Model type and stimulus dimensions

The following fundamental parameters appear near the top of the model parameter file.

  mod_type   me_v5        # Motion energy V5 model

  sscale         0.1      # Spatial resolution of stimulus (degr/pix)
  tscale         0.002    # Temporal resolution of stimulus (sec/frame)
  xn            32        # Stimulus image width (pix)
  yn            32        # Stimulus image height (pix)
  tn           128        # Stimulus time duration (frames)

(To view the full parameter file in HTML markup, click here. To download the actual parameter file, click here.)

The first parameter, mod_type, in the parameter block above sets the overall model type to be 'me_v5'. The next four parameters specify the stimulus size and resolution, and thus determine the size of the patch of visual field that the model will process. Based on the example values above, the stimulus will consist of image frames that are 32 by 32 pixels (xn, yn), and there will be 128 of these frames (tn) in each stimulus "movie". The size of the pixels are 1/10 of a degree (sscale), and each frame lasts for 2 ms (tscale). Thus, the stimulus images represent a 3.2 by 3.2 degree image patch, and the simulation will last 256 ms overall, processing the stream of images are delivered in that time.

The value of tn is typically a run-time decision, and is set on the command line depending on the nature of the visual stimulus that is being tested. Nevertheless, a default value must be specified in the model file (.moo file).