Placeholder Image

字幕列表 影片播放

  • NARRATOR: The name "Perkins" carved in stone.

  • Below a gothic tower a boy navigates with a cane.

  • A title: Visual Acuity Testing:

  • Acuity Cards and Testing Procedures

  • with D. Luisa Mayer, Ph.D.

  • - About in the '80s, people were really interested

  • in using various techniques, electrophysiology

  • but of course behavior, to measure visual functions

  • in babies that had visual problems.

  • And particularly to measure visual acuity,

  • which is the most important, probably,

  • most important visual function that you can measure

  • in an individual with a visual impairment.

  • And there were a number of different researchers

  • who were interested in this and in doing this,

  • modifying procedures and so forth.

  • And at that time, in the '80s,

  • I had finished my dissertation,

  • my OPL research, and came to Children's Hospital in Boston

  • to work with Anne Fulton, who was the first ophthalmologist

  • to bring FPL into a clinical setting.

  • And she did that with Velma Dobson's collaboration.

  • They brought an apparatus, which I help set up

  • at Children's Hospital in Boston in '79, I believe.

  • And they did some feasibility studies

  • and showed that you could actually use it

  • to test babies using shortcut procedures.

  • You couldn't do full psychometric functions

  • with babies in a clinic because you can't spend days

  • testing their vision.

  • But you could do it with shortcut procedures.

  • Teller was having troubles about our using FPL and OPL.

  • I brought OPL to Children's Hospital

  • and we're able to test most kids with that.

  • Not everybody, but a lot of kids.

  • But she felt there were problems with bringing FPL and OPL

  • into the clinic and using it to test babies.

  • And she felt that there really needed to be

  • a real clinical test of visual acuity

  • that could be applied in all kinds of clinical settings.

  • And she did something quite shocking to those of us

  • who worked with her because she went against

  • her psychophysics and against the scientific basis

  • of FPL and OPL.

  • And she did away with it.

  • And she said there's a lot of information that observer

  • is getting from the baby that doesn't get conveyed,

  • it doesn't get used in forced choice preferential looking.

  • For example, those very strong looks to big stripes

  • and the subtler looks to smaller stripes

  • and the subtle things that babies do,

  • because sometimes they raise their eyebrows

  • when they see stripes, and maybe that's a strong cue

  • that, yeah, they really saw and were interested in the stripes.

  • She felt that subtle, more subjective information

  • could be translated into a test where the observer

  • now controlled the testing.

  • And how did they do that?

  • She had the stripes mounted on gray cardboard

  • where the stripes were in one position

  • and that gray stimulus was still over there

  • and it was a gray surround and a gray card.

  • So the stimulus was the same, but the stripes

  • were on individual cards.

  • Every card had a different sized stripe.

  • And the observer was to use those cards

  • in whatever way they needed to, to find out if the baby

  • detected the stripes.

  • No longer was she saying the stripes were on

  • the right or the left.

  • NARRATOR: A photograph shows Davida Teller

  • with a number of rectangular cards fanned out

  • in front of her.

  • The width of the stripe stimulus varies from card to card.

  • - Well, we were really upset because we felt

  • that put too much control into the hands of the observer,

  • who could be biased.

  • We were observers, after all.

  • And as somebody looking at a little baby,

  • you want them to have 20/20 vision.

  • So you're going to say we felt that baby sees that card,

  • of course.

  • And Teller said, "Well, yes, there is the problem of bias,

  • "but the procedure can be designed so that

  • you minimize bias."

  • And how do you do that?

  • You keep the person showing the stripes ignorant or unaware

  • of where the stripes are until the person finishes testing.

  • Test this stripe multiple times showing it to the right

  • and the left, making sure that they think the baby sees it

  • because the baby looks strongly that way, flip the card,

  • the baby looks strongly this way, flip it again.

  • The baby looks strongly that way.

  • "I really think the stripes are there."

  • Check, "Oh, I was right."

  • And that kind of information the observer could

  • keep themselves objective to a certain point,

  • then confirm their judgment,

  • so give themselves feedback about it.

  • The corollary of that is that if you didn't think

  • the baby saw the stripes or you weren't sure,

  • you didn't check the front because you could

  • really be biased by that.

  • There are only two choices, after all.

  • So 50% of the time you could guess it correctly

  • and the baby's not seeing it.

  • So those two controls really, really answered

  • the question about controlling for observer bias,

  • at least to some degree.

  • The first iteration of this was to present these cards

  • in an apparatus very much like

  • the forced choice preferential looking apparatus,

  • but a portable one that could be put on a table.

  • And we called it a stage because it turns out

  • it looked sort of like a puppet stage.

  • And it had a rectangular window in the middle,

  • behind which the cards could be presented

  • and moved away and then flipped and presented again.

  • And so that made it very convenient

  • and made it pretty easy, easy to fit on a table

  • in some room.

  • You could close it up, it was portable,

  • you could put it away.

  • One really wonderful thing that happened was that window.

  • Because now that window opened the observer's view

  • and the baby's view of the observer

  • and also to a puppet show.

  • So often we would hold up a little puppet

  • to keep the baby's interest or sometimes our face

  • looking at them was enough to look through that window.

  • And then the card would go up and the baby

  • would still be looking,

  • sometimes waiting for you to show up again,

  • so you had to be a little careful

  • about how animated you were.

  • That's one of my problems is I would sit there

  • and play with the babies, but not do my test.

  • NARRATOR: We see a video clip of Luisa Mayer

  • administering an acuity card test to a young boy.

  • The boy sits on his mother's lap in front of the stage

  • as the cards are presented to him in the window.

  • When the cards are removed from the window and rotated,

  • the observer is able to engage the child's attention

  • before presenting the card again in a new orientation.

  • The child is observed through a peephole

  • in the middle of each acuity card.

  • - But that was a great, really clinically important thing

  • that happened just by chance, essentially.

  • It wasn't planned, so it was really, really nice.

  • So essentially the acuity card procedure--

  • which is what it was called-- wasn't very much different

  • in terms of the stimulus and the presentation mode

  • and the observer's task, in a sense, from FPL,

  • except that all of the psychophysics

  • and all of the counting and percent correct

  • and statistics that were done didn't need to be done.

  • In the '80s, Teller and Dobson and their colleagues

  • in Seattle did a number of studies of the ACP,

  • the acuity card procedure,

  • essentially to show how or to find out

  • how did it compare with FPL and OPL,

  • and it turns out the norms are very close.

  • They agree rather well.

  • And in a direct comparison between FPL

  • and the acuity card procedure in babies with visual problems,

  • it turned out they agreed remarkably well.

  • These were monocular tests,

  • you know, babies being tested one eye at a time.

  • They also discovered that test re-test reliability

  • was pretty good, which is very important.

  • You need to know you're going to get