Abstract
Goals: Visual performance varies at isoeccentric locations. Discriminability is higher along the horizontal than the vertical meridian, and along the lower than the upper vertical meridian (e.g., Carrasco, Talgar & Cameron, 2001). This pattern of asymmetries is referred to as a "performance field." Performance is often limited by internal noise, which is considered to be independent of task, reflecting neuronal properties such as their density, gain, and physiological thresholds (Pelli & Farell, 1999). Internal noise arises from pre-cortical and cortical sources, which are reflected in monocular and binocular noise measurements, respectively (Raghavan, 1995). Here, we measure internal noise as a function of isoeccentric position, monocularly and binocularly, allowing us to identify the source of the noise that shapes performance fields. Methods: Observers performed a discrimination task at four locations (North, South, East & West; 5.5º eccentricity), with target contrast controlled by an adaptive staircase. All stimuli were presented stereoscopically with noise that was uncorrelated across both eyes. We measured thresholds monocularly and binocularly, in high (27%) and low (0%) contrast noise. By assessing threshold elevation as a function of external noise, we estimated the internal noise for each location and viewing condition. Results: We observed the classic shape of performance fields: better performance along the horizontal than the vertical meridian, and along the lower than the upper vertical meridian, under both monocular and binocular conditions. At each location, the monocular and binocular levels of internal noise were similar; binocular integration added no net noise. Thus, the noise that shapes visual performance fields is monocular and, therefore, pre-cortical. That noise may reflect differences in cell density at the level of the retinal ganglion cells and the LGN (e.g., Connolly & Van Essen, 1984; Curcio & Allen, 1990). Our findings indicate that performance fields result from early signal processing stages.
Meeting abstract presented at VSS 2012