by the Candy Witch. Stability of belief was assessed over the course of 3 weeks and again 1‐year later. First year results revealed three patterns of belief: stable belief, wavering belief and stable non‐belief. First year belief status was not related to age, but older children from the stable belief group were more likely than younger children to disbelieve 1‐year later. The discussion presents a new proposal for the trajectory from belief to disbelief and an updated perspective on the role of individual differences in belief.     

Some bodies are easier to read: The effect of consistency and regularity on children's reading

Patterson and Morton (1985) proposed a model for the skilled reading of words and non-words that accommodates two non-lexical routines. One is the grapheme-to-phoneme correspondence system which utilizes the regularity of letter to sound correspondences for single letters and digraphs. The other is a system of “bodies”--the vowel and terminal letters of a monomorphemic, monosyllabic word. The idea of the body segment, as Patterson and Morton use it, is to capture consistency effects in reading aloud--that is, the fact that the spelling-sound pattern of words with similar written endings to the target affects the speed and accuracy of its reading.

In this study consistency and regularity are examined as separate factors in children's reading, by devising stimuli in accordance with the different types of three-letter ending that are proposed within the body sub-system.

A group of 87 children aged seven to nine (reading age range: 6;6 to 13;7) was sub-divided according to reading ability and given words and non-words to read aloud. In all the children, performance was affected by body type for both words and non-words, but the better readers were most affected. The implications of these results for a radical distributed model of reading acquisition (Seidenberg & McClelland, 1989) are considered.