Comparison of contrast sensitivity tests


The aim of the experiment was to determine if comparable contrast sensitivity measurements could be taken using different psychophysical methods.

Contrast Sensitivity is defined as "the ability of the eye to detect luminance contrast". (Millodot, M, 2009) It is important as it makes objects stand out from one another. Contrast Sensitivity testing provides a better understanding of the patients performance of real-world tasks than visual acuity measurements. (Lay et al, 2009) It is also a more reliable way of detecting visual pathway changes than VA. VA detects changes caused by spherical blur, however not all abnormalities are caused by this, such as, cataracts, glaucoma or diabetic retinopathy ( 2010). On the other hand, Contrast Sensitivity tests are sensitive enough to identify subtle changes in patients. (Lay et al, 2009) Poor contrast sensitivity may cause difficulties while using stairs or stepping off curbs as patients may find it difficult to see the edges due to them being of the same contrast. Night time driving may also be difficult as they would have difficulty seeing pedestrians in poorly lit areas and may have difficulty with traffic lights. (Stephenson, M, 2010)

We tested the Contrast Sensitivity Function which is defined as "the graphical representation of contrast sensitivity as a function of spatial frequency". Spatial Frequency is defined as "rate of alteration of the luminance in a visual stimulus as a function of length, expressed in cycles per degree". (Millodot, M, 2009) There are a number of ways to measure threshold; Method of Ascending Limits, Method of Descending Limits, Method of constant stimuli, Method of adjustment and Forced choice method.

In the Method of Ascending Limits a stimulus is presented to the observer at an intensity lower than their threshold and this intensity is increased until the observer can see it. The Method of Descending Limits is the opposite. The stimulus is presented at a level above threshold until the observer can no longer see it. The staircase method uses both these principles. The intensity of the stimulus is increased until the observer can see it and recorded. It is then decreased until the observer cannot see it. An average of these reversals is taken as the threshold.

The subject changes the stimulus intensity until it is just visible in the Method of Adjustment. This allows for a quick and efficient acquisition of contrast threshold. (Kalloniatis et al, 2010)

The Forced Choice Method gives the subject a limited number of choices, one of which is the stimulus. The greater the number of choices the less likely the subject will be able to guess the correct answer and will make the result more reliable. In a two alternative forced choice, the subject already has a 50% chance of getting it right and the threshold is therefore considered 75% as seen in the diagram below. (Scwartz, S, 1999) (Kalloniatis et al, 2010)

The tests we used for our experiment were the Pelli-Robson chart, the Freiburg Landolt-C Contrast Sensitivity Test and The Gabori Contrast Sensitivity Test. The Pelli-Robson Chart is a 26-alternative forced-choice method, as any of the 26 letters of the alphabet can be used. (Lay et al, 2009).The Freiburg Landolt C Contrast Sensitivity Test is an adaptable staircase, 8- alternative forced- choice method due to the 8 orientations of the gap. The Gabori uses the method of ascending limits. The stimulus is presented to the observer at an intensity below their threshold and the intensity increases until the observer can see it. We set out to determine whether these tests gave similar results despite using different psychophysical methods.


The tests were carried out on five female patients aged 20, with refractive errors ranging from -2.75D to plano. A comprehensive medical history of the patients showed that they were all in good ocular health, with no systemic conditions and were not taking any medications which are known to affect the eye. There was no history of ocular pathology or surgery which is known to affect contrast sensitivity. The tests were performed monocularly in both the subjects' right and left eye. Each of the subjects right and left eyes were tested once for each of the three tests. Refractive correction was worn when required.

Pelli-Robson Chart is a 26-alternative forced-choice method, as any of the 26 letters of the alphabet can be used. It is 86x63cm and consists of 8 lines of letters, each letter of the same size. Each line consists of two subsets of three letters, with each subset of decreasing contrast sensitivity. Contrast Sensitivity ranges from high to low, down the chart. Log Contrast sensitivity scores range from 0.00 to 2.25. The subject stood at a distance of 1m from the chart with her left eye occluded. The subjects score is allocated based on the faintest triplet where 2 of 3 letters are correctly named.

The Freiburg Landolt-C Contrast Sensitivity Test uses Landolt-C optotypes to calculate the contrast threshold of the subject. There are 8 gap directions and the subject has to detect the correct one, recording it using the corresponding number on the keyboard as seen in the diagram. The result is displayed onscreen. The screen size was 49cm2.

The test makes 24 presentations of each direction. The Contrast C diameter is 60 mins of arc. The subject is at a viewing distance of 3m and had one eye occluded. The test was done once for both eyes.

The Gabori Test is used to test for contrast sensitivity function. It has a screen size of 49cm2. The subject sits 57cm from the screen with their left eye occluded. The subject is required to look directly at the centre and to react as quickly as possible to gratings that slowly come into view from the grey background by clicking on them with the computer mouse. The result is then plotted on a graph at a number of spatial frequencies with their corresponding contrast sensitivity as shown in the appendices.

The spatial frequencies were given for Gabori on the graph but had to be calculated for the Pelli-Robson and Freiburg Tests by measuring the width of the letters on the Pelli-Robson Chart and the size of the gap of the Landolt C in the Freiburg.

The subjects pupil sizes were also recorded for each test using a pd rule and luminance levels were measured using the Spectra Scan PR650. These are shown in the Appendices.


I recorded the results for each individual's right and left eyes.

The contrast sensitivity for the Pelli-Robson Chart was recorded in Log Contrast Sensitivity. To convert it to contrast sensitivity I had to get the anti-log of the score.

For example, Subject A's result for her right eye was 1.5 Log Contrast Sensitivity. To get the contrast sensitivity we got the antilog of 1.5, which gives us a contrast sensitivity of 31.6227766.

The contrast sensitivity for the Freiburg Landolt-C Contrast Sensitivity Test was expressed as threshold contrast in %. To calculate the contrast sensitivity we divided 100 by the score.

For example, subject A's result for her right eye was 0.58%. To get the contrast sensitivity we divided 100 by 0.58%, giving us a contrast sensitivity of 172.41.

The Gabori Test gave us a number of contrast sensitivity results for the subjects' right eye, but at different spatial frequencies. See Appendices for a table of results for the Pelli-Robson chart and the Freiburg Test and a graph for the Gabori Test.

Once we had converted all the results so that they were in the same units, we were able to calculate the mean contrast sensitivity of all five subjects for each of the three tests and compile a graph as seen below.

For the Pelli-Robson chart, a is the width of the letters, while for the Freiburg test a is the size of the gap in the Landolt C. b is the testing distance. Spatial frequency is equal to the inverse of the spatial cycle, T (deg). Spatial cycle, T, equates to two times the size, S. Thus Spatial Frequency,F, is equal to half the Size, S. These calculations can be seen in the Appendices.


It was found that the Pelli-Robson Tests and the Gabori Test gave similar results while the Freiburg Landolt-C Test gave a significantly different result. It gave a significant overestimation of the contrast sensitivity when compared to the other two tests. Also when taking into account the uncertainty of the position of the stimulus presented and thus the delay in response time, we were expecting an underestimation of the subjects contrast sensitivity. However, this was not the case. At a similar spatial frequency to the Pelli-Robson Chart it gave similar contrast sensitivity.

This caused us to go have a look at the other measurements we had taken, in particular the luminance and pupil size of each subject for each test.

It was found that the luminance of the Pelli-Robson test was greatest and the Gabori was least and therefore pupil size was smallest with Pelli-Robson and biggest with Gabori, which is what is to be expected. This can thus suggest a reason for our findings. Alfonso et al (2007) did a study on the "Correlation of pupil size with visual acuity and contrast sensitivity after implantation of an apodized diffractive intraocular lens" and found that contrast sensitivity was better with larger pupils than smaller pupils. This would suggest that contrast sensitivity would be greatest with the Gabori test and explains why it had a similar result to the Pelli-Robson.

The Gabori test may also have had a higher result than expected due to the error of habitation. The subject may have developed a habit of responding to the stimulus and respond to a stimulus above their threshold.

The Freiburg may have had a significantly larger result due to the luminance being lower than the luminance for the Pelli-Robson Chart. To determine whether the Freiburg test can be compared to the other tests it would need to be repeated at a luminance level similar to the Pelli-Robson.

Also each test was only carried out once for each subject and would need to be carried out a number of times so an average may be taken to get a more accurate result.


It was found that the Freiburg Landolt-C Contrast Sensitivity Test gave a significantly different result than the Pelli-Robson Chart and the Gabori Test. However it would not be justifiable to say that it is therefore not comparable to the Pelli-Robson and Gabori as all three tests were carried out under different luminances. As previous studies have shown there is a correlation between luminance and contrast sensitivity and this experiment would need to be carried out again with the same luminance for each test.

  1. Millodot, M. 2009. Dictionary of Optometry and Visual Science. 7th ed. China. Butterworth Heinemann Elsevier.
  2. Lay, M. Wickware, E. & Rosenfield, M. 2009. Visual Acuity and Contrast Sensitivity. In M. Rosenfield, ed & N. Logan, ed. 2009. Optometry: Science, Techniques and Clinical Management. China. Butterworth Heinemann Elsevier. Ch12.
  3. Schwartz, S.H. 2004.Visual Perception; a clinical orientation. 3rd ed. U.S.A. Appleton & Lange.
  4. VectorVision 1850 Livingston Rd, Ste E, Greenville, OH 45331, Contrast Sensitivity Background, Available at [Accessed 13 March 2010]
  5. Stephenson, M. Contrast Sensitivity Testing in Eye Exams. [Online] (Updated March 2010) Available at [Accessed 13 March 2010]
  6. Kalloniatis, M & Luu, C. Principles of Pyschophysics.[Online] (Updated March 2010) Available at [Accessed 23 March 2010]
  7. Alfonso, J.F. Fernndez-Vega, L. Baamonde, M.B. & Monts-Mic, R. 2007. Correlation of Pupil Size with Visual Acuity and Contrast Sensitivity after Implantation of an Apdized Difference Intraocular Lens. Journal of Cataract & Refractive Surgery. 33 (3) pp430-438.

Please be aware that the free essay that you were just reading was not written by us. This essay, and all of the others available to view on the website, were provided to us by students in exchange for services that we offer. This relationship helps our students to get an even better deal while also contributing to the biggest free essay resource in the UK!