Abstract Reference: Investigative Ophthalmology and Visual Science, Vol. 39, No. 4, p. S1109, 1998.

Purpose. Orientation selective and local sign mechanisms have been proposed as the basis for vernier acuity judgments (Waugh & Levi, 1993, Vision. Res.; Beard et al., 1997, Vision. Res.). Linear image features contributing to discrimination can be determined for a two choice task by adding external noise to the images and then averaging the noises separately for the four types of stimulus/response trials (Ahumada, 1996 ECVP). This method is applied to a vernier acuity task with different spatial separations to compare the predictions of the two theories.

Methods. Three well-practiced observers were presented around 5000 trials of a vernier stimulus consisting of two dark horizontal lines (5 min by 0.9 min) within additive low-contrast white noise. Two spatial separations were tested, abutting and a 10 min horizontal separation. The task was to determine whether the target lines were aligned or vertically offset. The noises were averaged separately for the four stimulus/response trial types (e.g., stimulus=offset, response=aligned). The sum of the two "not aligned" images was then subtracted from the sum of the "aligned" images to obtain an overall image. Spatially smoothed images were quantized according to expected variability in the smoothed images to allow estimation of the statistical significance of image features.

Results. The response images from the 10 min separation condition are consistent with the local sign theory, having the appearance of two linear operators measuring vertical position with opposite sign. The images from the abutting stimulus have the same appearance with the two operators closer together. The image predicted by an oriented filter model is similar, but has its greatest weight in the abutting region, while the response images fall to non-significance there.

Conclusions. The response correlation image method, previously demonstrated for letter discrimination, clarifies the features used in vernier acuity.

Supported by NASA/AOS RTOP 548-50-12.

The main point of the talk is that by adding noise to the stimuli in a discrimination task it is possible to reveal the stimulus features used by the observer to perform the discrimination. Rather than making single changes to the stimulus to see whether the observers responses are consistent with hypothetical features being used (and risk the observer changing the features) we add noise which changes the stimulus in all directions to reveal the features (Watson, 1998 ARVO).

The talk has two parts. The first part illustrates the method in the case that the discrimination task is vernier acuity. The second part illustrates some statistical analyses that allow an assessment of the fraction of response variance that is controlled by the added noise and the extent to which that variance is predictable from a given model.

The discrimination task is a vernier task in which the observer has to discriminate the right line being vertically offset from the case in which there is no offset. We look at two cases, one in which the lines are abutting and the other in which there is a 10 arc min separation between them.

There were three observers, a male student assistant PW, a female student assistant DF, and the second author BLB. The Vernier stimulus consisted of two dark (-40% contrast) horizontal lines (5.0 min by 0.93 min, 16 pixels by 3 pixels). The two horizontal separations were 0 arc min and 10.2 arc min. The vertical offset was either 0 or 0.31 arc min. The additive white noise had a uniform amplitude distribution and it filled a square of 39.7 by 39.7 sec (128 by 128 pixels).

On each trial the observer indicated by a key press whether the stimulus was "aligned" or "offset". Trials were run in blocks of 100.

Initially, the noise had a peak contrast of 0.25 and the stimulus duration was 0.5 sec. To increase the number of errors, after 1900 trials the signal duration was lowered to 0.3 sec for observer DF, and after 1000 more trials her noise peak contrast was increased from 0.25 to 0.35 (her last 2000 trials had the 0.3 sec duration and the 0.35 noise contrast). For observer BLB the noise contrast was increased to 0.4 after 1800 trials. For observer PW, the initial conditions were maintained throughout.