Study conclusions should reflect results

Weak Results, Misleading Conclusions

The authors fail to control for relevant "third" variables that could easily explain the weak correlations that they find. Family violence exposure for instance, peer group influences, certainly genetic influences on aggressive behavior are just a few relevant variables that ought either be controlled or at minimum acknowledged as alternate causal agents for (very small) link between video game...

Physician pay should reflect workloads, say economists.

Should Neural Network Architecture Reflect Linguistic Structure?

I explore the hypothesis that conventional neural network models (e.g., recurrent neural networks) are incorrectly biased for making linguistically sensible generalizations when learning, and that a better class of models is based on architectures that reflect hierarchical structures for which considerable behavioral evidence exists. I focus on the problem of modeling and representing the meani...

What’s in a name? – Diatom classification should reflect systematic relationships

Large numbers of diatom taxa are currently being described each year and molecular data sets are providing phylogenetic evidence that challenges the traditional systematic arrangement of diatoms, but is such information being integrated into the classification? The traditional diatom classification originated as an aid to identification rather than as an arrangement expressing perceived relatio...

Protocol Organs: Modularity should reflect function, not timing

This paper proposes a functional architecture for protocol systems, as opposed to the conventional temporally layered one. Each block of information to be communicated is operated on by several organs, e.g. for routing, integrity and flow control. The architecture retains a skeleton protocol stack to determine the sequence of these operations. However, unlike conventional stacks, the operators ...