A class for presenting grating stimuli
Categories: Stimuli Works in: PsychoPy
Name of this component (alphanumeric or _, no spaces)
How do you want to define your start point?
Options: - time (s) - frame N - condition
How do you want to define your end point?
Options: - duration (s) - duration (frames) - time (s) - frame N - condition
When does the component start?
When does the component end? (blank is endless)
(Optional) expected start (s), purely for representing in the timeline
(Optional) expected duration (s), purely for representing in the timeline
Type of noise (Binary, Normal, Gabor, Isotropic, White, Coloured, Filtered, Image)
Options: - Binary - Normal - Uniform - Gabor - Isotropic - White - Filtered - Image
Which image component should be randomised? [Amplitude,Phase]. Randomizing amplitude will keep the phase spectrum of the image but set the amplitude spectrum to random values [0…1]. This keeps spatial structure in tact. Randoming the phase spectrum will keep the amplitude spectrum of the image but set the phase spectrum to random values [-pi…pi] in radians. This makes a noise sample with no obvious structure.
Options: - Phase - Amplitude
Store the onset/offset times in the data file (as well as in the log file).
Synchronize times with screen refresh (good for visual stimuli and responses based on them)
Disable this component
Units of dimensions for this stimulus
Options: - from exp settings - deg - cm - pix - norm - height - degFlatPos - degFlat
Position of this stimulus (e.g. [1,2] )
Size of this stimulus (either a single value or x,y pair, e.g. 2.5, [1,2]
Orientation of this stimulus (in deg)
Options: - -360 - 360
(Binary, Normal and Uniform only) Size of noise elements in the stimulus units, for pixelated noise.
Foreground color of this stimulus (e.g. $[1,1,0], red )
In what format (color space) have you specified the colors? (rgb, dkl, lms, hsv)
Options: - rgb - dkl - lms - hsv
Opacity of the stimulus (1=opaque, 0=fully transparent, 0.5=translucent). Leave blank for each color to have its own opacity (recommended if any color is None).
OpenGL Blendmode [avg, add (avg is most common mode in PsychoPy, add is used if you want to generate the sum of two components)]
Options: - avg - add
Michaelson contrast of the image
An image from which to derive the frequency spectrum for the noise. Give filename (including path)
An image to define the alpha mask (ie shape)- gauss, circle… or a filename (including path)
Final spatial frequency of image in 1 or 2 dimensions, e.g. 4 or [2,3]. Use None to set to 1 copy of noise per unit length of image or 1 copy of noise per image if units=pix. Set to 1/size (or [1/size,1/size]) where size is a number (or variable) equal to the size of the stimulus to get one copy of noise per image regardless of the units.
Spatial positioning of the noise within the stimulus (wraps in range 0-1.0)
Resolution of the texture for standard ones such as sin, sqr etc. For most cases a value of 256 pixels will suffice
Options: - 32 - 64 - 128 - 256 - 512 - 1024
How should the image be interpolated if/when rescaled
Options: - linear - nearest
Apply filter to noise sample? [Butterworth, Gabor, Isoptopic]. A filter with parameters taken from the either the Filtered (Butterworth) or Gabor/Isotropic tab will be applied to OTHER noise types. [NOTE: if noise of the same type as the filter is requested the filter is applied, once only, to a white noise sample.]
Options: - None - Butterworth - Gabor - Isotropic
Base spatial frequency in cycles per unit length If units = pix this value should be < 0.5.
Spatial frequency bandwidth in octaves - Full width half height
Orientation bandwidth in degrees (Gabor only) - Full width half height
Orientation of Gabor filter in degrees. Used to set the orientation of a Gabor filter to be applied to another noise sample with a different overall orientation. The best way to set the orientation of a Gabor noise sample is to leave this as 0 degree and use the overall orientation on the Advanced tab to vary the dominant orientation of the noise. If using this setting for orientation it is strongly recommended to set the interpolation method to ‘linear’ on the Advanced tab to avoid pixelization.
Exponent for the slope of the filter’s amplitude spectrum (A=f^Exponent). 0 = flat, -1 = slope of 1/f. When used on its own the ‘filtered’ noise type applies the filter to white noise so the resulting noise samples have the spectral properties of the filter. When filtering a noise sample of another type this term takes the original spectrum and multiplies it by a ramp in frequency space with values set by the exponent. It does not force the spectrum to a specific slope.
Order of filter - higher = steeper fall off, zero = no filter
Upper cutoff frequency in cycles per unit length. Set very high to avoid an upper cutoff and make a high pass filter.
Lower cutoff frequency in cycles per unit length. Set to zero to avoid a lower cuttoff and make a low pass filter.
Truncate high and low values beyond stated standard deviations from mean and rescale greyscale range. This is not used at all for ‘binary’ or ‘uniform’ noise and scales rather than clips ‘normal’ noise). The higher this is the lower the final RMS contrast. If very low noise may appear binarised. NOTE: If a filter is used clipping and rescaling are applied after the filter, regardless of the noise type.
How to update noise if not otherwise required by other changes (none, repeat, N-frames, Seconds)
Options: - None - Repeat - N-frames - Seconds
How often to update noise (in frames or seconds) - can be a variable, ignored if any noise characteristic is updating on every frame