On Tue, 01 Sep 2009 09:34:13 -0700, Michael Smith wrote: >Mike Palij wrote another extended response. Man. Are you retired? lol.
No, that's Stephen Black's job. I'm teaching 3 courses this semester including a lab course. I'm just thoughtful, analytical, and verbose. >My main point was that scientists no less than anyone else have >biases. This is not a new point but your handling of it is "inelegant". >And, like everyone else, tend to select only the references >(which presumably contain evidence) that supports their biases. Of >course, this is not what ideally happens in science but real life >isn't that ideal. This is a broad brush which ignores that many researchers are in fact self-critical and do acknowledge other viewpoints. It should be noted that in setting up a particular research project, the hypotheses being tested will usually be constructed in the context of a particular theory and may have less relevance in other theoretical contexts. Consider the following unresolved theoretical positions: (1) Are mental representations analog (i.e., in the form of images or discrete pieces that need to be stored and retrieved) or are they digital (basically, in abstract form such as the weights in a neural network)? A researcher who believes that mental representations are analog will ask different questions that a researcher who assumes that all mental representations are digital. For example, if one is concerned with whether 2-D and 3-D visual mental images have different psychological properties will probably find much more relevant research among other analog researchers (e.g., Roger Shepard, Stephen Kosslyn) than among digital/abstract researchers (e.g., Zenon Pylyshyn). In either cases, because the assumptions and predictions are different, using references in a "compatible" research area is more likely to occur than using references in an incompatible research area even though one is familiar with it. This might seem like a "confirmation bias" but it is not. (2) What is the nature of the psychophysical law? Is the relationship between level of stimulus energy and our subjective magnitude of it a logarithmic function (i.e., Fechner's Law), a power law (i.e., Stevens' Law), or some much more complex function because we need to describe the function relating stimulus energy to subjective magnitude and a function relating subjective magnitude to response magnitude (i.e., Shepard's position)? One is likely to operate within one of these "traditions" and rely upon work in one's area but they are also likely to know the other positions (if for no other reason then having to handle reviewers comments about other perspectives). [snip] >Anyway, I am assuming that Michael Sylvester meant something >along these lines when bemoaning perhaps this kind of use of >"the reference". It's never a good idea to assume that you know what Dr. Sylvester is thinking (especially if you a a eurocentric type) or why he is doing so. An alternative explanation for "the reference" is that he can't be bothered to search the relevant literatures and databases because that would take time away from more important activities such as DJ'ing and hanging out with biker gangs. :-) >Of course, I agree, that if you are going to do science then you need >to disseminate and communicate results hence the necessity of >references. I don't think anyone would argue that references aren't >the butter of the 'bread and butter' of the scientific community. It >is just the rather uncritical and routine reliance on them that is >being contested (and suggested as being what probably happens a great >deal more than anyone cares to admit). Methinks thou oversimplies things but I don't have the time right now to explain it to you. I'll leave that to Chris Green. ;-) >Wow. I must be getting close to retirement too! In more ways than one. ;-) -Mike Palij New York university m...@nyu.edu --- To make changes to your subscription contact: Bill Southerly (bsouthe...@frostburg.edu)
