CORTICAL PERIMETER

Short Info


● Cortical Perimeter is a software tool for the evaluation of contrast sensitivity of the visual field using ordinary computer displays or flat panel TV displays. Contrast sensitivity refers to the ability to see low contrasts, i.e. small differences in brightness.

● If a sufficiently large display is used, a full 30-deg-field can be measured. Alternatively, the measurement can be limited to a smaller area around the fovea, e.g. the macula area.

● In addition, it is also possible to evaluate the Letter Contrast Sensitivity Function (LCSF) of central vision.

● Further, by means of the Training mode, Cortical Perimeter can be used for training possible residual vision at defective visual field locations.



Visual field testing — key features

Fast perimetry mode. Four nearby measurement points on the visual field are combined to increase measurement speed. The increased speed is, therefore, achieved at the expense of reduced spatial accuracy.

Medium speed perimetry mode. Sensitivity of each measurement point is estimated independently.

Detection task. The observer is required to indicate whether she/he has seen a static or flickering ring by pressing the space bar on the keyboard. Observer's response criterion is controlled using catch trials.

Identification task. The observer is required to indicate the identity of a static or flickering letter by clicking a button marked by that letter. If the observer feels her/himself completely uncertain about the target identity, she/he should choose one of the alternatives on the basis of a guess of the most likely letter. If the observer categorically refuses to do that, the Next button can be pressed, which is always interpreted as a wrong response.

Stimulus properties. All stimuli have been filtered to a 1.5 octave spatial frequency pass-band having a centre frequency equal to 3 c/object width and height. In visual field testing, the stimulus presentation duration is 167 ms. When flicker is used, its rate is 12 Hz.

Letter stimuli are filtered Sloan or Sloan-like letters: A, B, D, H, K, N, O, R, S, U, V, and Z. The ring stimulus is the same as the filtered Sloan letter O.

The measurement can be done using different combinations of stimuli and tasks: Detection of static rings, Detection of flickering rings, Identification of static letters, or Identification of flickering letters.

M-scaling. The stimulus size is increased towards larger eccentricities (eccentricity = sideward angular distance from the fixation point) in order to obtain a roughly equal cortical representation as well as a roughly equal sensitivity across the visual field if the field is a healthy one.

Algorithm. The algorithm employed - a variant of the stochastic approximation method (Robbins and Monro, 1951) - searches for threshold contrasts at up to 73 points in the visual field in parallel. Target contrast changes depending on the responses given by the observer. The results, which can be stored to the hard disk, are displayed both graphically as a contrast sensitivity map and as alphanumerically.

Display. The program adapts to any display size. Large displays allow measurement of larger visual field areas. In the current program version, the maximum eccentricity is 30 deg at all meridians. A full 30-deg-field with 73 points altogether can be obtained with a 55-inch display (e.g. a flat panel TV screen), if the viewing distance is 40 cm. A 30-inch display is sufficient for measuring a full 20-deg-field. With the Fast perimetry and the detection task the 73-point-field can be measured in just under 4 minutes.

Sound signals. The program uses sound signals to indicate the start and end of the measurement and the start of the presentation of each stimulus.

 

Visual Field Examples

Visual field of a subject having a cortical V1 lesion (A) and a healthy visual field (B).

 

Letter contrast sensitivity function (LCSF)

Letter contrast sensitivity function for various spatial frequencies can be measured using the same band-pass filtered letters as used in the visual field measurement. Different spatial frequencies are achieved using letters of different sizes. Spatial frequency increases with decreasing letter size.

In the LCSF measurement each stimulus is shown for 500 ms. When the measurement is finished, the program automatically draws the result as a graph and gives the results alphanumerically, too. The graphs and alphanumerical results can be stored on the hard disk.

The start of a stimulus is indicated by a mid tone sound. A correct response is followed by a high tone sound and a wrong response is followed by a low tone sound.

An example of a Letter Contrast Sensitivity Function (LCSF).

 

Training mode

Some scientific studies suggest that the function of peripheral vision at defective areas after cortical lesion can be improved by intensive training at least in some cases (see e.g. Raninen et al. Temporal sensitivity in hemianopic visual field can be improved by long-term training using flicker stimulation. JNNP, 2006. http://dx.doi.org/10.1136/jnnp.2006.099366, and
Henriksson et al. Training-induced cortical representation of a hemianopic hemifield. JNNP, 2006. http://dx.doi.org/10.1136/jnnp.2006.099374).

The Training mode uses methods similar to those that have been found successful in the above-mentioned studies. Vision is trained in defective visual field locations by measuring contrast sensitivity repeatedly for a number of times during weeks or months. It should be noted that the effectiveness of the method has not been generally recognized and further research is needed on this subject.

In the detection task of the training mode, the observer hears two sound signals. The stimulus is shown after one of them at random. The observer is required to indicate whether the stimulus was presented after the first or second sound by pressing either key '1' or '2' on the keyboard or, alternatively, graphical button '1' or '2' on the display.

 

Calibration of the display

In order to get accurate results it is important to calibrate the display. The program needs to know the pixel size and the approximate shape of luminance response function of the display, which is characterized by a single number called gamma.

A very quick and easy way to do the calibration is the Visual calibration [Menu: Calibration -> Visual Calibration]. The gamma value obtained in this way is reasonably close to the true value.

Another, but somewhat more laborious, possibility is to use a luminance meter (photometer) to measure the objective luminance values for several display brightness values. Access this option by choosing Calibration->Photometer Calibration Aid. If you choose this, you will have to calculate the gamma value based on the measured luminance values. The gamma and pixel size values are then given to the program by using the Save Calibration menu-item in the Calibration menu, which saves the calibration to the hard disk.

If no calibration is done, the program assumes that the gamma is equal to 2.2 and the pixel size is equal to 0.0275 cm.