David
A good post and it reflects my experience of the Holmes Wright test some 6 years ago. As usual the CAA does not seem to be using the tests in the manner that they were originally designed to be used, Holmes and Wright, whoever they were designed the lanterns but I am almost sure that the validity of the results achieved by a person would be based upon carrying out the test in a particular way. I understand that it is common in research that they would validify the findings and therefore the ability of the lantern to produce correct results using by prescribing the method in which the test is carried out. This is usually the method and 'rules' that have been used during research and hence those used to produce results demonstrating the validity of the test.
Anyway.......the JAR manual states the following in relation to the Holmes Wright lantern, and if what you say is correct the test certainly hasn't been carried out in accordance with this. An extract for the manual is as follows:
The Holmes-Wright lantern has an aperture size of 1.6 mm, corresponding to a visual angle of 0.9 minutes of arc. The light intensity is 2 000 m-candelas for demonstration, 200 m-candelas for daylight testing and 20 m_-candelas for testing in complete darkness. [The latter not used by City University or by the UK CAA.] The lantern is easy to use. The examinee is placed in front of the lantern at a distance of 6 metres.
Five different colours are presented: two red, two green, and one white light stimulus in nine different combinations, each presenting two colours (which may be identical in some of the presentations). The 2 x 9 fixed stimuli are presented for two to three seconds each and the examinee must identify the colour of each without delay. If all colours are correctly identified, the lantern test has been passed. If the examinee makes two or more errors, the lantern test has not been passed and the examinee is classified as colour unsafe. If the examinee makes one mistake or shows uncertainty during the test run, the lantern test is re-performed by executing two consecutive
runs of the nine presentations. No errors or mistakes are allowed during this second run.
I wonder what Mr Holmes and Mr Wright criteria were to claim the validity of the lantern? Bet you it is not the same as the CAA!!!!