Human Color Vision by Jan Kremers Rigmor C. Baraas & N. Justin Marshall

Author:Jan Kremers, Rigmor C. Baraas & N. Justin Marshall
Language: eng
Format: epub
Publisher: Springer International Publishing, Cham

6.4 The Role of Individual Differences in Theory Development

More than 50 years ago, Cronbach [161] pointed to a curious and unfortunate theoretical schism dividing psychology into two largely separate schools of thought he referred to as “correlational psychology” and “experimental psychology.” The main difference between them is their perspective on the importance of individual differences. It is probably fair to say that research on perception has been, and to some extent still is, deeply entrenched in the one-sided perspective of the “experimental psychology” described by Cronbach and has yet to fully exploit the benefits of complementing traditional experimental studies with analyses of natural variation [162–164].

This is particularly tangible in the literature on color constancy and color appearance. Many studies have been performed with a small number of observers, quite often only the authors and perhaps a couple of naive observers. This common practice is probably more strongly rooted in tradition [165] along with practical issues3 than based on a principled scientific deliberations. One could argue that the fair inter-observer agreement evident in many published studies suggests that individual differences are small and thus insignificant. The idea that inter-observer agreement tends to be good may, however, be a self-perpetuating prejudice caused and maintained by a publication bias. Whenever large individual differences are observed, many perceptual psychologists are probably prone to think that something went wrong in the experiment and hesitate to publish the results, particularly if data suggesting a higher precision have already been published. In the early days of perception science, large inter-individual variability was considered “prima facie evidence that the attempted isolation of critical determining factors had failed and that uncontrolled disturbing processes had supervened” ([165], p. 73) and even today, this kind of attitude may be encountered ([166], p.101).

In the literature on color constancy and color induction, there is often good agreement among the few observers participating in single studies, but vastly different results across studies [54]. More recent research documents surprisingly large individual differences within studies [96, 97, 167–169], which suggests that publication bias may indeed have been a real issue. As a notable exception in color constancy research, Allen et al. [167, 168] used individual differences in working memory performance to successfully account for individual differences in color constancy, specifically in how scene complexity affects color constancy performance (see also [170]).

The tendency to neglect individual differences in studies of color constancy and color induction is not entirely irrational. First, it is well known that the phenomena under study depend crucially on a host of stimulus variables [53, 54]. Thus, even small differences in the stimuli between studies may explain why they produce rather diverging results. Second, it is also clear that different methods for measuring how color appearance depends on the stimulus can lead to dramatically different results [84]. Thus, any differences observed across different studies can plausibly be attributed to the effect of known or unknown differences in the experimental variables.

It is also well known that the results of asymmetric color matching experiments may depend on instructions ([35, 171–173]).

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.